Remove AuthenticationError handling from integration managers

The AuthenticationError exception occurs when we don't have access to a repo
(e.g., user token lacks permissions, repo moved to different org). In this
situation, we cannot reliably send a message back to the user because:

1. The error happens when trying to fetch issue/PR data using the user's token
2. Sending messages requires the GitHub App installation token for the repo
3. If we don't have access to the repo, the installation token likely won't
   work either (especially if repo moved to different org)
4. Setting a helpful message that we can't actually deliver is deceptive

Instead, let AuthenticationError fall through to the generic exception handler,
which will:
- Log the full error with stack trace for debugging
- Attempt to send the generic 'Uh oh!' error message (which may also fail)
- At least be honest that we can't help the user in this scenario

This reverts the AuthenticationError handling from GitHub, GitLab, and Slack
integration managers.

Addresses review feedback from @neubig on PR #11473.

.github/scripts/update_pr_description.sh

@@ -17,9 +17,6 @@ DOCKER_RUN_COMMAND="docker run -it --rm \
   --name openhands-app-${SHORT_SHA} \
   docker.openhands.dev/openhands/openhands:${SHORT_SHA}"
 
-# Define the uvx command
-UVX_RUN_COMMAND="uvx --python 3.12 --from git+https://github.com/OpenHands/OpenHands@${BRANCH_NAME}#subdirectory=openhands-cli openhands"
-
 # Get the current PR body
 PR_BODY=$(gh pr view "$PR_NUMBER" --json body --jq .body)
 
@@ -37,11 +34,6 @@ GUI with Docker:
 \`\`\`
 ${DOCKER_RUN_COMMAND}
 \`\`\`
-
-CLI with uvx:
-\`\`\`
-${UVX_RUN_COMMAND}
-\`\`\`
 EOF
 )
 else
@@ -57,11 +49,6 @@ GUI with Docker:
 \`\`\`
 ${DOCKER_RUN_COMMAND}
 \`\`\`
-
-CLI with uvx:
-\`\`\`
-${UVX_RUN_COMMAND}
-\`\`\`
 EOF
 )
 fi

.github/workflows/clean-up.yml

@@ -1,69 +0,0 @@
-# Workflow that cleans up outdated and old workflows to prevent out of disk issues
-name: Delete old workflow runs
-
-# This workflow is currently only triggered manually
-on:
-  workflow_dispatch:
-    inputs:
-      days:
-        description: 'Days-worth of runs to keep for each workflow'
-        required: true
-        default: '30'
-      minimum_runs:
-        description: 'Minimum runs to keep for each workflow'
-        required: true
-        default: '10'
-      delete_workflow_pattern:
-        description: 'Name or filename of the workflow (if not set, all workflows are targeted)'
-        required: false
-      delete_workflow_by_state_pattern:
-        description: 'Filter workflows by state: active, deleted, disabled_fork, disabled_inactivity, disabled_manually'
-        required: true
-        default: "ALL"
-        type: choice
-        options:
-          - "ALL"
-          - active
-          - deleted
-          - disabled_inactivity
-          - disabled_manually
-      delete_run_by_conclusion_pattern:
-        description: 'Remove runs based on conclusion: action_required, cancelled, failure, skipped, success'
-        required: true
-        default: 'ALL'
-        type: choice
-        options:
-          - 'ALL'
-          - 'Unsuccessful: action_required,cancelled,failure,skipped'
-          - action_required
-          - cancelled
-          - failure
-          - skipped
-          - success
-      dry_run:
-        description: 'Logs simulated changes, no deletions are performed'
-        required: false
-
-jobs:
-  del_runs:
-    runs-on: blacksmith-4vcpu-ubuntu-2204
-    permissions:
-      actions: write
-      contents: read
-    steps:
-      - name: Delete workflow runs
-        uses: Mattraks/delete-workflow-runs@v2
-        with:
-          token: ${{ github.token }}
-          repository: ${{ github.repository }}
-          retain_days: ${{ github.event.inputs.days }}
-          keep_minimum_runs: ${{ github.event.inputs.minimum_runs }}
-          delete_workflow_pattern: ${{ github.event.inputs.delete_workflow_pattern }}
-          delete_workflow_by_state_pattern: ${{ github.event.inputs.delete_workflow_by_state_pattern }}
-          delete_run_by_conclusion_pattern: >-
-            ${{
-              startsWith(github.event.inputs.delete_run_by_conclusion_pattern, 'Unsuccessful:')
-              && 'action_required,cancelled,failure,skipped'
-              || github.event.inputs.delete_run_by_conclusion_pattern
-            }}
-          dry_run: ${{ github.event.inputs.dry_run }}

.github/workflows/cli-build-binary-and-optionally-release.yml

@@ -1,122 +0,0 @@
-# Workflow that builds and tests the CLI binary executable
-name: CLI - Build binary and optionally release
-
-# Run on pushes to main branch and CLI tags, and on pull requests when CLI files change
-on:
-  push:
-    branches:
-      - main
-    tags:
-      - "*-cli"
-  pull_request:
-    paths:
-      - "openhands-cli/**"
-
-permissions:
-  contents: write       # needed to create releases or upload assets
-
-# Cancel previous runs if a new commit is pushed
-concurrency:
-  group: ${{ github.workflow }}-${{ (github.head_ref && github.ref) || github.run_id }}
-  cancel-in-progress: true
-
-jobs:
-  build-binary:
-    name: Build binary executable
-    strategy:
-      matrix:
-        include:
-          # Build on Ubuntu 22.04 for maximum GLIBC compatibility (GLIBC 2.31)
-          - os: ubuntu-22.04
-            platform: linux
-            artifact_name: openhands-cli-linux
-          # Build on macOS for macOS users
-          - os: macos-15
-            platform: macos
-            artifact_name: openhands-cli-macos
-    runs-on: ${{ matrix.os }}
-
-    steps:
-      - name: Checkout repository
-        uses: actions/checkout@v4
-        with:
-          fetch-depth: 0
-
-      - name: Set up Python
-        uses: actions/setup-python@v5
-        with:
-          python-version: 3.12
-
-      - name: Install uv
-        uses: astral-sh/setup-uv@v3
-        with:
-          version: "latest"
-
-      - name: Install dependencies
-        working-directory: openhands-cli
-        run: |
-          uv sync
-
-      - name: Build binary executable
-        working-directory: openhands-cli
-        run: |
-          ./build.sh --install-pyinstaller | tee output.log
-          echo "Full output:"
-          cat output.log
-
-          if grep -q "❌" output.log; then
-            echo "❌ Found failure marker in output"
-            exit 1
-          fi
-
-          echo "✅ Build & test finished without ❌ markers"
-
-      - name: Verify binary files exist
-        run: |
-          if ! ls openhands-cli/dist/openhands* 1> /dev/null 2>&1; then
-            echo "❌ No binaries found to upload!"
-            exit 1
-          fi
-          echo "✅ Found binaries to upload."
-
-      - name: Upload binary artifact
-        uses: actions/upload-artifact@v4
-        with:
-          name: ${{ matrix.artifact_name }}
-          path: openhands-cli/dist/openhands*
-          retention-days: 30
-
-  create-github-release:
-    name: Create GitHub Release
-    runs-on: ubuntu-latest
-    needs: build-binary
-    if: startsWith(github.ref, 'refs/tags/')
-    steps:
-      - name: Checkout repository
-        uses: actions/checkout@v4
-
-      - name: Download all artifacts
-        uses: actions/download-artifact@v4
-        with:
-          path: artifacts
-
-      - name: Prepare release assets
-        run: |
-          mkdir -p release-assets
-          # Copy binaries with appropriate names for release
-          if [ -f artifacts/openhands-cli-linux/openhands ]; then
-            cp artifacts/openhands-cli-linux/openhands release-assets/openhands-linux
-          fi
-          if [ -f artifacts/openhands-cli-macos/openhands ]; then
-            cp artifacts/openhands-cli-macos/openhands release-assets/openhands-macos
-          fi
-          ls -la release-assets/
-
-      - name: Create GitHub Release
-        uses: softprops/action-gh-release@v2
-        with:
-          files: release-assets/*
-          draft: true
-          prerelease: false
-        env:
-          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}

.github/workflows/dispatch-to-docs.yml

@@ -1,23 +0,0 @@
-name: Dispatch to docs repo
-
-on:
-  push:
-    branches: [main]
-    paths:
-      - 'docs/**'
-  workflow_dispatch:
-
-jobs:
-  dispatch:
-    runs-on: ubuntu-latest
-    strategy:
-      matrix:
-        repo: ["OpenHands/docs"]
-    steps:
-      - name: Push to docs repo
-        uses: peter-evans/repository-dispatch@v3
-        with:
-          token: ${{ secrets.ALLHANDS_BOT_GITHUB_PAT }}
-          repository: ${{ matrix.repo }}
-          event-type: update
-          client-payload: '{"ref": "${{ github.ref }}", "sha": "${{ github.sha }}", "module": "openhands", "branch": "main"}'

.github/workflows/lint.yml

@@ -72,21 +72,3 @@ jobs:
       - name: Run pre-commit hooks
         working-directory: ./enterprise
         run: pre-commit run --all-files --show-diff-on-failure --config ./dev_config/python/.pre-commit-config.yaml
-
-  lint-cli-python:
-    name: Lint CLI python
-    runs-on: blacksmith-4vcpu-ubuntu-2204
-    steps:
-      - uses: actions/checkout@v4
-        with:
-          fetch-depth: 0
-      - name: Set up python
-        uses: useblacksmith/setup-python@v6
-        with:
-          python-version: 3.12
-          cache: "pip"
-      - name: Install pre-commit
-        run: pip install pre-commit==4.2.0
-      - name: Run pre-commit hooks
-        working-directory: ./openhands-cli
-        run: pre-commit run --all-files --config ./dev_config/python/.pre-commit-config.yaml

.github/workflows/mdx-lint.yml

@@ -1,70 +0,0 @@
-# Workflow that checks MDX format in docs/ folder
-name: MDX Lint
-
-# Run on pushes to main and on pull requests that modify docs/ files
-on:
-  push:
-    branches:
-      - main
-    paths:
-      - 'docs/**/*.mdx'
-  pull_request:
-    paths:
-      - 'docs/**/*.mdx'
-
-# If triggered by a PR, it will be in the same group. However, each commit on main will be in its own unique group
-concurrency:
-  group: ${{ github.workflow }}-${{ (github.head_ref && github.ref) || github.run_id }}
-  cancel-in-progress: true
-
-jobs:
-  mdx-lint:
-    name: Lint MDX files
-    runs-on: blacksmith-4vcpu-ubuntu-2204
-    steps:
-      - uses: actions/checkout@v4
-
-      - name: Install Node.js 22
-        uses: useblacksmith/setup-node@v5
-        with:
-          node-version: 22
-
-      - name: Install MDX dependencies
-        run: |
-          npm install @mdx-js/mdx@3 glob@10
-
-      - name: Validate MDX files
-        run: |
-          node -e "
-          const {compile} = require('@mdx-js/mdx');
-          const fs = require('fs');
-          const path = require('path');
-          const glob = require('glob');
-
-          async function validateMDXFiles() {
-            const files = glob.sync('docs/**/*.mdx');
-            console.log('Found', files.length, 'MDX files to validate');
-
-            let hasErrors = false;
-
-            for (const file of files) {
-              try {
-                const content = fs.readFileSync(file, 'utf8');
-                await compile(content);
-                console.log('✅ MDX parsing successful for', file);
-              } catch (err) {
-                console.error('❌ MDX parsing failed for', file, ':', err.message);
-                hasErrors = true;
-              }
-            }
-
-            if (hasErrors) {
-              console.error('\\n❌ Some MDX files have parsing errors. Please fix them before merging.');
-              process.exit(1);
-            } else {
-              console.log('\\n✅ All MDX files are valid!');
-            }
-          }
-
-          validateMDXFiles();
-          "

.github/workflows/py-tests.yml

@@ -101,56 +101,11 @@ jobs:
           path: ".coverage.enterprise.${{ matrix.python_version }}"
           include-hidden-files: true
 
-  # Run CLI unit tests
-  test-cli-python:
-    name: CLI Unit Tests
-    runs-on: blacksmith-4vcpu-ubuntu-2404
-    strategy:
-      matrix:
-        python-version: ["3.12"]
-    steps:
-      - name: Checkout repository
-        uses: actions/checkout@v4
-        with:
-          fetch-depth: 0
-
-      - name: Set up Python
-        uses: useblacksmith/setup-python@v6
-        with:
-          python-version: ${{ matrix.python-version }}
-
-      - name: Install uv
-        uses: astral-sh/setup-uv@v3
-        with:
-          version: "latest"
-
-      - name: Install dependencies
-        working-directory: ./openhands-cli
-        run: |
-          uv sync --group dev
-
-      - name: Run CLI unit tests
-        working-directory: ./openhands-cli
-        env:
-          # write coverage to repo root so the merge step finds it
-          COVERAGE_FILE: "${{ github.workspace }}/.coverage.openhands-cli.${{ matrix.python-version }}"
-        run: |
-          uv run pytest --forked -n auto -s \
-            -p no:ddtrace -p no:ddtrace.pytest_bdd -p no:ddtrace.pytest_benchmark \
-            tests --cov=openhands_cli --cov-branch
-
-      - name: Store coverage file
-        uses: actions/upload-artifact@v4
-        with:
-          name: coverage-openhands-cli
-          path: ".coverage.openhands-cli.${{ matrix.python-version }}"
-          include-hidden-files: true
-
   coverage-comment:
     name: Coverage Comment
     if: github.event_name == 'pull_request'
     runs-on: ubuntu-latest
-    needs: [test-on-linux, test-enterprise, test-cli-python]
+    needs: [test-on-linux, test-enterprise]
 
     permissions:
       pull-requests: write
@@ -164,9 +119,6 @@ jobs:
           pattern: coverage-*
           merge-multiple: true
 
-      - name: Create symlink for CLI source files
-        run: ln -sf openhands-cli/openhands_cli openhands_cli
-
       - name: Coverage comment
         id: coverage_comment
         uses: py-cov-action/python-coverage-comment-action@v3

.github/workflows/pypi-release.yml

@@ -10,7 +10,6 @@ on:
         type: choice
         options:
           - app server
-          - cli
         default: app server
   push:
     tags:
@@ -39,36 +38,3 @@ jobs:
         run: ./build.sh
       - name: publish
         run: poetry publish -u __token__ -p ${{ secrets.PYPI_TOKEN }}
-
-  release-cli:
-    name: Publish CLI to PyPI
-    runs-on: ubuntu-latest
-    # Run when manually dispatched for "cli" OR for tag pushes that contain '-cli'
-    if: |
-      (github.event_name == 'workflow_dispatch' && github.event.inputs.reason == 'cli')
-      || (github.event_name == 'push' && startsWith(github.ref, 'refs/tags/') && contains(github.ref, '-cli'))
-    steps:
-      - name: Checkout repository
-        uses: actions/checkout@v4
-
-      - name: Set up Python
-        uses: actions/setup-python@v5
-        with:
-          python-version: 3.12
-
-      - name: Install uv
-        uses: astral-sh/setup-uv@v3
-        with:
-          version: "latest"
-
-      - name: Build CLI package
-        working-directory: openhands-cli
-        run: |
-          # Clean dist directory to avoid conflicts with binary builds
-          rm -rf dist/
-          uv build
-
-      - name: Publish CLI to PyPI
-        working-directory: openhands-cli
-        run: |
-          uv publish --token ${{ secrets.PYPI_TOKEN_OPENHANDS }}

.github/workflows/run-eval.yml

@@ -1,135 +0,0 @@
-# Run evaluation on a PR, after releases, or manually
-name: Run Eval
-
-# Runs when a PR is labeled with one of the "run-eval-" labels, after releases, or manually triggered
-on:
-  pull_request:
-    types: [labeled]
-  release:
-    types: [published]
-  workflow_dispatch:
-    inputs:
-      branch:
-        description: 'Branch to evaluate'
-        required: true
-        default: 'main'
-      eval_instances:
-        description: 'Number of evaluation instances'
-        required: true
-        default: '50'
-        type: choice
-        options:
-          - '1'
-          - '2'
-          - '50'
-          - '100'
-      reason:
-        description: 'Reason for manual trigger'
-        required: false
-        default: ''
-
-env:
-  # Environment variable for the master GitHub issue number where all evaluation results will be commented
-  # This should be set to the issue number where you want all evaluation results to be posted
-  MASTER_EVAL_ISSUE_NUMBER: ${{ vars.MASTER_EVAL_ISSUE_NUMBER || '0' }}
-
-jobs:
-  trigger-job:
-    name: Trigger remote eval job
-    if: ${{ (github.event_name == 'pull_request' && (github.event.label.name == 'run-eval-1' || github.event.label.name == 'run-eval-2' || github.event.label.name == 'run-eval-50' || github.event.label.name == 'run-eval-100')) || github.event_name == 'release' || github.event_name == 'workflow_dispatch' }}
-    runs-on: blacksmith-4vcpu-ubuntu-2204
-
-    steps:
-      - name: Checkout branch
-        uses: actions/checkout@v4
-        with:
-          ref: ${{ github.event_name == 'pull_request' && github.head_ref || (github.event_name == 'workflow_dispatch' && github.event.inputs.branch) || github.ref }}
-
-      - name: Set evaluation parameters
-        id: eval_params
-        run: |
-          REPO_URL="https://github.com/${{ github.repository }}"
-          echo "Repository URL: $REPO_URL"
-
-          # Determine branch based on trigger type
-          if [[ "${{ github.event_name }}" == "pull_request" ]]; then
-            EVAL_BRANCH="${{ github.head_ref }}"
-            echo "PR Branch: $EVAL_BRANCH"
-          elif [[ "${{ github.event_name }}" == "workflow_dispatch" ]]; then
-            EVAL_BRANCH="${{ github.event.inputs.branch }}"
-            echo "Manual Branch: $EVAL_BRANCH"
-          else
-            # For release events, use the tag name or main branch
-            EVAL_BRANCH="${{ github.ref_name }}"
-            echo "Release Branch/Tag: $EVAL_BRANCH"
-          fi
-
-          # Determine evaluation instances based on trigger type
-          if [[ "${{ github.event_name }}" == "pull_request" ]]; then
-            if [[ "${{ github.event.label.name }}" == "run-eval-1" ]]; then
-              EVAL_INSTANCES="1"
-            elif [[ "${{ github.event.label.name }}" == "run-eval-2" ]]; then
-              EVAL_INSTANCES="2"
-            elif [[ "${{ github.event.label.name }}" == "run-eval-50" ]]; then
-              EVAL_INSTANCES="50"
-            elif [[ "${{ github.event.label.name }}" == "run-eval-100" ]]; then
-              EVAL_INSTANCES="100"
-            fi
-          elif [[ "${{ github.event_name }}" == "workflow_dispatch" ]]; then
-            EVAL_INSTANCES="${{ github.event.inputs.eval_instances }}"
-          else
-            # For release events, default to 50 instances
-            EVAL_INSTANCES="50"
-          fi
-
-          echo "Evaluation instances: $EVAL_INSTANCES"
-          echo "repo_url=$REPO_URL" >> $GITHUB_OUTPUT
-          echo "eval_branch=$EVAL_BRANCH" >> $GITHUB_OUTPUT
-          echo "eval_instances=$EVAL_INSTANCES" >> $GITHUB_OUTPUT
-
-      - name: Trigger remote job
-        run: |
-          # Determine PR number for the remote evaluation system
-          if [[ "${{ github.event_name }}" == "pull_request" ]]; then
-            PR_NUMBER="${{ github.event.pull_request.number }}"
-          else
-            # For non-PR triggers, use the master issue number as PR number
-            PR_NUMBER="${{ env.MASTER_EVAL_ISSUE_NUMBER }}"
-          fi
-
-          curl -X POST \
-            -H "Authorization: Bearer ${{ secrets.PAT_TOKEN }}" \
-            -H "Accept: application/vnd.github+json" \
-            -d "{\"ref\": \"main\", \"inputs\": {\"github-repo\": \"${{ steps.eval_params.outputs.repo_url }}\", \"github-branch\": \"${{ steps.eval_params.outputs.eval_branch }}\", \"pr-number\": \"${PR_NUMBER}\", \"eval-instances\": \"${{ steps.eval_params.outputs.eval_instances }}\"}}" \
-            https://api.github.com/repos/OpenHands/evaluation/actions/workflows/create-branch.yml/dispatches
-
-          # Send Slack message
-          if [[ "${{ github.event_name }}" == "pull_request" ]]; then
-            TRIGGER_URL="https://github.com/${{ github.repository }}/pull/${{ github.event.pull_request.number }}"
-            slack_text="PR $TRIGGER_URL has triggered evaluation on ${{ steps.eval_params.outputs.eval_instances }} instances..."
-          elif [[ "${{ github.event_name }}" == "release" ]]; then
-            TRIGGER_URL="https://github.com/${{ github.repository }}/releases/tag/${{ github.ref_name }}"
-            slack_text="Release $TRIGGER_URL has triggered evaluation on ${{ steps.eval_params.outputs.eval_instances }} instances..."
-          else
-            TRIGGER_URL="https://github.com/${{ github.repository }}/actions/runs/${{ github.run_id }}"
-            slack_text="Manual trigger (${{ github.event.inputs.reason || 'No reason provided' }}) has triggered evaluation on ${{ steps.eval_params.outputs.eval_instances }} instances for branch ${{ steps.eval_params.outputs.eval_branch }}..."
-          fi
-
-          curl -X POST -H 'Content-type: application/json' --data '{"text":"'"$slack_text"'"}' \
-            https://hooks.slack.com/services/${{ secrets.SLACK_TOKEN }}
-
-      - name: Comment on issue/PR
-        uses: KeisukeYamashita/create-comment@v1
-        with:
-          # For PR triggers, comment on the PR. For other triggers, comment on the master issue
-          number: ${{ github.event_name == 'pull_request' && github.event.pull_request.number || env.MASTER_EVAL_ISSUE_NUMBER }}
-          unique: false
-          comment: |
-            **Evaluation Triggered**
-
-            **Trigger:** ${{ github.event_name == 'pull_request' && format('Pull Request #{0}', github.event.pull_request.number) || (github.event_name == 'release' && 'Release') || format('Manual Trigger: {0}', github.event.inputs.reason || 'No reason provided') }}
-            **Branch:** ${{ steps.eval_params.outputs.eval_branch }}
-            **Instances:** ${{ steps.eval_params.outputs.eval_instances }}
-            **Commit:** ${{ github.sha }}
-
-            Running evaluation on the specified branch. Once eval is done, the results will be posted here.

COMMUNITY.md

@@ -1,43 +1,45 @@
-# 🙌 The OpenHands Community
+# The OpenHands Community
 
-The OpenHands community is built around the belief that (1) AI and AI agents are going to fundamentally change the way
-we build software, and (2) if this is true, we should do everything we can to make sure that the benefits provided by
-such powerful technology are accessible to everyone.
+OpenHands is a community of engineers, academics, and enthusiasts reimagining software development for an AI-powered world.
 
-If this resonates with you, we'd love to have you join us in our quest!
+## Mission
 
-## 🤝 How to Join
+It’s very clear that AI is changing software development. We want the developer community to drive that change organically, through open source.
 
-Check out our [How to Join the Community section.](https://github.com/OpenHands/OpenHands?tab=readme-ov-file#-how-to-join-the-community)
+So we’re not just building friendly interfaces for AI-driven development. We’re publishing _building blocks_ that empower developers to create new experiences, tailored to your own habits, needs, and imagination.
 
-## 💪 Becoming a Contributor
+## Ethos
 
-We welcome contributions from everyone! Whether you're a developer, a researcher, or simply enthusiastic about advancing
-the field of software engineering with AI, there are many ways to get involved:
+We have two core values: **high openness** and **high agency**. While we don’t expect everyone in the community to embody these values, we want to establish them as norms.
 
-- **Code Contributions:** Help us develop new core functionality, improve our agents, improve the frontend and other
-interfaces, or anything else that would help make OpenHands better.
-- **Research and Evaluation:** Contribute to our understanding of LLMs in software engineering, participate in
-evaluating the models, or suggest improvements.
-- **Feedback and Testing:** Use the OpenHands toolset, report bugs, suggest features, or provide feedback on usability.
+### High Openness
 
-For details, please check [CONTRIBUTING.md](./CONTRIBUTING.md).
+We welcome anyone and everyone into our community by default. You don’t have to be a software developer to help us build. You don’t have to be pro-AI to help us learn.
 
-## Code of Conduct
+Our plans, our work, our successes, and our failures are all public record. We want the world to see not just the fruits of our work, but the whole process of growing it.
 
-We have a [Code of Conduct](./CODE_OF_CONDUCT.md) that we expect all contributors to adhere to.
-Long story short, we are aiming for an open, welcoming, diverse, inclusive, and healthy community.
-All contributors are expected to contribute to building this sort of community.
+We welcome thoughtful criticism, whether it’s a comment on a PR or feedback on the community as a whole.
 
-## 🛠️ Becoming a Maintainer
+### High Agency
 
-For contributors who have made significant and sustained contributions to the project, there is a possibility of joining
-the maintainer team. The process for this is as follows:
+Everyone should feel empowered to contribute to OpenHands. Whether it’s by making a PR, hosting an event, sharing feedback, or just asking a question, don’t hold back!
 
-1. Any contributor who has made sustained and high-quality contributions to the codebase can be nominated by any
-maintainer. If you feel that you may qualify you can reach out to any of the maintainers that have reviewed your PRs and ask if you can be nominated.
-2. Once a maintainer nominates a new maintainer, there will be a discussion period among the maintainers for at least 3 days.
-3. If no concerns are raised the nomination will be accepted by acclamation, and if concerns are raised there will be a discussion and possible vote.
+OpenHands gives everyone the building blocks to create state-of-the-art developer experiences. We experiment constantly and love building new things.
 
-Note that just making many PRs does not immediately imply that you will become a maintainer. We will be looking
-at sustained high-quality contributions over a period of time, as well as good teamwork and adherence to our [Code of Conduct](./CODE_OF_CONDUCT.md).
+Coding, development practices, and communities are changing rapidly. We won’t hesitate to change direction and make big bets.
+
+## Relationship to All Hands
+
+OpenHands is supported by the for-profit organization [All Hands AI, Inc](https://www.all-hands.dev/).
+
+All Hands was founded by three of the first major contributors to OpenHands:
+
+- Xingyao Wang, a UIUC PhD candidate who got OpenHands to the top of the SWE-bench leaderboards
+- Graham Neubig, a CMU Professor who rallied the academic community around OpenHands
+- Robert Brennan, a software engineer who architected the user-facing features of OpenHands
+
+All Hands is an important part of the OpenHands ecosystem. We’ve raised over $20M--mainly to hire developers and researchers who can work on OpenHands full-time, and to provide them with expensive infrastructure. ([Join us!](https://allhandsai.applytojob.com/apply/))
+
+But we see OpenHands as much larger, and ultimately more important, than All Hands. When our financial responsibility to investors is at odds with our social responsibility to the community—as it inevitably will be, from time to time—we promise to navigate that conflict thoughtfully and transparently.
+
+At some point, we may transfer custody of OpenHands to an open source foundation. But for now, the [Benevolent Dictator approach](http://www.catb.org/~esr/writings/cathedral-bazaar/homesteading/ar01s16.html) helps us move forward with speed and intention. If we ever forget the “benevolent” part, please: fork us.

Development.md

@@ -91,14 +91,14 @@ make run
 #### Option B: Individual Server Startup
 
 - **Start the Backend Server:** If you prefer, you can start the backend server independently to focus on
-backend-related tasks or configurations.
+  backend-related tasks or configurations.
 
   ```bash
   make start-backend
   ```
 
 - **Start the Frontend Server:** Similarly, you can start the frontend server on its own to work on frontend-related
-components or interface enhancements.
+  components or interface enhancements.
   ```bash
   make start-frontend
   ```
@@ -110,6 +110,7 @@ You can use OpenHands to develop and improve OpenHands itself! This is a powerfu
 #### Quick Start
 
 1. **Build and run OpenHands:**
+
    ```bash
    export INSTALL_DOCKER=0
    export RUNTIME=local
@@ -117,6 +118,7 @@ You can use OpenHands to develop and improve OpenHands itself! This is a powerfu
    ```
 
 2. **Access the interface:**
+
    - Local development: http://localhost:3001
    - Remote/cloud environments: Use the appropriate external URL
 
@@ -199,6 +201,6 @@ Here's a guide to the important documentation files in the repository:
 - [/containers/README.md](./containers/README.md): Information about Docker containers and deployment
 - [/tests/unit/README.md](./tests/unit/README.md): Guide to writing and running unit tests
 - [/evaluation/README.md](./evaluation/README.md): Documentation for the evaluation framework and benchmarks
-- [/microagents/README.md](./microagents/README.md): Information about the microagents architecture and implementation
+- [/skills/README.md](./skills/README.md): Information about the skills architecture and implementation
 - [/openhands/server/README.md](./openhands/server/README.md): Server implementation details and API documentation
 - [/openhands/runtime/README.md](./openhands/runtime/README.md): Documentation for the runtime environment and execution model

README.md

@@ -1,22 +1,18 @@
 <a name="readme-top"></a>
 
 <div align="center">
-  <img src="https://raw.githubusercontent.com/All-Hands-AI/docs/main/openhands/static/img/logo.png" alt="Logo" width="200">
-  <h1 align="center">OpenHands: Code Less, Make More</h1>
+  <img src="https://raw.githubusercontent.com/OpenHands/docs/main/openhands/static/img/logo.png" alt="Logo" width="200">
+  <h1 align="center" style="border-bottom: none">OpenHands: AI-Driven Development</h1>
 </div>
 
 
 <div align="center">
-  <a href="https://github.com/OpenHands/OpenHands/graphs/contributors"><img src="https://img.shields.io/github/contributors/OpenHands/OpenHands?style=for-the-badge&color=blue" alt="Contributors"></a>
-  <a href="https://github.com/OpenHands/OpenHands/stargazers"><img src="https://img.shields.io/github/stars/OpenHands/OpenHands?style=for-the-badge&color=blue" alt="Stargazers"></a>
-  <a href="https://github.com/OpenHands/OpenHands/blob/main/LICENSE"><img src="https://img.shields.io/github/license/OpenHands/OpenHands?style=for-the-badge&color=blue" alt="MIT License"></a>
+  <a href="https://github.com/OpenHands/OpenHands/blob/main/LICENSE"><img src="https://img.shields.io/badge/LICENSE-MIT-20B2AA?style=for-the-badge" alt="MIT License"></a>
+  <a href="https://docs.google.com/spreadsheets/d/1wOUdFCMyY6Nt0AIqF705KN4JKOWgeI4wUGUP60krXXs/edit?gid=811504672#gid=811504672"><img src="https://img.shields.io/badge/SWEBench-72.8-00cc00?logoColor=FFE165&style=for-the-badge" alt="Benchmark Score"></a>
   <br/>
-  <a href="https://all-hands.dev/joinslack"><img src="https://img.shields.io/badge/Slack-Join%20Us-red?logo=slack&logoColor=white&style=for-the-badge" alt="Join our Slack community"></a>
-  <a href="https://github.com/OpenHands/OpenHands/blob/main/CREDITS.md"><img src="https://img.shields.io/badge/Project-Credits-blue?style=for-the-badge&color=FFE165&logo=github&logoColor=white" alt="Credits"></a>
-  <br/>
-  <a href="https://docs.all-hands.dev/usage/getting-started"><img src="https://img.shields.io/badge/Documentation-000?logo=googledocs&logoColor=FFE165&style=for-the-badge" alt="Check out the documentation"></a>
-  <a href="https://arxiv.org/abs/2407.16741"><img src="https://img.shields.io/badge/Paper%20on%20Arxiv-000?logoColor=FFE165&logo=arxiv&style=for-the-badge" alt="Paper on Arxiv"></a>
-  <a href="https://docs.google.com/spreadsheets/d/1wOUdFCMyY6Nt0AIqF705KN4JKOWgeI4wUGUP60krXXs/edit?gid=0#gid=0"><img src="https://img.shields.io/badge/Benchmark%20score-000?logoColor=FFE165&logo=huggingface&style=for-the-badge" alt="Evaluation Benchmark Score"></a>
+  <a href="https://docs.openhands.dev/sdk"><img src="https://img.shields.io/badge/Documentation-000?logo=googledocs&logoColor=FFE165&style=for-the-badge" alt="Check out the documentation"></a>
+  <a href="https://arxiv.org/abs/2511.03690"><img src="https://img.shields.io/badge/Paper-000?logoColor=FFE165&logo=arxiv&style=for-the-badge" alt="Tech Report"></a>
+
 
   <!-- Keep these links. Translations will automatically update with the README. -->
   <a href="https://www.readme-i18n.com/OpenHands/OpenHands?lang=de">Deutsch</a> |
@@ -28,157 +24,63 @@
   <a href="https://www.readme-i18n.com/OpenHands/OpenHands?lang=ru">Русский</a> |
   <a href="https://www.readme-i18n.com/OpenHands/OpenHands?lang=zh">中文</a>
 
-  <hr>
 </div>
 
-Welcome to OpenHands (formerly OpenDevin), a platform for software development agents powered by AI.
+<hr>
 
-OpenHands agents can do anything a human developer can: modify code, run commands, browse the web,
-call APIs, and yes—even copy code snippets from StackOverflow.
+🙌 Welcome to OpenHands, a [community](COMMUNITY.md) focused on AI-driven development. We’d love for you to [join us on Slack](https://dub.sh/openhands).
 
-Learn more at [docs.all-hands.dev](https://docs.all-hands.dev), or [sign up for OpenHands Cloud](https://app.all-hands.dev) to get started.
+There are a few ways to work with OpenHands:
 
+### OpenHands Software Agent SDK
+The SDK is a composable Python library that contains all of our agentic tech. It's the engine that powers everything else below.
 
-> [!IMPORTANT]
-> **Upcoming change**: We are renaming our GitHub Org from `All-Hands-AI` to `OpenHands` on October 20th, 2025.
-> Check the [tracking issue](https://github.com/All-Hands-AI/OpenHands/issues/11376) for more information.
+Define agents in code, then run them locally, or scale to 1000s of agents in the cloud.
 
+[Check out the docs](https://docs.openhands.dev/sdk) or [view the source](https://github.com/OpenHands/software-agent-sdk/)
 
-> [!IMPORTANT]
-> Using OpenHands for work? We'd love to chat! Fill out
-> [this short form](https://docs.google.com/forms/d/e/1FAIpQLSet3VbGaz8z32gW9Wm-Grl4jpt5WgMXPgJ4EDPVmCETCBpJtQ/viewform)
-> to join our Design Partner program, where you'll get early access to commercial features and the opportunity to provide input on our product roadmap.
+### OpenHands CLI
+The CLI is the easiest way to start using OpenHands. The experience will be familiar to anyone who has worked
+with e.g. Claude Code or Codex. You can power it with Claude, GPT, or any other LLM.
 
-## ☁️ OpenHands Cloud
-The easiest way to get started with OpenHands is on [OpenHands Cloud](https://app.all-hands.dev),
-which comes with $10 in free credits for new users.
+[Check out the docs](https://docs.openhands.dev/openhands/usage/run-openhands/cli-mode) or [view the source](https://github.com/OpenHands/OpenHands-CLI)
 
-## 💻 Running OpenHands Locally
+### OpenHands Local GUI
+Use the Local GUI for running agents on your laptop. It comes with a REST API and a single-page React application.
+The experience will be familiar to anyone who has used Devin or Jules.
 
-### Option 1: CLI Launcher (Recommended)
+[Check out the docs](https://docs.openhands.dev/openhands/usage/run-openhands/local-setup) or view the source in this repo.
 
-The easiest way to run OpenHands locally is using the CLI launcher with [uv](https://docs.astral.sh/uv/). This provides better isolation from your current project's virtual environment and is required for OpenHands' default MCP servers.
+### OpenHands Cloud
+This is a deployment of OpenHands GUI, running on hosted infrastructure.
 
-**Install uv** (if you haven't already):
+You can try it with a free $10 credit by [signing in with your GitHub account](https://app.all-hands.dev).
 
-See the [uv installation guide](https://docs.astral.sh/uv/getting-started/installation/) for the latest installation instructions for your platform.
+OpenHands Cloud comes with source-available features and integrations:
+- Integrations with Slack, Jira, and Linear
+- Multi-user support
+- RBAC and permissions
+- Collaboration features (e.g., conversation sharing)
 
-**Launch OpenHands**:
-```bash
-# Launch the GUI server
-uvx --python 3.12 openhands serve
+### OpenHands Enterprise
+Large enterprises can work with us to self-host OpenHands Cloud in their own VPC, via Kubernetes.
+OpenHands Enterprise can also work with the CLI and SDK above.
 
-# Or launch the CLI
-uvx --python 3.12 openhands
-```
+OpenHands Enterprise is source-available--you can see all the source code here in the enterprise/ directory,
+but you'll need to purchase a license if you want to run it for more than one month.
 
-You'll find OpenHands running at [http://localhost:3000](http://localhost:3000) (for GUI mode)!
+Enterprise contracts also come with extended support and access to our research team.
 
-### Option 2: Docker
+Learn more at [openhands.dev/enterprise](https://openhands.dev/enterprise)
 
-<details>
-<summary>Click to expand Docker command</summary>
+### Everything Else
 
-You can also run OpenHands directly with Docker:
+Check out our [Product Roadmap](https://github.com/orgs/openhands/projects/1), and feel free to
+[open up an issue](https://github.com/OpenHands/OpenHands/issues) if there's something you'd like to see!
 
-```bash
-docker pull docker.openhands.dev/openhands/runtime:0.62-nikolaik
+You might also be interested in our [evaluation infrastructure](https://github.com/OpenHands/benchmarks), our [chrome extension](https://github.com/OpenHands/openhands-chrome-extension/), or our [Theory-of-Mind module](https://github.com/OpenHands/ToM-SWE).
 
-docker run -it --rm --pull=always \
-    -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.openhands.dev/openhands/runtime:0.62-nikolaik \
-    -e LOG_ALL_EVENTS=true \
-    -v /var/run/docker.sock:/var/run/docker.sock \
-    -v ~/.openhands:/.openhands \
-    -p 3000:3000 \
-    --add-host host.docker.internal:host-gateway \
-    --name openhands-app \
-    docker.openhands.dev/openhands/openhands:0.62
-```
+All our work is available under the MIT license, except for the `enterprise/` directory in this repository (see the [enterprise license](enterprise/LICENSE) for details).
+The core `openhands` and `agent-server` Docker images are fully MIT-licensed as well.
 
-</details>
-
-> **Note**: If you used OpenHands before version 0.44, you may want to run `mv ~/.openhands-state ~/.openhands` to migrate your conversation history to the new location.
-
-> [!WARNING]
-> On a public network? See our [Hardened Docker Installation Guide](https://docs.all-hands.dev/usage/runtimes/docker#hardened-docker-installation)
-> to secure your deployment by restricting network binding and implementing additional security measures.
-
-### Getting Started
-
-When you open the application, you'll be asked to choose an LLM provider and add an API key.
-[Anthropic's Claude Sonnet 4.5](https://www.anthropic.com/api) (`anthropic/claude-sonnet-4-5-20250929`)
-works best, but you have [many options](https://docs.all-hands.dev/usage/llms).
-
-See the [Running OpenHands](https://docs.all-hands.dev/usage/installation) guide for
-system requirements and more information.
-
-## 💡 Other ways to run OpenHands
-
-> [!WARNING]
-> OpenHands is meant to be run by a single user on their local workstation.
-> It is not appropriate for multi-tenant deployments where multiple users share the same instance. There is no built-in authentication, isolation, or scalability.
->
-> If you're interested in running OpenHands in a multi-tenant environment, check out the source-available, commercially-licensed
-> [OpenHands Cloud Helm Chart](https://github.com/openHands/OpenHands-cloud)
-
-You can [connect OpenHands to your local filesystem](https://docs.all-hands.dev/usage/runtimes/docker#connecting-to-your-filesystem),
-interact with it via a [friendly CLI](https://docs.all-hands.dev/usage/how-to/cli-mode),
-run OpenHands in a scriptable [headless mode](https://docs.all-hands.dev/usage/how-to/headless-mode),
-or run it on tagged issues with [a github action](https://docs.all-hands.dev/usage/how-to/github-action).
-
-Visit [Running OpenHands](https://docs.all-hands.dev/usage/installation) for more information and setup instructions.
-
-If you want to modify the OpenHands source code, check out [Development.md](https://github.com/OpenHands/OpenHands/blob/main/Development.md).
-
-Having issues? The [Troubleshooting Guide](https://docs.all-hands.dev/usage/troubleshooting) can help.
-
-## 📖 Documentation
-
-To learn more about the project, and for tips on using OpenHands,
-check out our [documentation](https://docs.all-hands.dev/usage/getting-started).
-
-There you'll find resources on how to use different LLM providers,
-troubleshooting resources, and advanced configuration options.
-
-## 🤝 How to Join the Community
-
-OpenHands is a community-driven project, and we welcome contributions from everyone. We do most of our communication
-through Slack, so this is the best place to start, but we also are happy to have you contact us on Github:
-
-- [Join our Slack workspace](https://all-hands.dev/joinslack) - Here we talk about research, architecture, and future development.
-- [Read or post Github Issues](https://github.com/OpenHands/OpenHands/issues) - Check out the issues we're working on, or add your own ideas.
-
-See more about the community in [COMMUNITY.md](./COMMUNITY.md) or find details on contributing in [CONTRIBUTING.md](./CONTRIBUTING.md).
-
-## 📈 Progress
-
-See the monthly OpenHands roadmap [here](https://github.com/orgs/OpenHands/projects/1) (updated at the maintainer's meeting at the end of each month).
-
-<p align="center">
-  <a href="https://star-history.com/#OpenHands/OpenHands&Date">
-    <img src="https://api.star-history.com/svg?repos=OpenHands/OpenHands&type=Date" width="500" alt="Star History Chart">
-  </a>
-</p>
-
-## 📜 License
-
-Distributed under the MIT License, with the exception of the `enterprise/` folder. See [`LICENSE`](./LICENSE) for more information.
-
-## 🙏 Acknowledgements
-
-OpenHands is built by a large number of contributors, and every contribution is greatly appreciated! We also build upon other open source projects, and we are deeply thankful for their work.
-
-For a list of open source projects and licenses used in OpenHands, please see our [CREDITS.md](./CREDITS.md) file.
-
-## 📚 Cite
-
-```
-@inproceedings{
-  wang2025openhands,
-  title={OpenHands: An Open Platform for {AI} Software Developers as Generalist Agents},
-  author={Xingyao Wang and Boxuan Li and Yufan Song and Frank F. Xu and Xiangru Tang and Mingchen Zhuge and Jiayi Pan and Yueqi Song and Bowen Li and Jaskirat Singh and Hoang H. Tran and Fuqiang Li and Ren Ma and Mingzhang Zheng and Bill Qian and Yanjun Shao and Niklas Muennighoff and Yizhe Zhang and Binyuan Hui and Junyang Lin and Robert Brennan and Hao Peng and Heng Ji and Graham Neubig},
-  booktitle={The Thirteenth International Conference on Learning Representations},
-  year={2025},
-  url={https://openreview.net/forum?id=OJd3ayDDoF}
-}
-```
+If you need help with anything, or just want to chat, [come find us on Slack](https://dub.sh/openhands).

containers/app/Dockerfile

@@ -73,7 +73,7 @@ ENV VIRTUAL_ENV=/app/.venv \
 
 COPY --chown=openhands:openhands --chmod=770 --from=backend-builder ${VIRTUAL_ENV} ${VIRTUAL_ENV}
 
-COPY --chown=openhands:openhands --chmod=770 ./microagents ./microagents
+COPY --chown=openhands:openhands --chmod=770 ./skills ./skills
 COPY --chown=openhands:openhands --chmod=770 ./openhands ./openhands
 COPY --chown=openhands:openhands --chmod=777 ./openhands/runtime/plugins ./openhands/runtime/plugins
 COPY --chown=openhands:openhands pyproject.toml poetry.lock README.md MANIFEST.in LICENSE ./

dev_config/python/.pre-commit-config.yaml

@@ -3,9 +3,9 @@ repos:
     rev: v5.0.0
     hooks:
       - id: trailing-whitespace
-        exclude: ^(docs/|modules/|python/|openhands-ui/|third_party/|enterprise/|openhands-cli/)
+        exclude: ^(docs/|modules/|python/|openhands-ui/|third_party/|enterprise/)
       - id: end-of-file-fixer
-        exclude: ^(docs/|modules/|python/|openhands-ui/|third_party/|enterprise/|openhands-cli/)
+        exclude: ^(docs/|modules/|python/|openhands-ui/|third_party/|enterprise/)
       - id: check-yaml
         args: ["--allow-multiple-documents"]
       - id: debug-statements
@@ -28,12 +28,12 @@ repos:
         entry: ruff check --config dev_config/python/ruff.toml
         types_or: [python, pyi, jupyter]
         args: [--fix, --unsafe-fixes]
-        exclude: ^(third_party/|enterprise/|openhands-cli/)
+        exclude: ^(third_party/|enterprise/)
       # Run the formatter.
       - id: ruff-format
         entry: ruff format --config dev_config/python/ruff.toml
         types_or: [python, pyi, jupyter]
-        exclude: ^(third_party/|enterprise/|openhands-cli/)
+        exclude: ^(third_party/|enterprise/)
 
   - repo: https://github.com/pre-commit/mirrors-mypy
     rev: v1.15.0

enterprise/experiments/experiment_manager.py

@@ -5,12 +5,8 @@ from experiments.constants import (
     EXPERIMENT_SYSTEM_PROMPT_EXPERIMENT,
 )
 from experiments.experiment_versions import (
-    handle_condenser_max_step_experiment,
     handle_system_prompt_experiment,
 )
-from experiments.experiment_versions._004_condenser_max_step_experiment import (
-    handle_condenser_max_step_experiment__v1,
-)
 
 from openhands.core.config.openhands_config import OpenHandsConfig
 from openhands.core.logger import openhands_logger as logger
@@ -31,10 +27,6 @@ class SaaSExperimentManager(ExperimentManager):
             )
             return agent
 
-        agent = handle_condenser_max_step_experiment__v1(
-            user_id, conversation_id, agent
-        )
-
         if EXPERIMENT_SYSTEM_PROMPT_EXPERIMENT:
             agent = agent.model_copy(
                 update={'system_prompt_filename': 'system_prompt_long_horizon.j2'}
@@ -60,20 +52,7 @@ class SaaSExperimentManager(ExperimentManager):
         """
         logger.debug(
             'experiment_manager:run_conversation_variant_test:started',
-            extra={'user_id': user_id},
-        )
-
-        # Skip all experiment processing if the experiment manager is disabled
-        if not ENABLE_EXPERIMENT_MANAGER:
-            logger.info(
-                'experiment_manager:run_conversation_variant_test:skipped',
-                extra={'reason': 'experiment_manager_disabled'},
-            )
-            return conversation_settings
-
-        # Apply conversation-scoped experiments
-        conversation_settings = handle_condenser_max_step_experiment(
-            user_id, conversation_id, conversation_settings
+            extra={'user_id': user_id, 'conversation_id': conversation_id},
         )
 
         return conversation_settings

enterprise/integrations/github/github_manager.py

@@ -292,18 +292,26 @@ class GithubManager(Manager):
                     f'[GitHub] Created conversation {conversation_id} for user {user_info.username}'
                 )
 
-                # Create a GithubCallbackProcessor
-                processor = GithubCallbackProcessor(
-                    github_view=github_view,
-                    send_summary_instruction=True,
-                )
+                from openhands.server.shared import ConversationStoreImpl, config
 
-                # Register the callback processor
-                register_callback_processor(conversation_id, processor)
-
-                logger.info(
-                    f'[Github] Registered callback processor for conversation {conversation_id}'
+                conversation_store = await ConversationStoreImpl.get_instance(
+                    config, github_view.user_info.keycloak_user_id
                 )
+                metadata = await conversation_store.get_metadata(conversation_id)
+
+                if metadata.conversation_version != 'v1':
+                    # Create a GithubCallbackProcessor
+                    processor = GithubCallbackProcessor(
+                        github_view=github_view,
+                        send_summary_instruction=True,
+                    )
+
+                    # Register the callback processor
+                    register_callback_processor(conversation_id, processor)
+
+                    logger.info(
+                        f'[Github] Registered callback processor for conversation {conversation_id}'
+                    )
 
                 # Send message with conversation link
                 conversation_link = CONVERSATION_URL.format(conversation_id)

enterprise/integrations/github/github_view.py

@@ -1,4 +1,4 @@
-from uuid import uuid4
+from uuid import UUID, uuid4
 
 from github import Github, GithubIntegration
 from github.Issue import Issue
@@ -26,10 +26,22 @@ from storage.proactive_conversation_store import ProactiveConversationStore
 from storage.saas_secrets_store import SaasSecretsStore
 from storage.saas_settings_store import SaasSettingsStore
 
+from openhands.agent_server.models import SendMessageRequest
+from openhands.app_server.app_conversation.app_conversation_models import (
+    AppConversationStartRequest,
+    AppConversationStartTaskStatus,
+)
+from openhands.app_server.config import get_app_conversation_service
+from openhands.app_server.services.injector import InjectorState
+from openhands.app_server.user.specifiy_user_context import USER_CONTEXT_ATTR
+from openhands.app_server.user.user_context import UserContext
+from openhands.app_server.user.user_models import UserInfo
 from openhands.core.logger import openhands_logger as logger
 from openhands.integrations.github.github_service import GithubServiceImpl
 from openhands.integrations.provider import PROVIDER_TOKEN_TYPE, ProviderType
 from openhands.integrations.service_types import Comment
+from openhands.sdk import TextContent
+from openhands.sdk.conversation.secret_source import SecretSource
 from openhands.server.services.conversation_service import (
     initialize_conversation,
     start_conversation,
@@ -43,6 +55,52 @@ from openhands.utils.async_utils import call_sync_from_async
 OH_LABEL, INLINE_OH_LABEL = get_oh_labels(HOST)
 
 
+class GithubUserContext(UserContext):
+    """User context for GitHub integration that provides user info without web request."""
+
+    def __init__(self, keycloak_user_id: str, git_provider_tokens: PROVIDER_TOKEN_TYPE):
+        self.keycloak_user_id = keycloak_user_id
+        self.git_provider_tokens = git_provider_tokens
+        self.settings_store = SaasSettingsStore(
+            user_id=self.keycloak_user_id,
+            session_maker=session_maker,
+            config=get_config(),
+        )
+
+        self.secrets_store = SaasSecretsStore(
+            self.keycloak_user_id, session_maker, get_config()
+        )
+
+    async def get_user_id(self) -> str | None:
+        return self.keycloak_user_id
+
+    async def get_user_info(self) -> UserInfo:
+        user_settings = await self.settings_store.load()
+        return UserInfo(
+            id=self.keycloak_user_id,
+            **user_settings.model_dump(context={'expose_secrets': True}),
+        )
+
+    async def get_authenticated_git_url(self, repository: str) -> str:
+        # This would need to be implemented based on the git provider tokens
+        # For now, return a basic HTTPS URL
+        return f'https://github.com/{repository}.git'
+
+    async def get_latest_token(self, provider_type: ProviderType) -> str | None:
+        # Return the appropriate token from git_provider_tokens
+        if provider_type == ProviderType.GITHUB and self.git_provider_tokens:
+            return self.git_provider_tokens.get(ProviderType.GITHUB)
+        return None
+
+    async def get_secrets(self) -> dict[str, SecretSource]:
+        # Return empty dict for now - GitHub integration handles secrets separately
+        user_secrets = await self.secrets_store.load()
+        return dict(user_secrets.custom_secrets) if user_secrets else {}
+
+    async def get_mcp_api_key(self) -> str | None:
+        raise NotImplementedError()
+
+
 async def get_user_proactive_conversation_setting(user_id: str | None) -> bool:
     """Get the user's proactive conversation setting.
 
@@ -76,6 +134,35 @@ async def get_user_proactive_conversation_setting(user_id: str | None) -> bool:
     return settings.enable_proactive_conversation_starters
 
 
+async def get_user_v1_enabled_setting(user_id: str | None) -> bool:
+    """Get the user's V1 conversation API setting.
+
+    Args:
+        user_id: The keycloak user ID
+
+    Returns:
+        True if V1 conversations are enabled for this user, False otherwise
+    """
+
+    # If no user ID is provided, we can't check user settings
+    if not user_id:
+        return False
+
+    config = get_config()
+    settings_store = SaasSettingsStore(
+        user_id=user_id, session_maker=session_maker, config=config
+    )
+
+    settings = await call_sync_from_async(
+        settings_store.get_user_settings_by_keycloak_id, user_id
+    )
+
+    if not settings or settings.v1_enabled is None:
+        return False
+
+    return settings.v1_enabled
+
+
 # =================================================
 # SECTION: Github view types
 # =================================================
@@ -159,6 +246,31 @@ class GithubIssue(ResolverViewInterface):
         git_provider_tokens: PROVIDER_TOKEN_TYPE,
         conversation_metadata: ConversationMetadata,
     ):
+        v1_enabled = await get_user_v1_enabled_setting(self.user_info.keycloak_user_id)
+
+        if v1_enabled:
+            try:
+                # Use V1 app conversation service
+                await self._create_v1_conversation(
+                    jinja_env, git_provider_tokens, conversation_metadata
+                )
+                return
+
+            except Exception as e:
+                logger.warning(f'Error checking V1 settings, falling back to V0: {e}')
+
+        # Use existing V0 conversation service
+        await self._create_v0_conversation(
+            jinja_env, git_provider_tokens, conversation_metadata
+        )
+
+    async def _create_v0_conversation(
+        self,
+        jinja_env: Environment,
+        git_provider_tokens: PROVIDER_TOKEN_TYPE,
+        conversation_metadata: ConversationMetadata,
+    ):
+        """Create conversation using the legacy V0 system."""
         custom_secrets = await self._get_user_secrets()
 
         user_instructions, conversation_instructions = await self._get_instructions(
@@ -177,6 +289,77 @@ class GithubIssue(ResolverViewInterface):
             conversation_instructions=conversation_instructions,
         )
 
+    async def _create_v1_conversation(
+        self,
+        jinja_env: Environment,
+        git_provider_tokens: PROVIDER_TOKEN_TYPE,
+        conversation_metadata: ConversationMetadata,
+    ):
+        """Create conversation using the new V1 app conversation system."""
+        user_instructions, conversation_instructions = await self._get_instructions(
+            jinja_env
+        )
+
+        # Create the initial message request
+        initial_message = SendMessageRequest(
+            role='user', content=[TextContent(text=user_instructions)]
+        )
+
+        # Create the GitHub V1 callback processor
+        github_callback_processor = self._create_github_v1_callback_processor()
+
+        # Get the app conversation service and start the conversation
+        injector_state = InjectorState()
+
+        # Create the V1 conversation start request with the callback processor
+        start_request = AppConversationStartRequest(
+            conversation_id=UUID(conversation_metadata.conversation_id),
+            system_message_suffix=conversation_instructions,
+            initial_message=initial_message,
+            selected_repository=self.full_repo_name,
+            git_provider=ProviderType.GITHUB,
+            title=f'GitHub Issue #{self.issue_number}: {self.title}',
+            trigger=ConversationTrigger.RESOLVER,
+            processors=[
+                github_callback_processor
+            ],  # Pass the callback processor directly
+        )
+
+        # Set up the GitHub user context for the V1 system
+        github_user_context = GithubUserContext(
+            keycloak_user_id=self.user_info.keycloak_user_id,
+            git_provider_tokens=git_provider_tokens,
+        )
+        setattr(injector_state, USER_CONTEXT_ATTR, github_user_context)
+
+        async with get_app_conversation_service(
+            injector_state
+        ) as app_conversation_service:
+            async for task in app_conversation_service.start_app_conversation(
+                start_request
+            ):
+                if task.status == AppConversationStartTaskStatus.ERROR:
+                    logger.error(f'Failed to start V1 conversation: {task.detail}')
+                    raise RuntimeError(
+                        f'Failed to start V1 conversation: {task.detail}'
+                    )
+
+    def _create_github_v1_callback_processor(self):
+        """Create a V1 callback processor for GitHub integration."""
+        from openhands.app_server.event_callback.github_v1_callback_processor import (
+            GithubV1CallbackProcessor,
+        )
+
+        # Create and return the GitHub V1 callback processor
+        return GithubV1CallbackProcessor(
+            github_view_data={
+                'issue_number': self.issue_number,
+                'full_repo_name': self.full_repo_name,
+                'installation_id': self.installation_id,
+            },
+            send_summary_instruction=self.send_summary_instruction,
+        )
+
 
 @dataclass
 class GithubIssueComment(GithubIssue):
@@ -292,6 +475,24 @@ class GithubInlinePRComment(GithubPRComment):
 
         return user_instructions, conversation_instructions
 
+    def _create_github_v1_callback_processor(self):
+        """Create a V1 callback processor for GitHub integration."""
+        from openhands.app_server.event_callback.github_v1_callback_processor import (
+            GithubV1CallbackProcessor,
+        )
+
+        # Create and return the GitHub V1 callback processor
+        return GithubV1CallbackProcessor(
+            github_view_data={
+                'issue_number': self.issue_number,
+                'full_repo_name': self.full_repo_name,
+                'installation_id': self.installation_id,
+                'comment_id': self.comment_id,
+            },
+            inline_pr_comment=True,
+            send_summary_instruction=self.send_summary_instruction,
+        )
+
 
 @dataclass
 class GithubFailingAction:

enterprise/migrations/versions/081_add_parent_conversation_id.py

@@ -0,0 +1,41 @@
+"""add parent_conversation_id to conversation_metadata
+
+Revision ID: 081
+Revises: 080
+Create Date: 2025-11-06 00:00:00.000000
+
+"""
+
+from typing import Sequence, Union
+
+import sqlalchemy as sa
+from alembic import op
+
+# revision identifiers, used by Alembic.
+revision: str = '081'
+down_revision: Union[str, None] = '080'
+branch_labels: Union[str, Sequence[str], None] = None
+depends_on: Union[str, Sequence[str], None] = None
+
+
+def upgrade() -> None:
+    """Upgrade schema."""
+    op.add_column(
+        'conversation_metadata',
+        sa.Column('parent_conversation_id', sa.String(), nullable=True),
+    )
+    op.create_index(
+        op.f('ix_conversation_metadata_parent_conversation_id'),
+        'conversation_metadata',
+        ['parent_conversation_id'],
+        unique=False,
+    )
+
+
+def downgrade() -> None:
+    """Downgrade schema."""
+    op.drop_index(
+        op.f('ix_conversation_metadata_parent_conversation_id'),
+        table_name='conversation_metadata',
+    )
+    op.drop_column('conversation_metadata', 'parent_conversation_id')

enterprise/migrations/versions/082_add_setting_up_skills_enum_value.py

@@ -0,0 +1,51 @@
+"""Add SETTING_UP_SKILLS to appconversationstarttaskstatus enum
+
+Revision ID: 082
+Revises: 081
+Create Date: 2025-11-19 12:00:00.000000
+
+"""
+
+from typing import Sequence, Union
+
+from alembic import op
+from sqlalchemy import text
+
+# revision identifiers, used by Alembic.
+revision: str = '082'
+down_revision: Union[str, Sequence[str], None] = '081'
+branch_labels: Union[str, Sequence[str], None] = None
+depends_on: Union[str, Sequence[str], None] = None
+
+
+def upgrade() -> None:
+    """Add SETTING_UP_SKILLS enum value to appconversationstarttaskstatus."""
+    # Check if the enum value already exists before adding it
+    # This handles the case where the enum was created with the value already included
+    connection = op.get_bind()
+    result = connection.execute(
+        text(
+            "SELECT 1 FROM pg_enum WHERE enumlabel = 'SETTING_UP_SKILLS' "
+            "AND enumtypid = (SELECT oid FROM pg_type WHERE typname = 'appconversationstarttaskstatus')"
+        )
+    )
+
+    if not result.fetchone():
+        # Add the new enum value only if it doesn't already exist
+        op.execute(
+            "ALTER TYPE appconversationstarttaskstatus ADD VALUE 'SETTING_UP_SKILLS'"
+        )
+
+
+def downgrade() -> None:
+    """Remove SETTING_UP_SKILLS enum value from appconversationstarttaskstatus.
+
+    Note: PostgreSQL doesn't support removing enum values directly.
+    This would require recreating the enum type and updating all references.
+    For safety, this downgrade is not implemented.
+    """
+    # PostgreSQL doesn't support removing enum values directly
+    # This would require a complex migration to recreate the enum
+    # For now, we'll leave this as a no-op since removing enum values
+    # is rarely needed and can be dangerous
+    pass

enterprise/migrations/versions/083_add_v1_enabled_to_user_settings.py

@@ -0,0 +1,35 @@
+"""Add v1_enabled column to user_settings
+
+Revision ID: 083
+Revises: 082
+Create Date: 2025-11-18 00:00:00.000000
+
+"""
+
+from typing import Sequence, Union
+
+import sqlalchemy as sa
+from alembic import op
+
+# revision identifiers, used by Alembic.
+revision: str = '083'
+down_revision: Union[str, None] = '082'
+branch_labels: Union[str, Sequence[str], None] = None
+depends_on: Union[str, Sequence[str], None] = None
+
+
+def upgrade() -> None:
+    """Add v1_enabled column to user_settings table."""
+    op.add_column(
+        'user_settings',
+        sa.Column(
+            'v1_enabled',
+            sa.Boolean(),
+            nullable=True,
+        ),
+    )
+
+
+def downgrade() -> None:
+    """Remove v1_enabled column from user_settings table."""
+    op.drop_column('user_settings', 'v1_enabled')

enterprise/poetry.lock

@@ -5820,14 +5820,14 @@ llama = ["llama-index (>=0.12.29,<0.13.0)", "llama-index-core (>=0.12.29,<0.13.0
 
 [[package]]
 name = "openhands-agent-server"
-version = "1.1.0"
+version = "1.3.0"
 description = "OpenHands Agent Server - REST/WebSocket interface for OpenHands AI Agent"
 optional = false
 python-versions = ">=3.12"
 groups = ["main"]
 files = [
-    {file = "openhands_agent_server-1.1.0-py3-none-any.whl", hash = "sha256:59a856883df23488c0723e47655ef21649a321fcd4709a25a4690866eff6ac88"},
-    {file = "openhands_agent_server-1.1.0.tar.gz", hash = "sha256:e39bebd39afd45cfcfd765005e7c4e5409e46678bd7612ae20bae79f7057b935"},
+    {file = "openhands_agent_server-1.3.0-py3-none-any.whl", hash = "sha256:2f87f790c740dc3fb81821c5f9fa375af875fbb937ebca3baa6dc5c035035b3c"},
+    {file = "openhands_agent_server-1.3.0.tar.gz", hash = "sha256:0a83ae77373f5c41d0ba0e22d8f0f6144d54d55784183a50b7c098c96cd5135c"},
 ]
 
 [package.dependencies]
@@ -5843,7 +5843,7 @@ wsproto = ">=1.2.0"
 
 [[package]]
 name = "openhands-ai"
-version = "0.0.0-post.5525+0b6631523"
+version = "0.62.0"
 description = "OpenHands: Code Less, Make More"
 optional = false
 python-versions = "^3.12,<3.14"
@@ -5860,6 +5860,7 @@ bashlex = "^0.18"
 boto3 = "*"
 browsergym-core = "0.13.3"
 deprecated = "*"
+deprecation = "^2.1.0"
 dirhash = "*"
 docker = "*"
 fastapi = "*"
@@ -5884,9 +5885,9 @@ memory-profiler = "^0.61.0"
 numpy = "*"
 openai = "1.99.9"
 openhands-aci = "0.3.2"
-openhands-agent-server = "1.1.0"
-openhands-sdk = "1.1.0"
-openhands-tools = "1.1.0"
+openhands-agent-server = "1.3.0"
+openhands-sdk = "1.3.0"
+openhands-tools = "1.3.0"
 opentelemetry-api = "^1.33.1"
 opentelemetry-exporter-otlp-proto-grpc = "^1.33.1"
 pathspec = "^0.12.1"
@@ -5942,17 +5943,18 @@ url = ".."
 
 [[package]]
 name = "openhands-sdk"
-version = "1.1.0"
+version = "1.3.0"
 description = "OpenHands SDK - Core functionality for building AI agents"
 optional = false
 python-versions = ">=3.12"
 groups = ["main"]
 files = [
-    {file = "openhands_sdk-1.1.0-py3-none-any.whl", hash = "sha256:4a984ce1687a48cf99a67fdf3d37b116f8b2840743d4807810b5024af6a1d57e"},
-    {file = "openhands_sdk-1.1.0.tar.gz", hash = "sha256:855e0d8f3657205e4119e50520c17e65b3358b1a923f7a051a82512a54bf426c"},
+    {file = "openhands_sdk-1.3.0-py3-none-any.whl", hash = "sha256:feee838346f8e60ea3e4d3391de7cb854314eb8b3c9e3dbbb56f98a784aadc56"},
+    {file = "openhands_sdk-1.3.0.tar.gz", hash = "sha256:2d060803a78de462121b56dea717a66356922deb02276f37b29fae8af66343fb"},
 ]
 
 [package.dependencies]
+deprecation = ">=2.1.0"
 fastmcp = ">=2.11.3"
 httpx = ">=0.27.0"
 litellm = ">=1.77.7.dev9"
@@ -5968,14 +5970,14 @@ boto3 = ["boto3 (>=1.35.0)"]
 
 [[package]]
 name = "openhands-tools"
-version = "1.1.0"
+version = "1.3.0"
 description = "OpenHands Tools - Runtime tools for AI agents"
 optional = false
 python-versions = ">=3.12"
 groups = ["main"]
 files = [
-    {file = "openhands_tools-1.1.0-py3-none-any.whl", hash = "sha256:767d6746f05edade49263aa24450a037485a3dc23379f56917ef19aad22033f9"},
-    {file = "openhands_tools-1.1.0.tar.gz", hash = "sha256:c2fadaa4f4e16e9a3df5781ea847565dcae7171584f09ef7c0e1d97c8dfc83f6"},
+    {file = "openhands_tools-1.3.0-py3-none-any.whl", hash = "sha256:f31056d87c3058ac92709f9161c7c602daeee3ed0cb4439097b43cda105ed03e"},
+    {file = "openhands_tools-1.3.0.tar.gz", hash = "sha256:3da46f09e28593677d3e17252ce18584fcc13caab1a73213e66bd7edca2cebe0"},
 ]
 
 [package.dependencies]

enterprise/server/auth/constants.py

@@ -30,3 +30,11 @@ JIRA_DC_CLIENT_SECRET = os.getenv('JIRA_DC_CLIENT_SECRET', '').strip()
 JIRA_DC_BASE_URL = os.getenv('JIRA_DC_BASE_URL', '').strip()
 JIRA_DC_ENABLE_OAUTH = os.getenv('JIRA_DC_ENABLE_OAUTH', '1') in ('1', 'true')
 AUTH_URL = os.getenv('AUTH_URL', '').rstrip('/')
+ROLE_CHECK_ENABLED = os.getenv('ROLE_CHECK_ENABLED', 'false').lower() in (
+    '1',
+    'true',
+    't',
+    'yes',
+    'y',
+    'on',
+)

enterprise/server/auth/saas_user_auth.py

@@ -203,6 +203,15 @@ class SaasUserAuth(UserAuth):
         self.settings_store = settings_store
         return settings_store
 
+    async def get_mcp_api_key(self) -> str:
+        api_key_store = ApiKeyStore.get_instance()
+        mcp_api_key = api_key_store.retrieve_mcp_api_key(self.user_id)
+        if not mcp_api_key:
+            mcp_api_key = api_key_store.create_api_key(
+                self.user_id, 'MCP_API_KEY', None
+            )
+        return mcp_api_key
+
     @classmethod
     async def get_instance(cls, request: Request) -> UserAuth:
         logger.debug('saas_user_auth_get_instance')
@@ -243,7 +252,12 @@ def get_api_key_from_header(request: Request):
     # This is a temp hack
     # Streamable HTTP MCP Client works via redirect requests, but drops the Authorization header for reason
     # We include `X-Session-API-Key` header by default due to nested runtimes, so it used as a drop in replacement here
-    return request.headers.get('X-Session-API-Key')
+    session_api_key = request.headers.get('X-Session-API-Key')
+    if session_api_key:
+        return session_api_key
+
+    # Fallback to X-Access-Token header as an additional option
+    return request.headers.get('X-Access-Token')
 
 
 async def saas_user_auth_from_bearer(request: Request) -> SaasUserAuth | None:

enterprise/server/routes/auth.py

@@ -12,6 +12,7 @@ from server.auth.constants import (
     KEYCLOAK_CLIENT_ID,
     KEYCLOAK_REALM_NAME,
     KEYCLOAK_SERVER_URL_EXT,
+    ROLE_CHECK_ENABLED,
 )
 from server.auth.gitlab_sync import schedule_gitlab_repo_sync
 from server.auth.saas_user_auth import SaasUserAuth
@@ -132,6 +133,12 @@ async def keycloak_callback(
 
     user_info = await token_manager.get_user_info(keycloak_access_token)
     logger.debug(f'user_info: {user_info}')
+    if ROLE_CHECK_ENABLED and 'roles' not in user_info:
+        return JSONResponse(
+            status_code=status.HTTP_401_UNAUTHORIZED,
+            content={'error': 'Missing required role'},
+        )
+
     if 'sub' not in user_info or 'preferred_username' not in user_info:
         return JSONResponse(
             status_code=status.HTTP_400_BAD_REQUEST,

enterprise/storage/saas_conversation_store.py

@@ -60,6 +60,7 @@ class SaasConversationStore(ConversationStore):
         kwargs.pop('reasoning_tokens', None)
         kwargs.pop('context_window', None)
         kwargs.pop('per_turn_token', None)
+        kwargs.pop('parent_conversation_id', None)
 
         return ConversationMetadata(**kwargs)
 

enterprise/storage/saas_settings_store.py

@@ -97,6 +97,10 @@ class SaasSettingsStore(SettingsStore):
             return settings
 
     async def store(self, item: Settings):
+        # Check if provider is OpenHands and generate API key if needed
+        if item and self._is_openhands_provider(item):
+            await self._ensure_openhands_api_key(item)
+
         with self.session_maker() as session:
             existing = None
             kwargs = {}
@@ -368,6 +372,30 @@ class SaasSettingsStore(SettingsStore):
     def _should_encrypt(self, key: str) -> bool:
         return key in ('llm_api_key', 'llm_api_key_for_byor', 'search_api_key')
 
+    def _is_openhands_provider(self, item: Settings) -> bool:
+        """Check if the settings use the OpenHands provider."""
+        return bool(item.llm_model and item.llm_model.startswith('openhands/'))
+
+    async def _ensure_openhands_api_key(self, item: Settings) -> None:
+        """Generate and set the OpenHands API key for the given settings.
+
+        First checks if an existing key with the OpenHands alias exists,
+        and reuses it if found. Otherwise, generates a new key.
+        """
+        # Generate new key if none exists
+        generated_key = await self._generate_openhands_key()
+        if generated_key:
+            item.llm_api_key = SecretStr(generated_key)
+            logger.info(
+                'saas_settings_store:store:generated_openhands_key',
+                extra={'user_id': self.user_id},
+            )
+        else:
+            logger.warning(
+                'saas_settings_store:store:failed_to_generate_openhands_key',
+                extra={'user_id': self.user_id},
+            )
+
     async def _create_user_in_lite_llm(
         self, client: httpx.AsyncClient, email: str | None, max_budget: int, spend: int
     ):
@@ -390,3 +418,55 @@ class SaasSettingsStore(SettingsStore):
             },
         )
         return response
+
+    async def _generate_openhands_key(self) -> str | None:
+        """Generate a new OpenHands provider key for a user."""
+        if not (LITE_LLM_API_KEY and LITE_LLM_API_URL):
+            logger.warning(
+                'saas_settings_store:_generate_openhands_key:litellm_config_not_found',
+                extra={'user_id': self.user_id},
+            )
+            return None
+
+        try:
+            async with httpx.AsyncClient(
+                verify=httpx_verify_option(),
+                headers={
+                    'x-goog-api-key': LITE_LLM_API_KEY,
+                },
+            ) as client:
+                response = await client.post(
+                    f'{LITE_LLM_API_URL}/key/generate',
+                    json={
+                        'user_id': self.user_id,
+                        'metadata': {'type': 'openhands'},
+                    },
+                )
+                response.raise_for_status()
+                response_json = response.json()
+                key = response_json.get('key')
+
+                if key:
+                    logger.info(
+                        'saas_settings_store:_generate_openhands_key:success',
+                        extra={
+                            'user_id': self.user_id,
+                            'key_length': len(key) if key else 0,
+                            'key_prefix': (
+                                key[:10] + '...' if key and len(key) > 10 else key
+                            ),
+                        },
+                    )
+                    return key
+                else:
+                    logger.error(
+                        'saas_settings_store:_generate_openhands_key:no_key_in_response',
+                        extra={'user_id': self.user_id, 'response_json': response_json},
+                    )
+                    return None
+        except Exception as e:
+            logger.exception(
+                'saas_settings_store:_generate_openhands_key:error',
+                extra={'user_id': self.user_id, 'error': str(e)},
+            )
+            return None

enterprise/storage/user_settings.py

@@ -38,3 +38,4 @@ class UserSettings(Base):  # type: ignore
     email_verified = Column(Boolean, nullable=True)
     git_user_name = Column(String, nullable=True)
     git_user_email = Column(String, nullable=True)
+    v1_enabled = Column(Boolean, nullable=True)

enterprise/tests/unit/experiments/test_saas_experiment_manager.py

@@ -92,11 +92,8 @@ def test_unknown_variant_returns_original_agent_without_changes(monkeypatch):
     assert getattr(result, 'condenser', None) is None
 
 
-@patch('experiments.experiment_manager.handle_condenser_max_step_experiment__v1')
 @patch('experiments.experiment_manager.ENABLE_EXPERIMENT_MANAGER', False)
-def test_run_agent_variant_tests_v1_noop_when_manager_disabled(
-    mock_handle_condenser,
-):
+def test_run_agent_variant_tests_v1_noop_when_manager_disabled():
     """If ENABLE_EXPERIMENT_MANAGER is False, the method returns the exact same agent and does not call the handler."""
     agent = make_agent()
     conv_id = uuid4()
@@ -109,8 +106,6 @@ def test_run_agent_variant_tests_v1_noop_when_manager_disabled(
 
     # Same object returned (no copy)
     assert result is agent
-    # Handler should not have been called
-    mock_handle_condenser.assert_not_called()
 
 
 @patch('experiments.experiment_manager.ENABLE_EXPERIMENT_MANAGER', True)
@@ -131,7 +126,3 @@ def test_run_agent_variant_tests_v1_calls_handler_and_sets_system_prompt(monkeyp
     # Should be a different instance than the original (copied after handler runs)
     assert result is not agent
     assert result.system_prompt_filename == 'system_prompt_long_horizon.j2'
-
-    # The condenser returned by the handler must be preserved after the system-prompt override copy
-    assert isinstance(result.condenser, LLMSummarizingCondenser)
-    assert result.condenser.max_size == 80

enterprise/tests/unit/test_github_view.py

@@ -1,7 +1,9 @@
 from unittest import TestCase, mock
+from unittest.mock import MagicMock, patch
 
-from integrations.github.github_view import GithubFactory, get_oh_labels
+from integrations.github.github_view import GithubFactory, GithubIssue, get_oh_labels
 from integrations.models import Message, SourceType
+from integrations.types import UserData
 
 
 class TestGithubLabels(TestCase):
@@ -75,3 +77,128 @@ class TestGithubCommentCaseInsensitivity(TestCase):
         self.assertTrue(GithubFactory.is_issue_comment(message_lower))
         self.assertTrue(GithubFactory.is_issue_comment(message_upper))
         self.assertTrue(GithubFactory.is_issue_comment(message_mixed))
+
+
+class TestGithubV1ConversationRouting(TestCase):
+    """Test V1 conversation routing logic in GitHub integration."""
+
+    def setUp(self):
+        """Set up test fixtures."""
+        # Create a proper UserData instance instead of MagicMock
+        user_data = UserData(
+            user_id=123, username='testuser', keycloak_user_id='test-keycloak-id'
+        )
+
+        # Create a mock raw_payload
+        raw_payload = Message(
+            source=SourceType.GITHUB,
+            message={
+                'payload': {
+                    'action': 'opened',
+                    'issue': {'number': 123},
+                }
+            },
+        )
+
+        self.github_issue = GithubIssue(
+            user_info=user_data,
+            full_repo_name='test/repo',
+            issue_number=123,
+            installation_id=456,
+            conversation_id='test-conversation-id',
+            should_extract=True,
+            send_summary_instruction=False,
+            is_public_repo=True,
+            raw_payload=raw_payload,
+            uuid='test-uuid',
+            title='Test Issue',
+            description='Test issue description',
+            previous_comments=[],
+        )
+
+    @patch('integrations.github.github_view.get_user_v1_enabled_setting')
+    @patch.object(GithubIssue, '_create_v0_conversation')
+    @patch.object(GithubIssue, '_create_v1_conversation')
+    async def test_create_new_conversation_routes_to_v0_when_disabled(
+        self, mock_create_v1, mock_create_v0, mock_get_v1_setting
+    ):
+        """Test that conversation creation routes to V0 when v1_enabled is False."""
+        # Mock v1_enabled as False
+        mock_get_v1_setting.return_value = False
+        mock_create_v0.return_value = None
+        mock_create_v1.return_value = None
+
+        # Mock parameters
+        jinja_env = MagicMock()
+        git_provider_tokens = MagicMock()
+        conversation_metadata = MagicMock()
+
+        # Call the method
+        await self.github_issue.create_new_conversation(
+            jinja_env, git_provider_tokens, conversation_metadata
+        )
+
+        # Verify V0 was called and V1 was not
+        mock_create_v0.assert_called_once_with(
+            jinja_env, git_provider_tokens, conversation_metadata
+        )
+        mock_create_v1.assert_not_called()
+
+    @patch('integrations.github.github_view.get_user_v1_enabled_setting')
+    @patch.object(GithubIssue, '_create_v0_conversation')
+    @patch.object(GithubIssue, '_create_v1_conversation')
+    async def test_create_new_conversation_routes_to_v1_when_enabled(
+        self, mock_create_v1, mock_create_v0, mock_get_v1_setting
+    ):
+        """Test that conversation creation routes to V1 when v1_enabled is True."""
+        # Mock v1_enabled as True
+        mock_get_v1_setting.return_value = True
+        mock_create_v0.return_value = None
+        mock_create_v1.return_value = None
+
+        # Mock parameters
+        jinja_env = MagicMock()
+        git_provider_tokens = MagicMock()
+        conversation_metadata = MagicMock()
+
+        # Call the method
+        await self.github_issue.create_new_conversation(
+            jinja_env, git_provider_tokens, conversation_metadata
+        )
+
+        # Verify V1 was called and V0 was not
+        mock_create_v1.assert_called_once_with(
+            jinja_env, git_provider_tokens, conversation_metadata
+        )
+        mock_create_v0.assert_not_called()
+
+    @patch('integrations.github.github_view.get_user_v1_enabled_setting')
+    @patch.object(GithubIssue, '_create_v0_conversation')
+    @patch.object(GithubIssue, '_create_v1_conversation')
+    async def test_create_new_conversation_fallback_on_v1_setting_error(
+        self, mock_create_v1, mock_create_v0, mock_get_v1_setting
+    ):
+        """Test that conversation creation falls back to V0 when _create_v1_conversation fails."""
+        # Mock v1_enabled as True so V1 is attempted
+        mock_get_v1_setting.return_value = True
+        # Mock _create_v1_conversation to raise an exception
+        mock_create_v1.side_effect = Exception('V1 conversation creation failed')
+        mock_create_v0.return_value = None
+
+        # Mock parameters
+        jinja_env = MagicMock()
+        git_provider_tokens = MagicMock()
+        conversation_metadata = MagicMock()
+
+        # Call the method
+        await self.github_issue.create_new_conversation(
+            jinja_env, git_provider_tokens, conversation_metadata
+        )
+
+        # Verify V1 was attempted first, then V0 was called as fallback
+        mock_create_v1.assert_called_once_with(
+            jinja_env, git_provider_tokens, conversation_metadata
+        )
+        mock_create_v0.assert_called_once_with(
+            jinja_env, git_provider_tokens, conversation_metadata
+        )

enterprise/tests/unit/test_saas_user_auth.py

@@ -535,3 +535,115 @@ def test_get_api_key_from_header_with_invalid_authorization_format():
 
     # Assert that None was returned
     assert api_key is None
+
+
+def test_get_api_key_from_header_with_x_access_token():
+    """Test that get_api_key_from_header extracts API key from X-Access-Token header."""
+    # Create a mock request with X-Access-Token header
+    mock_request = MagicMock(spec=Request)
+    mock_request.headers = {'X-Access-Token': 'access_token_key'}
+
+    # Call the function
+    api_key = get_api_key_from_header(mock_request)
+
+    # Assert that the API key was correctly extracted
+    assert api_key == 'access_token_key'
+
+
+def test_get_api_key_from_header_priority_authorization_over_x_access_token():
+    """Test that Authorization header takes priority over X-Access-Token header."""
+    # Create a mock request with both headers
+    mock_request = MagicMock(spec=Request)
+    mock_request.headers = {
+        'Authorization': 'Bearer auth_api_key',
+        'X-Access-Token': 'access_token_key',
+    }
+
+    # Call the function
+    api_key = get_api_key_from_header(mock_request)
+
+    # Assert that the API key from Authorization header was used
+    assert api_key == 'auth_api_key'
+
+
+def test_get_api_key_from_header_priority_x_session_over_x_access_token():
+    """Test that X-Session-API-Key header takes priority over X-Access-Token header."""
+    # Create a mock request with both headers
+    mock_request = MagicMock(spec=Request)
+    mock_request.headers = {
+        'X-Session-API-Key': 'session_api_key',
+        'X-Access-Token': 'access_token_key',
+    }
+
+    # Call the function
+    api_key = get_api_key_from_header(mock_request)
+
+    # Assert that the API key from X-Session-API-Key header was used
+    assert api_key == 'session_api_key'
+
+
+def test_get_api_key_from_header_all_three_headers():
+    """Test header priority when all three headers are present."""
+    # Create a mock request with all three headers
+    mock_request = MagicMock(spec=Request)
+    mock_request.headers = {
+        'Authorization': 'Bearer auth_api_key',
+        'X-Session-API-Key': 'session_api_key',
+        'X-Access-Token': 'access_token_key',
+    }
+
+    # Call the function
+    api_key = get_api_key_from_header(mock_request)
+
+    # Assert that the API key from Authorization header was used (highest priority)
+    assert api_key == 'auth_api_key'
+
+
+def test_get_api_key_from_header_invalid_authorization_fallback_to_x_access_token():
+    """Test that invalid Authorization header falls back to X-Access-Token."""
+    # Create a mock request with invalid Authorization header and X-Access-Token
+    mock_request = MagicMock(spec=Request)
+    mock_request.headers = {
+        'Authorization': 'InvalidFormat api_key',
+        'X-Access-Token': 'access_token_key',
+    }
+
+    # Call the function
+    api_key = get_api_key_from_header(mock_request)
+
+    # Assert that the API key from X-Access-Token header was used
+    assert api_key == 'access_token_key'
+
+
+def test_get_api_key_from_header_empty_headers():
+    """Test that empty header values are handled correctly."""
+    # Create a mock request with empty header values
+    mock_request = MagicMock(spec=Request)
+    mock_request.headers = {
+        'Authorization': '',
+        'X-Session-API-Key': '',
+        'X-Access-Token': 'access_token_key',
+    }
+
+    # Call the function
+    api_key = get_api_key_from_header(mock_request)
+
+    # Assert that the API key from X-Access-Token header was used
+    assert api_key == 'access_token_key'
+
+
+def test_get_api_key_from_header_bearer_with_empty_token():
+    """Test that Bearer header with empty token falls back to other headers."""
+    # Create a mock request with Bearer header with empty token
+    mock_request = MagicMock(spec=Request)
+    mock_request.headers = {
+        'Authorization': 'Bearer ',
+        'X-Access-Token': 'access_token_key',
+    }
+
+    # Call the function
+    api_key = get_api_key_from_header(mock_request)
+
+    # Assert that empty string from Bearer is returned (current behavior)
+    # This tests the current implementation behavior
+    assert api_key == ''

evaluation/benchmarks/swefficiency/README.md

@@ -0,0 +1,65 @@
+# SWE-fficiency Evaluation
+
+This folder contains the OpenHands inference generation of the [SWE-fficiency benchmark](https://swefficiency.com/) ([paper](https://arxiv.org/pdf/2507.12415v1)).
+
+The evaluation consists of three steps:
+
+1. Environment setup: [install python environment](../../README.md#development-environment) and [configure LLM config](../../README.md#configure-openhands-and-your-llm).
+2. [Run inference](#running-inference-locally-with-docker): Generate a edit patch for each Github issue
+3. [Evaluate patches](#evaluate-generated-patches)
+
+## Setup Environment and LLM Configuration
+
+Please follow instruction [here](../../README.md#setup) to setup your local development environment and LLM.
+
+## Running inference Locally with Docker
+
+Make sure your Docker daemon is running, and you have ample disk space (at least 200-500GB, depends on the SWE-PErf set you are running on) for the instance-level docker image.
+
+When the `run_infer.sh` script is started, it will automatically pull the relevant SWE-Perf images.
+For example, for instance ID `scikit-learn_scikit-learn-11674`, it will try to pull our pre-build docker image `betty1202/sweb.eval.x86_64.scikit-learn_s_scikit-learn-11674` from DockerHub.
+This image will be used create an OpenHands runtime image where the agent will operate on.
+
+```bash
+./evaluation/benchmarks/swefficiency/scripts/run_infer.sh [model_config] [git-version] [agent] [eval_limit] [max_iter] [num_workers] [dataset] [dataset_split] [n_runs] [mode]
+
+# Example
+./evaluation/benchmarks/swefficiency/scripts/run_infer.sh llm.eval_gpt4_1106_preview HEAD CodeActAgent 500 100 1 swefficiency/swefficiency test
+```
+
+where `model_config` is mandatory, and the rest are optional.
+
+- `model_config`, e.g. `eval_gpt4_1106_preview`, is the config group name for your
+LLM settings, as defined in your `config.toml`.
+- `git-version`, e.g. `HEAD`, is the git commit hash of the OpenHands version you would
+like to evaluate. It could also be a release tag like `0.6.2`.
+- `agent`, e.g. `CodeActAgent`, is the name of the agent for benchmarks, defaulting
+to `CodeActAgent`.
+- `eval_limit`, e.g. `10`, limits the evaluation to the first `eval_limit` instances. By
+default, the script evaluates the entire SWE-Perf test set (140 issues). Note:
+in order to use `eval_limit`, you must also set `agent`.
+- `max_iter`, e.g. `20`, is the maximum number of iterations for the agent to run. By
+default, it is set to 100.
+- `num_workers`, e.g. `3`, is the number of parallel workers to run the evaluation. By
+default, it is set to 1.
+- `dataset`, a huggingface dataset name. e.g. `SWE-Perf/SWE-Perf`, specifies which dataset to evaluate on.
+- `dataset_split`, split for the huggingface dataset. e.g., `test`, `dev`. Default to `test`.
+
+- `n_runs`, e.g. `3`, is the number of times to run the evaluation. Default is 1.
+- `mode`, e.g. `swt`, `swt-ci`, or `swe`, specifies the evaluation mode. Default is `swe`.
+
+> [!CAUTION]
+> Setting `num_workers` larger than 1 is not officially tested, YMMV.
+
+
+Let's say you'd like to run 10 instances using `llm.eval_gpt4_1106_preview` and CodeActAgent,
+
+then your command would be:
+
+```bash
+./evaluation/benchmarks/swe_bench/scripts/run_infer.sh llm.eval_gpt4_1106_preview HEAD CodeActAgent 10
+```
+
+### 2. Run the SWE-fficiency benchmark official evaluation
+
+Once the output is converted, use the [official SWE-fficiency benchmark evaluation](https://github.com/swefficiency/swefficiency) to evaluate it.

evaluation/benchmarks/swefficiency/binary_patch_utils.py

@@ -0,0 +1,52 @@
+"""
+Utilities for handling binary files and patch generation in SWE-bench evaluation.
+"""
+
+
+def remove_binary_diffs(patch_text):
+    """
+    Remove binary file diffs from a git patch.
+
+    Args:
+        patch_text (str): The git patch text
+
+    Returns:
+        str: The cleaned patch text with binary diffs removed
+    """
+    lines = patch_text.splitlines()
+    cleaned_lines = []
+    block = []
+    is_binary_block = False
+
+    for line in lines:
+        if line.startswith('diff --git '):
+            if block and not is_binary_block:
+                cleaned_lines.extend(block)
+            block = [line]
+            is_binary_block = False
+        elif 'Binary files' in line:
+            is_binary_block = True
+            block.append(line)
+        else:
+            block.append(line)
+
+    if block and not is_binary_block:
+        cleaned_lines.extend(block)
+    return '\n'.join(cleaned_lines)
+
+
+def remove_binary_files_from_git():
+    """
+    Generate a bash command to remove binary files from git staging.
+
+    Returns:
+        str: A bash command that removes binary files from git staging
+    """
+    return """
+    for file in $(git status --porcelain | grep -E "^(M| M|\\?\\?|A| A)" | cut -c4-); do
+        if [ -f "$file" ] && (file "$file" | grep -q "executable" || git check-attr binary "$file" | grep -q "binary: set"); then
+            git rm -f "$file" 2>/dev/null || rm -f "$file"
+            echo "Removed: $file"
+        fi
+    done
+    """.strip()

evaluation/benchmarks/swefficiency/run_infer.py

@@ -0,0 +1,960 @@
+import asyncio
+import copy
+import functools
+import json
+import multiprocessing
+import os
+import tempfile
+from typing import Any, Literal
+
+import pandas as pd
+import toml
+from datasets import load_dataset
+
+import openhands.agenthub
+from evaluation.benchmarks.swe_bench.binary_patch_utils import (
+    remove_binary_diffs,
+    remove_binary_files_from_git,
+)
+from evaluation.utils.shared import (
+    EvalException,
+    EvalMetadata,
+    EvalOutput,
+    assert_and_raise,
+    codeact_user_response,
+    get_default_sandbox_config_for_eval,
+    get_metrics,
+    is_fatal_evaluation_error,
+    make_metadata,
+    prepare_dataset,
+    reset_logger_for_multiprocessing,
+    run_evaluation,
+    update_llm_config_for_completions_logging,
+)
+from openhands.controller.state.state import State
+from openhands.core.config import (
+    AgentConfig,
+    OpenHandsConfig,
+    get_evaluation_parser,
+    get_llm_config_arg,
+)
+from openhands.core.config.condenser_config import NoOpCondenserConfig
+from openhands.core.config.utils import get_condenser_config_arg
+from openhands.core.logger import openhands_logger as logger
+from openhands.core.main import create_runtime, run_controller
+from openhands.critic import AgentFinishedCritic
+from openhands.events.action import CmdRunAction, FileReadAction, MessageAction
+from openhands.events.observation import (
+    CmdOutputObservation,
+    ErrorObservation,
+    FileReadObservation,
+)
+from openhands.events.serialization.event import event_from_dict, event_to_dict
+from openhands.runtime.base import Runtime
+from openhands.utils.async_utils import call_async_from_sync
+from openhands.utils.shutdown_listener import sleep_if_should_continue
+
+USE_HINT_TEXT = os.environ.get('USE_HINT_TEXT', 'false').lower() == 'true'
+RUN_WITH_BROWSING = os.environ.get('RUN_WITH_BROWSING', 'false').lower() == 'true'
+BenchMode = Literal['swe', 'swt', 'swt-ci']
+
+
+AGENT_CLS_TO_FAKE_USER_RESPONSE_FN = {
+    'CodeActAgent': codeact_user_response,
+}
+
+
+def _get_swebench_workspace_dir_name(instance: pd.Series) -> str:
+    return f'{instance.repo}__{instance.version}'.replace('/', '__')
+
+
+def get_instruction(instance: pd.Series, metadata: EvalMetadata) -> MessageAction:
+    workspace_dir_name = _get_swebench_workspace_dir_name(instance)
+
+    # TODO: Change to testbed?
+    instruction = f"""
+<uploaded_files>
+/workspace/{workspace_dir_name}
+</uploaded_files>
+
+I’ve uploaded a python code repository in the directory workspace_dir_name. Consider the following performance workload and `workload()` function showing an specific usage of the repository:
+<performance_workload>
+{instance.workload}
+</performance_workload>
+
+Can you help me implement the necessary changes to the repository so that the runtime of the `workload()` function is faster? Basic guidelines:
+1. Your task is to make changes to non-test files in the /workspace directory to improve the performance of the code running in `workload()`. Please do not directly change the implementation of the `workload()` function to optimize things: I want you to focus on making the workload AS IS run faster by only editing the repository containing code that the `workload()` function calls.
+2. Make changes while ensuring the repository is functionally equivalent to the original: your changes should not introduce new bugs or cause already-passing tests to begin failing after your changes. However, you do not need to worry about tests that already fail without any changes made. For relevant test files you find in the repository, you can run them via the bash command `{instance.test_cmd} <test_file>` to check for correctness. Note that running all the tests may take a long time, so you need to determine which tests are relevant to your changes.
+3. Make sure the `workload()` function improves in performance after you make changes to the repository. The workload can potentially take some time to run, so please allow it to finish and be generous with setting your timeout parameter (a timeout value of 3600 or larger here is encouraged): for faster iteration, you should adjust the workload script to use fewer iterations. Before you complete your task, please make sure to check that the **original performance workload** and `workload()` function runs successfully and the performance is improved.
+4. You may need to reinstall/rebuild the repo for your changes to take effect before testing if you made non-Python changes. Reinstalling may take a long time to run (a timeout value of 3600 or larger here is encouraged), so please be patient with running it and allow it to complete if possible. You can reinstall the repository by running the bash command `{instance.rebuild_cmd}` in the workspace directory.
+5. All the dependencies required to run the `workload()` function are already installed in the environment. You should not install or upgrade any dependencies.
+
+Follow these steps to improve performance:
+1. As a first step, explore the repository structure.
+2. Create a Python script to reproduce the performance workload, execute it with python <workload_file>, and examine the printed output metrics.
+3. Edit the source code of the repository to improve performance. Please do not change the contents of the `workload()` function itself, but focus on optimizing the code in the repository that the original `workload()` function uses.
+4. If non-Python changes were made, rebuild the repo to make sure the changes take effect.
+5. Rerun your script to confirm that performance has improved.
+6. If necessary, identify any relevant test files in the repository related to your changes and verify that test statuses did not change after your modifications.
+7. After each attempted change, please reflect on the changes attempted and the performance impact observed. If the performance did not improve, consider alternative approaches or optimizations.
+8. Once you are satisfied, please use the finish command to complete your task.
+
+Please remember that you should not change the implementation of the `workload()` function. The performance improvement should solely come from editing the source files in the code repository.
+"""
+
+    if RUN_WITH_BROWSING:
+        instruction += (
+            '<IMPORTANT!>\nYou SHOULD NEVER attempt to browse the web. </IMPORTANT!>\n'
+        )
+
+    return MessageAction(content=instruction)
+
+
+def get_instance_docker_image(
+    instance_id: str,
+) -> str:
+    return f'ghcr.io/swefficiency/swefficiency-images:{instance_id}'
+
+
+def get_config(
+    instance: pd.Series,
+    metadata: EvalMetadata,
+    cpu_group: list[int] | None = None,
+) -> OpenHandsConfig:
+    # We use a different instance image for the each instance of swe-bench eval
+    base_container_image = get_instance_docker_image(
+        instance['instance_id'],
+    )
+    logger.info(
+        f'Using instance container image: {base_container_image}. '
+        f'Please make sure this image exists. '
+        f'Submit an issue on https://github.com/All-Hands-AI/OpenHands if you run into any issues.'
+    )
+
+    sandbox_config = get_default_sandbox_config_for_eval()
+    sandbox_config.base_container_image = base_container_image
+    sandbox_config.enable_auto_lint = True
+    sandbox_config.use_host_network = False
+    sandbox_config.timeout = 3600
+
+    # Control container cleanup behavior via environment variable
+    # Default to False for multiprocessing stability to prevent cascade failures
+    sandbox_config.rm_all_containers = True
+
+    sandbox_config.platform = 'linux/amd64'
+    sandbox_config.remote_runtime_resource_factor = 4.0
+    sandbox_config.runtime_startup_env_vars.update(
+        {
+            'NO_CHANGE_TIMEOUT_SECONDS': '900',  # 15 minutes
+        }
+    )
+
+    if cpu_group is not None:
+        print(f'Configuring Docker runtime with CPU group: {cpu_group}')
+        sandbox_config.docker_runtime_kwargs = {
+            # HACK: Use the cpu_group if provided, otherwise use all available CPUs
+            'cpuset_cpus': ','.join(map(str, cpu_group)),
+            'nano_cpus': int(1e9 * len(cpu_group)),  # optional: hard cap to vCPU count
+            'mem_limit': '16g',
+        }
+
+    # Note: We keep rm_all_containers = False for worker process safety
+
+    config = OpenHandsConfig(
+        default_agent=metadata.agent_class,
+        run_as_openhands=False,
+        max_iterations=metadata.max_iterations,
+        runtime=os.environ.get('RUNTIME', 'docker'),
+        sandbox=sandbox_config,
+        # do not mount workspace
+        workspace_base=None,
+        workspace_mount_path=None,
+    )
+    config.set_llm_config(
+        update_llm_config_for_completions_logging(
+            metadata.llm_config, metadata.eval_output_dir, instance['instance_id']
+        )
+    )
+    agent_config = AgentConfig(
+        enable_jupyter=False,
+        enable_browsing=RUN_WITH_BROWSING,
+        enable_llm_editor=False,
+        enable_mcp=False,
+        condenser=metadata.condenser_config,
+        enable_prompt_extensions=False,
+    )
+    config.set_agent_config(agent_config)
+    return config
+
+
+def initialize_runtime(
+    runtime: Runtime,
+    instance: pd.Series,  # this argument is not required
+    metadata: EvalMetadata,
+):
+    """Initialize the runtime for the agent.
+
+    This function is called before the runtime is used to run the agent.
+    """
+    logger.info('-' * 30)
+    logger.info('BEGIN Runtime Initialization Fn')
+    logger.info('-' * 30)
+    workspace_dir_name = _get_swebench_workspace_dir_name(instance)
+    obs: CmdOutputObservation
+
+    # Set instance id and git configuration
+    action = CmdRunAction(
+        command=f"""echo 'export SWE_INSTANCE_ID={instance['instance_id']}' >> ~/.bashrc && echo 'export PIP_CACHE_DIR=~/.cache/pip' >> ~/.bashrc && echo "alias git='git --no-pager'" >> ~/.bashrc && git config --global core.pager "" && git config --global diff.binary false"""
+    )
+    action.set_hard_timeout(600)
+    logger.info(action, extra={'msg_type': 'ACTION'})
+    obs = runtime.run_action(action)
+    logger.info(obs, extra={'msg_type': 'OBSERVATION'})
+    assert_and_raise(
+        obs.exit_code == 0,
+        f'Failed to export SWE_INSTANCE_ID and configure git: {str(obs)}',
+    )
+
+    action = CmdRunAction(command="""export USER=$(whoami); echo USER=${USER} """)
+    action.set_hard_timeout(600)
+    logger.info(action, extra={'msg_type': 'ACTION'})
+    obs = runtime.run_action(action)
+    logger.info(obs, extra={'msg_type': 'OBSERVATION'})
+    assert_and_raise(obs.exit_code == 0, f'Failed to export USER: {str(obs)}')
+
+    # inject the init script
+    script_dir = os.path.dirname(__file__)
+
+    # inject the instance info
+    action = CmdRunAction(command='mkdir -p /swe_util/eval_data/instances')
+    action.set_hard_timeout(600)
+    logger.info(action, extra={'msg_type': 'ACTION'})
+    obs = runtime.run_action(action)
+    logger.info(obs, extra={'msg_type': 'OBSERVATION'})
+    assert_and_raise(
+        obs.exit_code == 0,
+        f'Failed to create /swe_util/eval_data/instances: {str(obs)}',
+    )
+
+    swe_instance_json_name = 'swe-bench-instance.json'
+    with tempfile.TemporaryDirectory() as temp_dir:
+        # Construct the full path for the desired file name within the temporary directory
+        temp_file_path = os.path.join(temp_dir, swe_instance_json_name)
+        # Write to the file with the desired name within the temporary directory
+        with open(temp_file_path, 'w') as f:
+            if not isinstance(instance, dict):
+                json.dump([instance.to_dict()], f)
+            else:
+                json.dump([instance], f)
+
+        # Copy the file to the desired location
+        runtime.copy_to(temp_file_path, '/swe_util/eval_data/instances/')
+
+        # inject the instance swe entry
+        runtime.copy_to(
+            str(os.path.join(script_dir, 'scripts/setup/instance_swe_entry.sh')),
+            '/swe_util/',
+        )
+
+    action = CmdRunAction(command='cat ~/.bashrc')
+    action.set_hard_timeout(600)
+    logger.info(action, extra={'msg_type': 'ACTION'})
+    obs = runtime.run_action(action)
+    logger.info(obs, extra={'msg_type': 'OBSERVATION'})
+    assert_and_raise(obs.exit_code == 0, f'Failed to cat ~/.bashrc: {str(obs)}')
+
+    action = CmdRunAction(command='source ~/.bashrc')
+    action.set_hard_timeout(600)
+    logger.info(action, extra={'msg_type': 'ACTION'})
+    obs = runtime.run_action(action)
+    logger.info(obs, extra={'msg_type': 'OBSERVATION'})
+    if isinstance(obs, ErrorObservation):
+        logger.error(f'Failed to source ~/.bashrc: {str(obs)}')
+    assert_and_raise(obs.exit_code == 0, f'Failed to source ~/.bashrc: {str(obs)}')
+
+    action = CmdRunAction(command='source /swe_util/instance_swe_entry.sh')
+    action.set_hard_timeout(600)
+    logger.info(action, extra={'msg_type': 'ACTION'})
+    obs = runtime.run_action(action)
+    logger.info(obs, extra={'msg_type': 'OBSERVATION'})
+    assert_and_raise(
+        obs.exit_code == 0,
+        f'Failed to source /swe_util/instance_swe_entry.sh: {str(obs)}',
+    )
+
+    action = CmdRunAction(command=f'cd /workspace/{workspace_dir_name}')
+    action.set_hard_timeout(600)
+    logger.info(action, extra={'msg_type': 'ACTION'})
+    obs = runtime.run_action(action)
+    logger.info(obs, extra={'msg_type': 'OBSERVATION'})
+    assert_and_raise(
+        obs.exit_code == 0,
+        f'Failed to cd to /workspace/{workspace_dir_name}: {str(obs)}',
+    )
+
+    action = CmdRunAction(command='git reset --hard')
+    action.set_hard_timeout(600)
+    logger.info(action, extra={'msg_type': 'ACTION'})
+    obs = runtime.run_action(action)
+    logger.info(obs, extra={'msg_type': 'OBSERVATION'})
+    assert_and_raise(obs.exit_code == 0, f'Failed to git reset --hard: {str(obs)}')
+
+    action = CmdRunAction(
+        command='for remote_name in $(git remote); do git remote remove "${remote_name}"; done'
+    )
+    action.set_hard_timeout(600)
+    logger.info(action, extra={'msg_type': 'ACTION'})
+    obs = runtime.run_action(action)
+    logger.info(obs, extra={'msg_type': 'OBSERVATION'})
+    assert_and_raise(obs.exit_code == 0, f'Failed to remove git remotes: {str(obs)}')
+
+    action = CmdRunAction(command='which python')
+    action.set_hard_timeout(600)
+    logger.info(action, extra={'msg_type': 'ACTION'})
+    obs = runtime.run_action(action)
+    logger.info(obs, extra={'msg_type': 'OBSERVATION'})
+    assert_and_raise(
+        obs.exit_code == 0 and 'testbed' in obs.content,
+        f'Expected to find python interpreter from testbed, but got: {str(obs)}',
+    )
+
+    logger.info('-' * 30)
+    logger.info('END Runtime Initialization Fn')
+    logger.info('-' * 30)
+
+
+def complete_runtime(
+    runtime: Runtime,
+    instance: pd.Series,  # this argument is not required, but it is used to get the workspace_dir_name
+) -> dict[str, Any]:
+    """Complete the runtime for the agent.
+
+    This function is called before the runtime is used to run the agent.
+    If you need to do something in the sandbox to get the correctness metric after
+    the agent has run, modify this function.
+    """
+    logger.info('-' * 30)
+    logger.info('BEGIN Runtime Completion Fn')
+    logger.info('-' * 30)
+    obs: CmdOutputObservation
+    workspace_dir_name = _get_swebench_workspace_dir_name(instance)
+
+    action = CmdRunAction(command=f'cd /workspace/{workspace_dir_name}')
+    action.set_hard_timeout(600)
+    logger.info(action, extra={'msg_type': 'ACTION'})
+    obs = runtime.run_action(action)
+    logger.info(obs, extra={'msg_type': 'OBSERVATION'})
+
+    if obs.exit_code == -1:
+        # The previous command is still running
+        # We need to kill previous command
+        logger.info('The previous command is still running, trying to kill it...')
+        action = CmdRunAction(command='C-c')
+        obs = runtime.run_action(action)
+        logger.info(obs, extra={'msg_type': 'OBSERVATION'})
+
+        # Then run the command again
+        action = CmdRunAction(command=f'cd /workspace/{workspace_dir_name}')
+        action.set_hard_timeout(600)
+        logger.info(action, extra={'msg_type': 'ACTION'})
+        obs = runtime.run_action(action)
+        logger.info(obs, extra={'msg_type': 'OBSERVATION'})
+
+    if obs.exit_code == -1:
+        # The previous command is still running
+        # We need to kill previous command
+        logger.info('The previous command is still running, trying to ctrl+z it...')
+        action = CmdRunAction(command='C-z')
+        obs = runtime.run_action(action)
+        logger.info(obs, extra={'msg_type': 'OBSERVATION'})
+
+        # Then run the command again
+        action = CmdRunAction(command=f'cd /workspace/{workspace_dir_name}')
+        action.set_hard_timeout(600)
+        logger.info(action, extra={'msg_type': 'ACTION'})
+        obs = runtime.run_action(action)
+        logger.info(obs, extra={'msg_type': 'OBSERVATION'})
+
+    assert_and_raise(
+        isinstance(obs, CmdOutputObservation) and obs.exit_code == 0,
+        f'Failed to cd to /workspace/{workspace_dir_name}: {str(obs)}',
+    )
+
+    action = CmdRunAction(command='git config --global core.pager ""')
+    action.set_hard_timeout(600)
+    logger.info(action, extra={'msg_type': 'ACTION'})
+    obs = runtime.run_action(action)
+    logger.info(obs, extra={'msg_type': 'OBSERVATION'})
+    assert_and_raise(
+        isinstance(obs, CmdOutputObservation) and obs.exit_code == 0,
+        f'Failed to git config --global core.pager "": {str(obs)}',
+    )
+
+    # First check for any git repositories in subdirectories
+    action = CmdRunAction(command='find . -type d -name .git -not -path "./.git"')
+    action.set_hard_timeout(600)
+    logger.info(action, extra={'msg_type': 'ACTION'})
+    obs = runtime.run_action(action)
+    logger.info(obs, extra={'msg_type': 'OBSERVATION'})
+    assert_and_raise(
+        isinstance(obs, CmdOutputObservation) and obs.exit_code == 0,
+        f'Failed to find git repositories: {str(obs)}',
+    )
+
+    git_dirs = [p for p in obs.content.strip().split('\n') if p]
+    if git_dirs:
+        # Remove all .git directories in subdirectories
+        for git_dir in git_dirs:
+            action = CmdRunAction(command=f'rm -rf "{git_dir}"')
+            action.set_hard_timeout(600)
+            logger.info(action, extra={'msg_type': 'ACTION'})
+            obs = runtime.run_action(action)
+            logger.info(obs, extra={'msg_type': 'OBSERVATION'})
+            assert_and_raise(
+                isinstance(obs, CmdOutputObservation) and obs.exit_code == 0,
+                f'Failed to remove git directory {git_dir}: {str(obs)}',
+            )
+
+    # add all files
+    action = CmdRunAction(command='git add -A')
+    action.set_hard_timeout(600)
+    logger.info(action, extra={'msg_type': 'ACTION'})
+    obs = runtime.run_action(action)
+    logger.info(obs, extra={'msg_type': 'OBSERVATION'})
+    assert_and_raise(
+        isinstance(obs, CmdOutputObservation) and obs.exit_code == 0,
+        f'Failed to git add -A: {str(obs)}',
+    )
+
+    # Remove binary files from git staging
+    action = CmdRunAction(command=remove_binary_files_from_git())
+    action.set_hard_timeout(600)
+    logger.info(action, extra={'msg_type': 'ACTION'})
+    obs = runtime.run_action(action)
+    logger.info(obs, extra={'msg_type': 'OBSERVATION'})
+    assert_and_raise(
+        isinstance(obs, CmdOutputObservation) and obs.exit_code == 0,
+        f'Failed to remove binary files: {str(obs)}',
+    )
+
+    n_retries = 0
+    git_patch = None
+    while n_retries < 5:
+        action = CmdRunAction(
+            command=f'git diff --no-color --cached {instance["base_commit"]} > patch.diff'
+        )
+        action.set_hard_timeout(max(300 + 100 * n_retries, 600))
+        logger.info(action, extra={'msg_type': 'ACTION'})
+        obs = runtime.run_action(action)
+        logger.info(obs, extra={'msg_type': 'OBSERVATION'})
+        n_retries += 1
+        if isinstance(obs, CmdOutputObservation):
+            if obs.exit_code == 0:
+                # Read the patch file
+                action = FileReadAction(path='patch.diff')
+                action.set_hard_timeout(max(300 + 100 * n_retries, 600))
+                logger.info(action, extra={'msg_type': 'ACTION'})
+                obs = runtime.run_action(action)
+                logger.info(obs, extra={'msg_type': 'OBSERVATION'})
+                if isinstance(obs, FileReadObservation):
+                    git_patch = obs.content
+                    break
+                elif isinstance(obs, ErrorObservation):
+                    # Fall back to cat "patch.diff" to get the patch
+                    assert 'File could not be decoded as utf-8' in obs.content
+                    action = CmdRunAction(command='cat patch.diff')
+                    action.set_hard_timeout(max(300 + 100 * n_retries, 600))
+                    logger.info(action, extra={'msg_type': 'ACTION'})
+                    obs = runtime.run_action(action)
+                    assert isinstance(obs, CmdOutputObservation) and obs.exit_code == 0
+                    logger.info(obs, extra={'msg_type': 'OBSERVATION'})
+                    git_patch = obs.content
+                    break
+                else:
+                    assert_and_raise(False, f'Unexpected observation type: {str(obs)}')
+            else:
+                logger.info('Failed to get git diff, retrying...')
+                sleep_if_should_continue(10)
+        elif isinstance(obs, ErrorObservation):
+            logger.error(f'Error occurred: {obs.content}. Retrying...')
+            sleep_if_should_continue(10)
+        else:
+            assert_and_raise(False, f'Unexpected observation type: {str(obs)}')
+
+    assert_and_raise(git_patch is not None, 'Failed to get git diff (None)')
+
+    # Remove binary diffs from the patch
+    git_patch = remove_binary_diffs(git_patch)
+
+    logger.info('-' * 30)
+    logger.info('END Runtime Completion Fn')
+    logger.info('-' * 30)
+    return {'git_patch': git_patch}
+
+
+class CPUGroupManager:
+    def __init__(self, cpu_groups_queue: multiprocessing.Queue):
+        self.cpu_groups_queue = cpu_groups_queue
+
+    def __enter__(self):
+        # Get the current CPU group for this worker]
+        if self.cpu_groups_queue is not None:
+            self.cpu_group = self.cpu_groups_queue.get()
+            logger.info(f'Worker started with CPU group: {self.cpu_group}')
+            return self.cpu_group
+        return None
+
+    def __exit__(self, exc_type, exc_value, traceback):
+        # Put the CPU group back into the queue for other workers to use
+        if self.cpu_groups_queue is not None:
+            self.cpu_groups_queue.put(self.cpu_group)
+            logger.info(f'Worker finished with CPU group: {self.cpu_group}')
+
+
+def cleanup_docker_resources_for_worker():
+    """Clean up Docker resources specific to this worker process.
+
+    This prevents cascade failures when one worker's container crashes.
+    Note: This only cleans up stale locks, not containers, to avoid
+    interfering with other workers. Container cleanup is handled
+    by the DockerRuntime.close() method based on configuration.
+    """
+
+    # Clean up any stale port locks from crashed processes
+    try:
+        from openhands.runtime.utils.port_lock import cleanup_stale_locks
+
+        cleanup_stale_locks(max_age_seconds=300)  # Clean up locks older than 5 minutes
+    except Exception as e:
+        logger.debug(f'Error cleaning up stale port locks: {e}')
+
+
+def process_instance(
+    instance: pd.Series,
+    metadata: EvalMetadata,
+    reset_logger: bool = True,
+    runtime_failure_count: int = 0,
+    cpu_groups_queue: multiprocessing.Queue = None,
+) -> EvalOutput:
+    # Clean up any Docker resources from previous failed runs
+    cleanup_docker_resources_for_worker()
+
+    # HACK: Use the global and get the cpu group for this worker.
+    with CPUGroupManager(cpu_groups_queue) as cpu_group:
+        config = get_config(instance, metadata, cpu_group=cpu_group)
+
+        # Setup the logger properly, so you can run multi-processing to parallelize the evaluation
+        if reset_logger:
+            log_dir = os.path.join(metadata.eval_output_dir, 'infer_logs')
+            reset_logger_for_multiprocessing(logger, instance.instance_id, log_dir)
+        else:
+            logger.info(f'Starting evaluation for instance {instance.instance_id}.')
+
+        metadata = copy.deepcopy(metadata)
+        metadata.details['runtime_failure_count'] = runtime_failure_count
+        metadata.details['remote_runtime_resource_factor'] = (
+            config.sandbox.remote_runtime_resource_factor
+        )
+
+        runtime = create_runtime(config, sid=None)
+        call_async_from_sync(runtime.connect)
+
+        try:
+            initialize_runtime(runtime, instance, metadata)
+
+            message_action = get_instruction(instance, metadata)
+
+            # Here's how you can run the agent (similar to the `main` function) and get the final task state
+            state: State | None = asyncio.run(
+                run_controller(
+                    config=config,
+                    initial_user_action=message_action,
+                    runtime=runtime,
+                    fake_user_response_fn=AGENT_CLS_TO_FAKE_USER_RESPONSE_FN[
+                        metadata.agent_class
+                    ],
+                )
+            )
+
+            # if fatal error, throw EvalError to trigger re-run
+            if is_fatal_evaluation_error(state.last_error):
+                raise EvalException('Fatal error detected: ' + state.last_error)
+
+            # ======= THIS IS SWE-Bench specific =======
+            # Get git patch
+            return_val = complete_runtime(runtime, instance)
+            git_patch = return_val['git_patch']
+            logger.info(
+                f'Got git diff for instance {instance.instance_id}:\n--------\n{git_patch}\n--------'
+            )
+        except Exception as e:
+            # Log the error but don't let it crash other workers
+            logger.error(
+                f'Error in worker processing instance {instance.instance_id}: {str(e)}'
+            )
+            raise
+        finally:
+            # Ensure runtime is properly closed to prevent cascade failures
+            try:
+                runtime.close()
+            except Exception as e:
+                logger.warning(
+                    f'Error closing runtime for {instance.instance_id}: {str(e)}'
+                )
+                # Don't re-raise - we want to continue cleanup
+
+        # ==========================================
+
+        # ======= Attempt to evaluate the agent's edits =======
+        # we use eval_infer.sh to evaluate the agent's edits, not here
+        # because the agent may alter the environment / testcases
+        test_result = {
+            'git_patch': git_patch,
+        }
+
+        # If you are working on some simpler benchmark that only evaluates the final model output (e.g., in a MessageAction)
+        # You can simply get the LAST `MessageAction` from the returned `state.history` and parse it for evaluation.
+        if state is None:
+            raise ValueError('State should not be None.')
+
+        # NOTE: this is NO LONGER the event stream, but an agent history that includes delegate agent's events
+        histories = [event_to_dict(event) for event in state.history]
+        metrics = get_metrics(state)
+
+        # Save the output
+        instruction = message_action.content
+        if message_action.image_urls:
+            instruction += (
+                '\n\n<image_urls>'
+                + '\n'.join(message_action.image_urls)
+                + '</image_urls>'
+            )
+        output = EvalOutput(
+            instance_id=instance.instance_id,
+            instruction=instruction,
+            instance=instance.to_dict(),  # SWE Bench specific
+            test_result=test_result,
+            metadata=metadata,
+            history=histories,
+            metrics=metrics,
+            error=state.last_error if state and state.last_error else None,
+        )
+        return output
+
+
+def filter_dataset(dataset: pd.DataFrame, filter_column: str) -> pd.DataFrame:
+    file_path = os.path.join(os.path.dirname(os.path.abspath(__file__)), 'config.toml')
+    if os.path.exists(file_path):
+        with open(file_path, 'r') as file:
+            data = toml.load(file)
+            if 'selected_ids' in data:
+                selected_ids = data['selected_ids']
+                logger.info(
+                    f'Filtering {len(selected_ids)} tasks from "selected_ids"...'
+                )
+                subset = dataset[dataset[filter_column].isin(selected_ids)]
+                logger.info(f'Retained {subset.shape[0]} tasks after filtering')
+                return subset
+            if 'selected_repos' in data:
+                # repos for the swe-bench instances:
+                # ['astropy/astropy', 'django/django', 'matplotlib/matplotlib', 'mwaskom/seaborn', 'pallets/flask', 'psf/requests', 'pydata/xarray', 'pylint-dev/pylint', 'pytest-dev/pytest', 'scikit-learn/scikit-learn', 'sphinx-doc/sphinx', 'sympy/sympy']
+                selected_repos = data['selected_repos']
+                if isinstance(selected_repos, str):
+                    selected_repos = [selected_repos]
+                assert isinstance(selected_repos, list)
+                logger.info(
+                    f'Filtering {selected_repos} tasks from "selected_repos"...'
+                )
+                subset = dataset[dataset['repo'].isin(selected_repos)]
+                logger.info(f'Retained {subset.shape[0]} tasks after filtering')
+                return subset
+
+    skip_ids = os.environ.get('SKIP_IDS', '').split(',')
+    if len(skip_ids) > 0:
+        logger.info(f'Filtering {len(skip_ids)} tasks from "SKIP_IDS"...')
+        return dataset[~dataset[filter_column].isin(skip_ids)]
+    return dataset
+
+
+def divide_cpus_among_workers(num_workers, num_cpus_per_worker=4, num_to_skip=0):
+    """Divide CPUs among workers, with better error handling for multiprocessing."""
+    try:
+        current_cpus = list(os.sched_getaffinity(0))
+    except AttributeError:
+        # os.sched_getaffinity not available on all platforms
+        import multiprocessing
+
+        current_cpus = list(range(multiprocessing.cpu_count()))
+
+    num_cpus = len(current_cpus)
+    if num_workers <= 0:
+        raise ValueError('Number of workers must be greater than 0')
+
+    # Chec that num worers and num_cpus_per_worker fit into available CPUs
+    total_cpus_needed = num_workers * num_cpus_per_worker + num_to_skip
+    if total_cpus_needed > num_cpus:
+        raise ValueError(
+            f'Not enough CPUs available. Requested {total_cpus_needed} '
+            f'CPUs (num_workers={num_workers}, num_cpus_per_worker={num_cpus_per_worker}, '
+            f'num_to_skip={num_to_skip}), but only {num_cpus} CPUs are available.'
+        )
+
+    # Divide this into groups, skipping the first `num_to_skip` CPUs.
+    available_cpus = current_cpus[num_to_skip:]
+    cpu_groups = [
+        available_cpus[i * num_cpus_per_worker : (i + 1) * num_cpus_per_worker]
+        for i in range(num_workers)
+    ]
+    print(
+        f'Divided {num_cpus} CPUs into {num_workers} groups, each with {num_cpus_per_worker} CPUs.'
+    )
+    print(f'CPU groups: {cpu_groups}')
+
+    return cpu_groups
+
+
+if __name__ == '__main__':
+    parser = get_evaluation_parser()
+    parser.add_argument(
+        '--dataset',
+        type=str,
+        default=None,
+        help='data set to evaluate on, for now use local.',
+    )
+    parser.add_argument(
+        '--split',
+        type=str,
+        default='test',
+        help='split to evaluate on',
+    )
+    parser.add_argument(
+        '--mode',
+        type=str,
+        default='swe',
+        help='mode to evaluate on',
+    )
+
+    args, _ = parser.parse_known_args()
+
+    # NOTE: It is preferable to load datasets from huggingface datasets and perform post-processing
+    # so we don't need to manage file uploading to OpenHands's repo
+
+    # dataset = load_dataset(args.dataset, split=args.split)
+    # swe_bench_tests = filter_dataset(dataset.to_pandas(), 'instance_id')
+    dataset = load_dataset(args.dataset, split=args.split)
+
+    # Convert dataset to pandas DataFrame if it is not already.
+    if not isinstance(dataset, pd.DataFrame):
+        dataset = dataset.to_pandas()
+
+    dataset['version'] = dataset['version'].astype(str)
+
+    # Convert created_at column to string.
+    dataset['created_at'] = dataset['created_at'].astype(str)
+
+    swe_bench_tests = filter_dataset(dataset, 'instance_id')
+
+    logger.info(
+        f'Loaded dataset {args.dataset} with split {args.split}: {len(swe_bench_tests)} tasks'
+    )
+
+    llm_config = None
+    if args.llm_config:
+        llm_config = get_llm_config_arg(args.llm_config)
+        llm_config.log_completions = True
+        # modify_params must be False for evaluation purpose, for reproducibility and accurancy of results
+        llm_config.modify_params = False
+
+    if llm_config is None:
+        raise ValueError(f'Could not find LLM config: --llm_config {args.llm_config}')
+
+    # Get condenser config from environment variable
+    condenser_name = os.environ.get('EVAL_CONDENSER')
+    if condenser_name:
+        condenser_config = get_condenser_config_arg(condenser_name)
+        if condenser_config is None:
+            raise ValueError(
+                f'Could not find Condenser config: EVAL_CONDENSER={condenser_name}'
+            )
+    else:
+        # If no specific condenser config is provided via env var, default to NoOpCondenser
+        condenser_config = NoOpCondenserConfig()
+        logger.debug(
+            'No Condenser config provided via EVAL_CONDENSER, using NoOpCondenser.'
+        )
+
+    details = {'mode': args.mode}
+    _agent_cls = openhands.agenthub.Agent.get_cls(args.agent_cls)
+
+    dataset_descrption = (
+        args.dataset.replace('/', '__') + '-' + args.split.replace('/', '__')
+    )
+    metadata = make_metadata(
+        llm_config,
+        dataset_descrption,
+        args.agent_cls,
+        args.max_iterations,
+        args.eval_note,
+        args.eval_output_dir,
+        details=details,
+        condenser_config=condenser_config,
+    )
+
+    output_file = os.path.join(metadata.eval_output_dir, 'output.jsonl')
+    print(f'### OUTPUT FILE: {output_file} ###')
+
+    # Run evaluation in iterative mode:
+    # If a rollout fails to output AgentFinishAction, we will try again until it succeeds OR total 3 attempts have been made.
+    ITERATIVE_EVAL_MODE = (
+        os.environ.get('ITERATIVE_EVAL_MODE', 'false').lower() == 'true'
+    )
+    ITERATIVE_EVAL_MODE_MAX_ATTEMPTS = int(
+        os.environ.get('ITERATIVE_EVAL_MODE_MAX_ATTEMPTS', '3')
+    )
+
+    # Get all CPUs and divide into groups of num_workers and put them into a multiprocessing.Queue.
+    cpu_groups_queue = None
+    cpu_groups_list = divide_cpus_among_workers(args.eval_num_workers, num_to_skip=8)
+    cpu_groups_queue = multiprocessing.Manager().Queue()
+    for cpu_group in cpu_groups_list:
+        cpu_groups_queue.put(cpu_group)
+
+    if not ITERATIVE_EVAL_MODE:
+        # load the dataset
+        instances = prepare_dataset(swe_bench_tests, output_file, args.eval_n_limit)
+
+        process_instance_with_cpu_groups = functools.partial(
+            process_instance,
+            cpu_groups_queue=cpu_groups_queue,
+        )
+
+        config = get_config(
+            instances.iloc[0],  # Use the first instance to get the config
+            metadata,
+            cpu_group=None,  # We will use the cpu_groups_queue to get the cpu group later
+        )
+
+        run_evaluation(
+            instances,
+            metadata,
+            output_file,
+            args.eval_num_workers,
+            process_instance_with_cpu_groups,
+            timeout_seconds=8
+            * 60
+            * 60,  # 8 hour PER instance should be more than enough
+            max_retries=3,
+        )
+    else:
+        critic = AgentFinishedCritic()
+
+        def get_cur_output_file_path(attempt: int) -> str:
+            return (
+                f'{output_file.removesuffix(".jsonl")}.critic_attempt_{attempt}.jsonl'
+            )
+
+        eval_ids = None
+        for attempt in range(1, ITERATIVE_EVAL_MODE_MAX_ATTEMPTS + 1):
+            cur_output_file = get_cur_output_file_path(attempt)
+            logger.info(
+                f'Running evaluation with critic {critic.__class__.__name__} for attempt {attempt} of {ITERATIVE_EVAL_MODE_MAX_ATTEMPTS}.'
+            )
+
+            # For deterministic eval, we set temperature to 0.1 for (>1) attempt
+            # so hopefully we get slightly different results
+            if attempt > 1 and metadata.llm_config.temperature == 0:
+                logger.info(
+                    f'Detected temperature is 0 for (>1) attempt {attempt}. Setting temperature to 0.1...'
+                )
+                metadata.llm_config.temperature = 0.1
+
+            # Load instances - at first attempt, we evaluate all instances
+            # On subsequent attempts, we only evaluate the instances that failed the previous attempt determined by critic
+            instances = prepare_dataset(
+                swe_bench_tests, cur_output_file, args.eval_n_limit, eval_ids=eval_ids
+            )
+            if len(instances) > 0 and not isinstance(
+                instances['PASS_TO_PASS'][instances['PASS_TO_PASS'].index[0]], str
+            ):
+                for col in ['PASS_TO_PASS', 'FAIL_TO_PASS']:
+                    instances[col] = instances[col].apply(lambda x: str(x))
+
+            # Run evaluation - but save them to cur_output_file
+            logger.info(
+                f'Evaluating {len(instances)} instances for attempt {attempt}...'
+            )
+            run_evaluation(
+                instances,
+                metadata,
+                cur_output_file,
+                args.eval_num_workers,
+                process_instance,
+                timeout_seconds=8
+                * 60
+                * 60,  # 8 hour PER instance should be more than enough
+                max_retries=1,
+            )
+
+            # When eval is done, we update eval_ids to the instances that failed the current attempt
+            instances_failed = []
+            logger.info(
+                f'Use critic {critic.__class__.__name__} to check {len(instances)} instances for attempt {attempt}...'
+            )
+            with open(cur_output_file, 'r') as f:
+                for line in f:
+                    instance = json.loads(line)
+                    try:
+                        history = [
+                            event_from_dict(event) for event in instance['history']
+                        ]
+                        critic_result = critic.evaluate(
+                            history, instance['test_result'].get('git_patch', '')
+                        )
+                        if not critic_result.success:
+                            instances_failed.append(instance['instance_id'])
+                    except Exception as e:
+                        logger.error(
+                            f'Error loading history for instance {instance["instance_id"]}: {e}'
+                        )
+                        instances_failed.append(instance['instance_id'])
+            logger.info(
+                f'{len(instances_failed)} instances failed the current attempt {attempt}: {instances_failed}'
+            )
+            eval_ids = instances_failed
+
+            # If no instances failed, we break
+            if len(instances_failed) == 0:
+                break
+
+        # Then we should aggregate the results from all attempts into the original output file
+        # and remove the intermediate files
+        logger.info(
+            'Aggregating results from all attempts into the original output file...'
+        )
+        fout = open(output_file, 'w')
+        added_instance_ids = set()
+        for attempt in reversed(range(1, ITERATIVE_EVAL_MODE_MAX_ATTEMPTS + 1)):
+            cur_output_file = get_cur_output_file_path(attempt)
+            if not os.path.exists(cur_output_file):
+                logger.warning(
+                    f'Intermediate output file {cur_output_file} does not exist. Skipping...'
+                )
+                continue
+
+            with open(cur_output_file, 'r') as f:
+                for line in f:
+                    instance = json.loads(line)
+                    # Also make sure git_patch is not empty - otherwise we fall back to previous attempt (empty patch is worse than anything else)
+                    if (
+                        instance['instance_id'] not in added_instance_ids
+                        and instance['test_result'].get('git_patch', '').strip()
+                    ):
+                        fout.write(line)
+                        added_instance_ids.add(instance['instance_id'])
+            logger.info(
+                f'Aggregated instances from {cur_output_file}. Total instances added so far: {len(added_instance_ids)}'
+            )
+        fout.close()
+        logger.info(
+            f'Done! Total {len(added_instance_ids)} instances added to {output_file}'
+        )

evaluation/benchmarks/swefficiency/scripts/run_infer.sh

@@ -0,0 +1,148 @@
+#!/usr/bin/env bash
+set -eo pipefail
+
+source "evaluation/utils/version_control.sh"
+
+MODEL_CONFIG=$1
+COMMIT_HASH=$2
+AGENT=$3
+EVAL_LIMIT=$4
+MAX_ITER=$5
+NUM_WORKERS=$6
+DATASET=$7
+SPLIT=$8
+N_RUNS=$9
+MODE=${10}
+
+
+if [ -z "$NUM_WORKERS" ]; then
+  NUM_WORKERS=1
+  echo "Number of workers not specified, use default $NUM_WORKERS"
+fi
+checkout_eval_branch
+
+if [ -z "$AGENT" ]; then
+  echo "Agent not specified, use default CodeActAgent"
+  AGENT="CodeActAgent"
+fi
+
+if [ -z "$MAX_ITER" ]; then
+  echo "MAX_ITER not specified, use default 100"
+  MAX_ITER=100
+fi
+
+if [ -z "$RUN_WITH_BROWSING" ]; then
+  echo "RUN_WITH_BROWSING not specified, use default false"
+  RUN_WITH_BROWSING=false
+fi
+
+
+if [ -z "$DATASET" ]; then
+  echo "DATASET not specified, use default princeton-nlp/SWE-bench_Lite"
+  DATASET="swefficiency/swefficiency"
+fi
+
+if [ -z "$SPLIT" ]; then
+  echo "SPLIT not specified, use default test"
+  SPLIT="test"
+fi
+
+if [ -z "$MODE" ]; then
+  MODE="swe"
+  echo "MODE not specified, use default $MODE"
+fi
+
+if [ -n "$EVAL_CONDENSER" ]; then
+  echo "Using Condenser Config: $EVAL_CONDENSER"
+else
+  echo "No Condenser Config provided via EVAL_CONDENSER, use default (NoOpCondenser)."
+fi
+
+export RUN_WITH_BROWSING=$RUN_WITH_BROWSING
+echo "RUN_WITH_BROWSING: $RUN_WITH_BROWSING"
+
+get_openhands_version
+
+echo "AGENT: $AGENT"
+echo "OPENHANDS_VERSION: $OPENHANDS_VERSION"
+echo "MODEL_CONFIG: $MODEL_CONFIG"
+echo "DATASET: $DATASET"
+echo "SPLIT: $SPLIT"
+echo "MAX_ITER: $MAX_ITER"
+echo "NUM_WORKERS: $NUM_WORKERS"
+echo "COMMIT_HASH: $COMMIT_HASH"
+echo "MODE: $MODE"
+echo "EVAL_CONDENSER: $EVAL_CONDENSER"
+
+# Default to NOT use Hint
+if [ -z "$USE_HINT_TEXT" ]; then
+  export USE_HINT_TEXT=false
+fi
+echo "USE_HINT_TEXT: $USE_HINT_TEXT"
+EVAL_NOTE="$OPENHANDS_VERSION"
+# if not using Hint, add -no-hint to the eval note
+if [ "$USE_HINT_TEXT" = false ]; then
+  EVAL_NOTE="$EVAL_NOTE-no-hint"
+fi
+
+if [ "$RUN_WITH_BROWSING" = true ]; then
+  EVAL_NOTE="$EVAL_NOTE-with-browsing"
+fi
+
+if [ -n "$EXP_NAME" ]; then
+  EVAL_NOTE="$EVAL_NOTE-$EXP_NAME"
+fi
+# if mode != swe, add mode to the eval note
+if [ "$MODE" != "swe" ]; then
+  EVAL_NOTE="${EVAL_NOTE}-${MODE}"
+fi
+# Add condenser config to eval note if provided
+if [ -n "$EVAL_CONDENSER" ]; then
+  EVAL_NOTE="${EVAL_NOTE}-${EVAL_CONDENSER}"
+fi
+
+# export RUNTIME="remote"
+# export SANDBOX_REMOTE_RUNTIME_API_URL="https://runtime.eval.all-hands.dev"
+export NO_CHANGE_TIMEOUT_SECONDS=900 # 15 minutes
+
+function run_eval() {
+  local eval_note="${1}"
+  COMMAND="poetry run python evaluation/benchmarks/swefficiency/run_infer.py \
+    --agent-cls $AGENT \
+    --llm-config $MODEL_CONFIG \
+    --max-iterations $MAX_ITER \
+    --eval-num-workers $NUM_WORKERS \
+    --eval-note $eval_note \
+    --dataset $DATASET \
+    --split $SPLIT \
+    --mode $MODE"
+
+  if [ -n "$EVAL_LIMIT" ]; then
+    echo "EVAL_LIMIT: $EVAL_LIMIT"
+    COMMAND="$COMMAND --eval-n-limit $EVAL_LIMIT"
+  fi
+
+  # Run the command
+  eval $COMMAND
+}
+
+unset SANDBOX_ENV_GITHUB_TOKEN # prevent the agent from using the github token to push
+if [ -z "$N_RUNS" ]; then
+  N_RUNS=1
+  echo "N_RUNS not specified, use default $N_RUNS"
+fi
+
+# Skip runs if the run number is in the SKIP_RUNS list
+# read from env variable SKIP_RUNS as a comma separated list of run numbers
+SKIP_RUNS=(${SKIP_RUNS//,/ })
+for i in $(seq 1 $N_RUNS); do
+  if [[ " ${SKIP_RUNS[@]} " =~ " $i " ]]; then
+    echo "Skipping run $i"
+    continue
+  fi
+  current_eval_note="$EVAL_NOTE-run_$i"
+  echo "EVAL_NOTE: $current_eval_note"
+  run_eval $current_eval_note
+done
+
+checkout_original_branch

evaluation/benchmarks/swefficiency/scripts/setup/instance_swe_entry.sh

@@ -0,0 +1,43 @@
+#!/usr/bin/env bash
+
+source ~/.bashrc
+SWEUTIL_DIR=/swe_util
+
+# FIXME: Cannot read SWE_INSTANCE_ID from the environment variable
+# SWE_INSTANCE_ID=django__django-11099
+if [ -z "$SWE_INSTANCE_ID" ]; then
+    echo "Error: SWE_INSTANCE_ID is not set." >&2
+    exit 1
+fi
+
+# Read the swe-bench-test-lite.json file and extract the required item based on instance_id
+item=$(jq --arg INSTANCE_ID "$SWE_INSTANCE_ID" '.[] | select(.instance_id == $INSTANCE_ID)' $SWEUTIL_DIR/eval_data/instances/swe-bench-instance.json)
+
+if [[ -z "$item" ]]; then
+  echo "No item found for the provided instance ID."
+  exit 1
+fi
+
+
+WORKSPACE_NAME=$(echo "$item" | jq -r '(.repo | tostring) + "__" + (.version | tostring) | gsub("/"; "__")')
+
+echo "WORKSPACE_NAME: $WORKSPACE_NAME"
+
+# Clear the workspace
+if [ -d /workspace ]; then
+    rm -rf /workspace/*
+else
+    mkdir /workspace
+fi
+# Copy repo to workspace
+if [ -d /workspace/$WORKSPACE_NAME ]; then
+    rm -rf /workspace/$WORKSPACE_NAME
+fi
+mkdir -p /workspace
+cp -r /testbed /workspace/$WORKSPACE_NAME
+
+# Activate instance-specific environment
+if [ -d /opt/miniconda3 ]; then
+    . /opt/miniconda3/etc/profile.d/conda.sh
+    conda activate testbed
+fi

evaluation/benchmarks/swefficiency/scripts/setup/prepare_swe_utils.sh

@@ -0,0 +1,27 @@
+#!/usr/bin/env bash
+
+set -e
+EVAL_WORKSPACE="evaluation/benchmarks/swe_bench/eval_workspace"
+mkdir -p $EVAL_WORKSPACE
+
+# 1. Prepare REPO
+echo "==== Prepare SWE-bench repo ===="
+OH_SWE_BENCH_REPO_PATH="https://github.com/All-Hands-AI/SWE-bench.git"
+OH_SWE_BENCH_REPO_BRANCH="eval"
+git clone -b $OH_SWE_BENCH_REPO_BRANCH $OH_SWE_BENCH_REPO_PATH $EVAL_WORKSPACE/OH-SWE-bench
+
+# 2. Prepare DATA
+echo "==== Prepare SWE-bench data ===="
+EVAL_IMAGE=ghcr.io/all-hands-ai/eval-swe-bench:builder_with_conda
+EVAL_WORKSPACE=$(realpath $EVAL_WORKSPACE)
+chmod +x $EVAL_WORKSPACE/OH-SWE-bench/swebench/harness/prepare_data.sh
+if [ -d $EVAL_WORKSPACE/eval_data ]; then
+    rm -r $EVAL_WORKSPACE/eval_data
+fi
+docker run \
+    -v $EVAL_WORKSPACE:/workspace \
+    -w /workspace \
+    -u $(id -u):$(id -g) \
+    -e HF_DATASETS_CACHE="/tmp" \
+    --rm -it $EVAL_IMAGE \
+    bash -c "cd OH-SWE-bench/swebench/harness && /swe_util/miniforge3/bin/conda run -n swe-bench-eval ./prepare_data.sh && mv eval_data /workspace/"

evaluation/benchmarks/swefficiency/scripts/setup/swe_entry.sh

@@ -0,0 +1,96 @@
+#!/usr/bin/env bash
+
+set -e
+
+# assert user name is `root`
+if [ "$USER" != "root" ]; then
+    echo "Error: This script is intended to be run by the 'root' user only." >&2
+    exit 1
+fi
+
+source ~/.bashrc
+
+SWEUTIL_DIR=/swe_util
+
+# Create logs directory
+LOG_DIR=/openhands/logs
+mkdir -p $LOG_DIR && chmod 777 $LOG_DIR
+
+# FIXME: Cannot read SWE_INSTANCE_ID from the environment variable
+# SWE_INSTANCE_ID=django__django-11099
+if [ -z "$SWE_INSTANCE_ID" ]; then
+    echo "Error: SWE_INSTANCE_ID is not set." >&2
+    exit 1
+fi
+
+# Read the swe-bench-test-lite.json file and extract the required item based on instance_id
+item=$(jq --arg INSTANCE_ID "$SWE_INSTANCE_ID" '.[] | select(.instance_id == $INSTANCE_ID)' $SWEUTIL_DIR/eval_data/instances/swe-bench-test-lite.json)
+
+if [[ -z "$item" ]]; then
+  echo "No item found for the provided instance ID."
+  exit 1
+fi
+
+CONDA_ENV_NAME=$(echo "$item" | jq -r '.repo + "__" + .version | gsub("/"; "__")')
+
+echo "CONDA_ENV_NAME: $CONDA_ENV_NAME"
+
+SWE_TASK_DIR=/openhands/swe_tasks
+mkdir -p $SWE_TASK_DIR
+# Dump test_patch to /workspace/test.patch
+echo "$item" | jq -r '.test_patch' > $SWE_TASK_DIR/test.patch
+# Dump patch to /workspace/gold.patch
+echo "$item" | jq -r '.patch' > $SWE_TASK_DIR/gold.patch
+# Dump the item to /workspace/instance.json except for the "test_patch" and "patch" fields
+echo "$item" | jq 'del(.test_patch, .patch)' > $SWE_TASK_DIR/instance.json
+
+# Clear the workspace
+rm -rf /workspace/*
+# Copy repo to workspace
+if [ -d /workspace/$CONDA_ENV_NAME ]; then
+    rm -rf /workspace/$CONDA_ENV_NAME
+fi
+cp -r $SWEUTIL_DIR/eval_data/testbeds/$CONDA_ENV_NAME /workspace
+
+# Reset swe-bench testbed and install the repo
+. $SWEUTIL_DIR/miniforge3/etc/profile.d/conda.sh
+conda config --set changeps1 False
+conda config --append channels conda-forge
+conda activate swe-bench-eval
+
+mkdir -p $SWE_TASK_DIR/reset_testbed_temp
+mkdir -p $SWE_TASK_DIR/reset_testbed_log_dir
+SWE_BENCH_DIR=/swe_util/OH-SWE-bench
+output=$(
+    export PYTHONPATH=$SWE_BENCH_DIR && \
+    cd $SWE_BENCH_DIR && \
+    python swebench/harness/reset_swe_env.py \
+    --swe_bench_tasks $SWEUTIL_DIR/eval_data/instances/swe-bench-test.json \
+    --temp_dir $SWE_TASK_DIR/reset_testbed_temp \
+    --testbed /workspace \
+    --conda_path $SWEUTIL_DIR/miniforge3 \
+    --instance_id $SWE_INSTANCE_ID \
+    --log_dir $SWE_TASK_DIR/reset_testbed_log_dir \
+    --timeout 900 \
+    --verbose
+)
+
+REPO_PATH=$(echo "$output" | awk -F': ' '/repo_path:/ {print $2}')
+TEST_CMD=$(echo "$output" | awk -F': ' '/test_cmd:/ {print $2}')
+echo "Repo Path: $REPO_PATH"
+echo "Test Command: $TEST_CMD"
+
+echo "export SWE_BENCH_DIR=\"$SWE_BENCH_DIR\"" >> ~/.bashrc
+echo "export REPO_PATH=\"$REPO_PATH\"" >> ~/.bashrc
+echo "export TEST_CMD=\"$TEST_CMD\"" >> ~/.bashrc
+
+if [[ "$REPO_PATH" == "None" ]]; then
+    echo "Error: Failed to retrieve repository path. Tests may not have passed or output was not as expected." >&2
+    exit 1
+fi
+
+# Activate instance-specific environment
+. $SWEUTIL_DIR/miniforge3/etc/profile.d/conda.sh
+conda activate $CONDA_ENV_NAME
+
+set +e

frontend/__tests__/components/features/chat/task-tracking-observation-content.test.tsx

@@ -8,10 +8,11 @@ vi.mock("react-i18next", () => ({
   useTranslation: () => ({
     t: (key: string) => {
       const translations: Record<string, string> = {
-        "TASK_TRACKING_OBSERVATION$TASK_LIST": "Task List",
-        "TASK_TRACKING_OBSERVATION$TASK_ID": "ID",
-        "TASK_TRACKING_OBSERVATION$TASK_NOTES": "Notes",
-        "TASK_TRACKING_OBSERVATION$RESULT": "Result",
+        TASK_TRACKING_OBSERVATION$TASK_LIST: "Task List",
+        TASK_TRACKING_OBSERVATION$TASK_ID: "ID",
+        TASK_TRACKING_OBSERVATION$TASK_NOTES: "Notes",
+        TASK_TRACKING_OBSERVATION$RESULT: "Result",
+        COMMON$TASKS: "Tasks",
       };
       return translations[key] || key;
     },
@@ -61,19 +62,26 @@ describe("TaskTrackingObservationContent", () => {
   it("renders task list when command is 'plan' and tasks exist", () => {
     render(<TaskTrackingObservationContent event={mockEvent} />);
 
-    expect(screen.getByText("Task List (3 items)")).toBeInTheDocument();
+    expect(screen.getByText("Tasks")).toBeInTheDocument();
     expect(screen.getByText("Implement feature A")).toBeInTheDocument();
     expect(screen.getByText("Fix bug B")).toBeInTheDocument();
     expect(screen.getByText("Deploy to production")).toBeInTheDocument();
   });
 
   it("displays correct status icons and badges", () => {
-    render(<TaskTrackingObservationContent event={mockEvent} />);
+    const { container } = render(
+      <TaskTrackingObservationContent event={mockEvent} />,
+    );
 
-    // Check for status text (the icons are emojis)
-    expect(screen.getByText("todo")).toBeInTheDocument();
-    expect(screen.getByText("in progress")).toBeInTheDocument();
-    expect(screen.getByText("done")).toBeInTheDocument();
+    // Status is represented by icons, not text. Verify task items are rendered with their titles
+    // which indicates the status icons are present (status affects icon rendering)
+    expect(screen.getByText("Implement feature A")).toBeInTheDocument();
+    expect(screen.getByText("Fix bug B")).toBeInTheDocument();
+    expect(screen.getByText("Deploy to production")).toBeInTheDocument();
+
+    // Verify task items are present (they contain the status icons)
+    const taskItems = container.querySelectorAll('[data-name="item"]');
+    expect(taskItems).toHaveLength(3);
   });
 
   it("displays task IDs and notes", () => {
@@ -84,14 +92,9 @@ describe("TaskTrackingObservationContent", () => {
     expect(screen.getByText("ID: task-3")).toBeInTheDocument();
 
     expect(screen.getByText("Notes: This is a test task")).toBeInTheDocument();
-    expect(screen.getByText("Notes: Completed successfully")).toBeInTheDocument();
-  });
-
-  it("renders result section when content exists", () => {
-    render(<TaskTrackingObservationContent event={mockEvent} />);
-
-    expect(screen.getByText("Result")).toBeInTheDocument();
-    expect(screen.getByText("Task tracking operation completed successfully")).toBeInTheDocument();
+    expect(
+      screen.getByText("Notes: Completed successfully"),
+    ).toBeInTheDocument();
   });
 
   it("does not render task list when command is not 'plan'", () => {
@@ -105,7 +108,7 @@ describe("TaskTrackingObservationContent", () => {
 
     render(<TaskTrackingObservationContent event={eventWithoutPlan} />);
 
-    expect(screen.queryByText("Task List")).not.toBeInTheDocument();
+    expect(screen.queryByText("Tasks")).not.toBeInTheDocument();
   });
 
   it("does not render task list when task list is empty", () => {
@@ -119,17 +122,6 @@ describe("TaskTrackingObservationContent", () => {
 
     render(<TaskTrackingObservationContent event={eventWithEmptyTasks} />);
 
-    expect(screen.queryByText("Task List")).not.toBeInTheDocument();
-  });
-
-  it("does not render result section when content is empty", () => {
-    const eventWithoutContent = {
-      ...mockEvent,
-      content: "",
-    };
-
-    render(<TaskTrackingObservationContent event={eventWithoutContent} />);
-
-    expect(screen.queryByText("Result")).not.toBeInTheDocument();
+    expect(screen.queryByText("Tasks")).not.toBeInTheDocument();
   });
 });

frontend/__tests__/components/features/home/repo-connector.test.tsx

@@ -71,6 +71,7 @@ beforeEach(() => {
     provider_tokens_set: {
       github: "some-token",
       gitlab: null,
+      azure_devops: null,
     },
   });
 });

frontend/__tests__/components/features/home/task-card.test.tsx

@@ -23,6 +23,7 @@ const MOCK_RESPOSITORIES: GitRepository[] = [
   { id: "2", full_name: "repo2", git_provider: "github", is_public: true },
   { id: "3", full_name: "repo3", git_provider: "gitlab", is_public: true },
   { id: "4", full_name: "repo4", git_provider: "gitlab", is_public: true },
+  { id: "5", full_name: "repo5", git_provider: "azure_devops", is_public: true },
 ];
 
 const renderTaskCard = (task = MOCK_TASK_1) => {

frontend/__tests__/posthog-tracking.test.tsx

@@ -0,0 +1,233 @@
+import {
+  describe,
+  it,
+  expect,
+  beforeAll,
+  afterAll,
+  afterEach,
+  vi,
+} from "vitest";
+import { screen, waitFor, render, cleanup } from "@testing-library/react";
+import { QueryClient, QueryClientProvider } from "@tanstack/react-query";
+import { createMockAgentErrorEvent } from "#/mocks/mock-ws-helpers";
+import { ConversationWebSocketProvider } from "#/contexts/conversation-websocket-context";
+import { conversationWebSocketTestSetup } from "./helpers/msw-websocket-setup";
+import { ConnectionStatusComponent } from "./helpers/websocket-test-components";
+
+// Mock the tracking function
+const mockTrackCreditLimitReached = vi.fn();
+
+// Mock useTracking hook
+vi.mock("#/hooks/use-tracking", () => ({
+  useTracking: () => ({
+    trackCreditLimitReached: mockTrackCreditLimitReached,
+    trackLoginButtonClick: vi.fn(),
+    trackConversationCreated: vi.fn(),
+    trackPushButtonClick: vi.fn(),
+    trackPullButtonClick: vi.fn(),
+    trackCreatePrButtonClick: vi.fn(),
+    trackGitProviderConnected: vi.fn(),
+    trackUserSignupCompleted: vi.fn(),
+    trackCreditsPurchased: vi.fn(),
+  }),
+}));
+
+// Mock useActiveConversation hook
+vi.mock("#/hooks/query/use-active-conversation", () => ({
+  useActiveConversation: () => ({
+    data: null,
+    isLoading: false,
+    error: null,
+  }),
+}));
+
+// MSW WebSocket mock setup
+const { wsLink, server: mswServer } = conversationWebSocketTestSetup();
+
+beforeAll(() => {
+  // The global MSW server from vitest.setup.ts is already running
+  // We just need to start our WebSocket-specific server
+  mswServer.listen({ onUnhandledRequest: "bypass" });
+});
+
+afterEach(() => {
+  // Clear all mocks before each test
+  mockTrackCreditLimitReached.mockClear();
+  mswServer.resetHandlers();
+  // Clean up any React components
+  cleanup();
+});
+
+afterAll(async () => {
+  // Close the WebSocket MSW server
+  mswServer.close();
+
+  // Give time for any pending WebSocket connections to close. This is very important to prevent serious memory leaks
+  await new Promise((resolve) => {
+    setTimeout(resolve, 500);
+  });
+});
+
+// Helper function to render components with all necessary providers
+function renderWithProviders(
+  children: React.ReactNode,
+  conversationId = "test-conversation-123",
+  conversationUrl = "http://localhost:3000/api/conversations/test-conversation-123",
+) {
+  const queryClient = new QueryClient({
+    defaultOptions: {
+      queries: { retry: false },
+      mutations: { retry: false },
+    },
+  });
+
+  return render(
+    <QueryClientProvider client={queryClient}>
+      <ConversationWebSocketProvider
+        conversationId={conversationId}
+        conversationUrl={conversationUrl}
+        sessionApiKey={null}
+      >
+        {children}
+      </ConversationWebSocketProvider>
+    </QueryClientProvider>,
+  );
+}
+
+describe("PostHog Analytics Tracking", () => {
+  describe("Credit Limit Tracking", () => {
+    it("should track credit_limit_reached when AgentErrorEvent contains budget error", async () => {
+      // Create a mock AgentErrorEvent with budget-related error message
+      const mockBudgetErrorEvent = createMockAgentErrorEvent({
+        error: "ExceededBudget: Task exceeded maximum budget of $10.00",
+      });
+
+      // Set up MSW to send the budget error event when connection is established
+      mswServer.use(
+        wsLink.addEventListener("connection", ({ client, server }) => {
+          server.connect();
+          // Send the mock budget error event after connection
+          client.send(JSON.stringify(mockBudgetErrorEvent));
+        }),
+      );
+
+      // Render with all providers
+      renderWithProviders(<ConnectionStatusComponent />);
+
+      // Wait for connection to be established
+      await waitFor(() => {
+        expect(screen.getByTestId("connection-state")).toHaveTextContent(
+          "OPEN",
+        );
+      });
+
+      // Wait for the tracking event to be captured
+      await waitFor(() => {
+        expect(mockTrackCreditLimitReached).toHaveBeenCalledWith(
+          expect.objectContaining({
+            conversationId: "test-conversation-123",
+          }),
+        );
+      });
+    });
+
+    it("should track credit_limit_reached when AgentErrorEvent contains 'credit' keyword", async () => {
+      // Create error with "credit" keyword (case-insensitive)
+      const mockCreditErrorEvent = createMockAgentErrorEvent({
+        error: "Insufficient CREDIT to complete this operation",
+      });
+
+      mswServer.use(
+        wsLink.addEventListener("connection", ({ client, server }) => {
+          server.connect();
+          client.send(JSON.stringify(mockCreditErrorEvent));
+        }),
+      );
+
+      renderWithProviders(<ConnectionStatusComponent />);
+
+      await waitFor(() => {
+        expect(screen.getByTestId("connection-state")).toHaveTextContent(
+          "OPEN",
+        );
+      });
+
+      await waitFor(() => {
+        expect(mockTrackCreditLimitReached).toHaveBeenCalledWith(
+          expect.objectContaining({
+            conversationId: "test-conversation-123",
+          }),
+        );
+      });
+    });
+
+    it("should NOT track credit_limit_reached for non-budget errors", async () => {
+      // Create a regular error without budget/credit keywords
+      const mockRegularErrorEvent = createMockAgentErrorEvent({
+        error: "Failed to execute command: Permission denied",
+      });
+
+      mswServer.use(
+        wsLink.addEventListener("connection", ({ client, server }) => {
+          server.connect();
+          client.send(JSON.stringify(mockRegularErrorEvent));
+        }),
+      );
+
+      renderWithProviders(<ConnectionStatusComponent />);
+
+      // Wait for connection and error to be processed
+      await waitFor(() => {
+        expect(screen.getByTestId("connection-state")).toHaveTextContent(
+          "OPEN",
+        );
+      });
+
+      // Verify that credit_limit_reached was NOT tracked
+      expect(mockTrackCreditLimitReached).not.toHaveBeenCalled();
+    });
+
+    it("should only track credit_limit_reached once per error event", async () => {
+      const mockBudgetErrorEvent = createMockAgentErrorEvent({
+        error: "Budget exceeded: $10.00 limit reached",
+      });
+
+      mswServer.use(
+        wsLink.addEventListener("connection", ({ client, server }) => {
+          server.connect();
+          // Send the same error event twice
+          client.send(JSON.stringify(mockBudgetErrorEvent));
+          client.send(
+            JSON.stringify({ ...mockBudgetErrorEvent, id: "different-id" }),
+          );
+        }),
+      );
+
+      renderWithProviders(<ConnectionStatusComponent />);
+
+      await waitFor(() => {
+        expect(screen.getByTestId("connection-state")).toHaveTextContent(
+          "OPEN",
+        );
+      });
+
+      await waitFor(() => {
+        expect(mockTrackCreditLimitReached).toHaveBeenCalledTimes(2);
+      });
+
+      // Both calls should be for credit_limit_reached (once per event)
+      expect(mockTrackCreditLimitReached).toHaveBeenNthCalledWith(
+        1,
+        expect.objectContaining({
+          conversationId: "test-conversation-123",
+        }),
+      );
+      expect(mockTrackCreditLimitReached).toHaveBeenNthCalledWith(
+        2,
+        expect.objectContaining({
+          conversationId: "test-conversation-123",
+        }),
+      );
+    });
+  });
+});

frontend/__tests__/routes/git-settings.test.tsx

@@ -124,6 +124,9 @@ describe("Content", () => {
     await screen.findByTestId("bitbucket-token-input");
     await screen.findByTestId("bitbucket-token-help-anchor");
 
+    await screen.findByTestId("azure-devops-token-input");
+    await screen.findByTestId("azure-devops-token-help-anchor");
+
     getConfigSpy.mockResolvedValue(VALID_SAAS_CONFIG);
     queryClient.invalidateQueries();
     rerender();
@@ -149,6 +152,13 @@ describe("Content", () => {
       expect(
         screen.queryByTestId("bitbucket-token-help-anchor"),
       ).not.toBeInTheDocument();
+
+      expect(
+        screen.queryByTestId("azure-devops-token-input"),
+      ).not.toBeInTheDocument();
+      expect(
+        screen.queryByTestId("azure-devops-token-help-anchor"),
+      ).not.toBeInTheDocument();
     });
   });
 
@@ -287,6 +297,7 @@ describe("Form submission", () => {
       github: { token: "test-token", host: "" },
       gitlab: { token: "", host: "" },
       bitbucket: { token: "", host: "" },
+      azure_devops: { token: "", host: "" },
     });
   });
 
@@ -308,6 +319,7 @@ describe("Form submission", () => {
       github: { token: "", host: "" },
       gitlab: { token: "test-token", host: "" },
       bitbucket: { token: "", host: "" },
+      azure_devops: { token: "", host: "" },
     });
   });
 
@@ -329,6 +341,29 @@ describe("Form submission", () => {
       github: { token: "", host: "" },
       gitlab: { token: "", host: "" },
       bitbucket: { token: "test-token", host: "" },
+      azure_devops: { token: "", host: "" },
+    });
+  });
+
+  it("should save the Azure DevOps token", async () => {
+    const saveProvidersSpy = vi.spyOn(SecretsService, "addGitProvider");
+    saveProvidersSpy.mockImplementation(() => Promise.resolve(true));
+    const getConfigSpy = vi.spyOn(OptionService, "getConfig");
+    getConfigSpy.mockResolvedValue(VALID_OSS_CONFIG);
+
+    renderGitSettingsScreen();
+
+    const azureDevOpsInput = await screen.findByTestId("azure-devops-token-input");
+    const submit = await screen.findByTestId("submit-button");
+
+    await userEvent.type(azureDevOpsInput, "test-token");
+    await userEvent.click(submit);
+
+    expect(saveProvidersSpy).toHaveBeenCalledWith({
+      github: { token: "", host: "" },
+      gitlab: { token: "", host: "" },
+      bitbucket: { token: "", host: "" },
+      azure_devops: { token: "test-token", host: "" },
     });
   });
 

frontend/__tests__/stores/use-event-store.test.ts

@@ -55,7 +55,7 @@ const mockObservationEvent: ObservationEvent = {
   tool_call_id: "call_123",
   observation: {
     kind: "ExecuteBashObservation",
-    output: "hello\n",
+    content: [{ type: "text", text: "hello\n" }],
     command: "echo hello",
     exit_code: 0,
     error: false,

frontend/__tests__/utils/convert-raw-providers-to-list.test.ts

@@ -7,6 +7,7 @@ describe("convertRawProvidersToList", () => {
     const example1: Partial<Record<Provider, string | null>> | undefined = {
       github: "test-token",
       gitlab: "test-token",
+      azure_devops: "test-token",
     };
     const example2: Partial<Record<Provider, string | null>> | undefined = {
       github: "",
@@ -14,9 +15,13 @@ describe("convertRawProvidersToList", () => {
     const example3: Partial<Record<Provider, string | null>> | undefined = {
       gitlab: null,
     };
+    const example4: Partial<Record<Provider, string | null>> | undefined = {
+      azure_devops: "test-token",
+    };
 
-    expect(convertRawProvidersToList(example1)).toEqual(["github", "gitlab"]);
+    expect(convertRawProvidersToList(example1)).toEqual(["github", "gitlab", "azure_devops"]);
     expect(convertRawProvidersToList(example2)).toEqual(["github"]);
     expect(convertRawProvidersToList(example3)).toEqual(["gitlab"]);
+    expect(convertRawProvidersToList(example4)).toEqual(["azure_devops"]);
   });
 });

frontend/__tests__/utils/handle-event-for-ui.test.ts

@@ -17,7 +17,7 @@ describe("handleEventForUI", () => {
     tool_call_id: "call_123",
     observation: {
       kind: "ExecuteBashObservation",
-      output: "hello\n",
+      content: [{ type: "text", text: "hello\n" }],
       command: "echo hello",
       exit_code: 0,
       error: false,

frontend/package.json

@@ -7,7 +7,7 @@
     "node": ">=22.0.0"
   },
   "dependencies": {
-    "@heroui/react": "^2.8.4",
+    "@heroui/react": "2.8.5",
     "@heroui/use-infinite-scroll": "^2.2.11",
     "@microlink/react-json-view": "^1.26.2",
     "@monaco-editor/react": "^4.7.0-rc.0",
@@ -38,7 +38,7 @@
     "jose": "^6.1.0",
     "lucide-react": "^0.544.0",
     "monaco-editor": "^0.53.0",
-    "posthog-js": "^1.290.0",
+    "posthog-js": "^1.298.1",
     "react": "^19.1.1",
     "react-dom": "^19.1.1",
     "react-highlight": "^0.15.0",

frontend/public/safari-pinned-tab.svg

@@ -1,32 +1,7 @@
-<svg version="1.2" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 1365 1365" width="1365" height="1365">
-	<title>safari-pinned-tab-svg</title>
-	<defs>
-		<clipPath clipPathUnits="userSpaceOnUse" id="cp1">
-			<path d="m655.2 313.1v822.23h-622.69v-822.23z"/>
-		</clipPath>
-		<clipPath clipPathUnits="userSpaceOnUse" id="cp2">
-			<path d="m1308.84 304v828.5h-617.24v-828.5z"/>
-		</clipPath>
-	</defs>
-	<style>
-		.s0 { fill: none }
-		.s1 { fill: #000000 }
-	</style>
-	<g id="surface1">
-		<path class="s0" d="m1258.6 499.2c-40.8-26.1-68 13.8-64.7 68.1l-0.3 0.4c0.1-56.6-7.3-119.2-31.7-169.8-8.7-17.9-26.2-47.4-61-33.5-15.2 6.1-29 24.4-21.9 71.7 0 0 8 49.7 6.5 112.2v0.8c-9.9-172.2-47.3-224.7-100.7-221.3-17.1 3.1-40.5 10.8-32.6 63.8 0 0 8.5 55.2 11.3 99.2l0.2 2.2h-0.2c-25.2-94.9-59-96.2-83.5-92.5-22.3 3.3-46.6 27.3-34.3 74.7 38.6 148.4 31 327.2 28.2 352.9-7.9-17.6-10.3-31.4-21.3-50.7-43.9-77.1-64.8-82.8-90.4-84-25.5-1.1-53 15.2-51.2 46.3 1.9 31.1 17.1 36.3 38.7 79.6 16.9 33.8 21.7 78 55.7 158.4 28.1 66.5 101.6 139.5 235.6 130.8 108.5-3.7 270.6-43.2 242.4-302.5-7-45-1.7-82.7 1.9-121.4 5.8-60 14.1-159.4-26.7-185.5z"/>
-		<path class="s0" d="m580.9 695.3c-25.7 1.7-46.4 7.7-89 85.7-10.6 19.4-12.7 33.3-20.3 50.9-3.3-25.5-14-204.2 21.9-353.4 11.4-47.5-13.3-71-35.6-73.9-24.6-3.3-58.5-1.3-81.9 94.6h-0.3l0.4-2.8c1.9-44 9.5-99.4 9.5-99.4 6.9-53.1-16.6-60.3-33.7-63.2-53.4-2.3-89.7 50.4-96.7 221.1h-0.2c-2.4-61.8 4.6-111 4.6-111 6.3-47.5-7.9-65.5-23.2-71.3-35-13.3-52 16.6-60.3 34.7-23.6 51-29.9 113.7-28.7 170.3l-0.4-0.4c2.4-54.2-25.6-93.7-65.9-66.9-40.3 26.9-30.1 126.2-23.4 186 4.5 38.6 10.4 76.2 4.1 121.4-23.5 259.7 139.3 296.1 247.8 297.8 134.1 6.1 206.3-68.3 233.3-135.4 32.5-80.9 36.6-125.3 52.8-159.3 20.8-43.8 36-49.2 37.3-80.3 1.3-31.1-26.5-46.9-52-45.3z"/>
-		<g id="Clip-Path" clip-path="url(#cp1)">
-			<g>
-				<path fill-rule="evenodd" class="s1" d="m634.4 695.7c11.9 12 17.8 27.9 17.1 45.7-1 24.6-9.5 37.8-19.4 53.2-5.8 9-12.3 19.2-19.8 34.9-6.6 13.8-11.2 30.4-17 51.5-7.6 27.3-17 61.2-35.3 106.7-14.1 35.1-71.4 146.1-231.3 147.6q-9.8 0.1-20-0.3c-93.6-1.5-164.2-28.5-209.3-80.6-46.9-54-65.8-134.2-56.4-238.4l0.1-0.9c5.2-37.6 1.5-69.5-2.6-103.3-0.5-4.4-1-8.7-1.5-13.1-9.4-82.5-15.4-173.1 31.8-204.6 27.9-18.5 49.3-11.1 59.6-4.9 1 0.6 1.9 1.2 2.9 1.9 4.5-32.3 12.6-63.9 25.6-92.2 11.2-24.2 37.1-62.2 83.8-44.5 7.6 3 17.3 8.8 24.8 20.2 6.7-14.7 14.5-26.6 23.5-35.8 16.6-17.2 37.4-25.4 61.6-24.3l2.1 0.2c39.2 6.5 55.9 35 49.4 84.9 0 0 0 0.3 0 0.6 20-17 40.4-16.9 56.1-14.8 17.2 2.2 32.8 12.1 42.8 27.2 8.6 13 17.1 35.8 8.7 70.7-22.3 92.9-26.1 198.5-25.2 268.8 36.3-61.5 59.9-74.1 93.2-76.2 20.8-1.3 41.3 6 54.8 19.8zm-316.9-329.4c-35.1 36.2-49.7 148.5-43.4 333.7 19.5-3 39.5-5.2 59.6-6.5 4.7-91.2 13.3-153.7 23.7-197q0-0.5 0-0.9c2-44.4 9.3-98.9 9.7-101.2 4.7-36.3-5.7-39.2-17-41.1-13.2-0.3-23.5 3.8-32.5 13zm-124.6 49.5c-50 108.3-16.2 277.2-8.5 303.8 16.1-4.8 33.8-9.2 52.4-13-2.1-57.7-2.3-108.1-0.6-151.9-2.3-62.3 4.4-111.4 4.7-113.5 1.8-13 4.2-44.5-11.1-50.3-10.9-4.1-22.8-5.7-36.9 24.8zm407.9 357.3c8.9-13.8 12.6-19.5 13.2-33.3 0.2-6.7-1.6-12-5.9-16.3-5.9-6.1-16-9.4-26.1-8.8-17.3 1.2-33.6 2.2-73.8 75.9-5.6 10.4-8.5 18.9-11.8 28.6-2.1 6.3-4.4 13.2-7.7 20.8-0.1 0.3-0.2 0.6-0.4 0.8-1.4 3.3-3 6.8-4.8 10.4-17.6 34.2-46.2 53.4-76.5 51.6-10.4-0.7-18.2-9.9-17.6-20.6 0.6-10.7 9.6-18.7 19.8-18.2 18 1.1 33.1-15.3 41.2-31.1 0.7-1.4 1.3-2.8 2-4.2-4-41.1-11.8-210.7 22.9-354.8 3.9-16.5 2.8-30.1-3.3-39.3-5.6-8.4-13.4-10.3-16.5-10.7-11.2-1.5-19.3-0.9-27.8 6.4-20.5 17.8-46.8 77.8-56.4 261.8 5.8 0 11.6 0 17.3 0.2 10.3 0.3 18.5 9.2 18.2 19.9-0.3 10.7-8.8 19.1-19.2 18.9-95.8-2.8-204.4 22.8-250.2 47.9-2.8 1.5-5.7 2.3-8.6 2.3-6.8 0.1-13.4-3.7-16.8-10.3-4.8-9.5-1.4-21.2 7.8-26.3 8.1-4.4 17.9-8.9 28.9-13.1-6.7-22.4-18.6-82-20.1-150.6-0.3-1.5-0.5-3-0.4-4.6 1.2-29.4-7.6-48.4-16.4-53.6-5.2-3.1-12.1-1.8-20.6 3.9-31.6 21-19.4 127.4-14.9 167.4 0.5 4.3 1 8.6 1.5 12.9 4.1 34.7 8.4 70.6 2.6 113-8.3 92.7 7.5 162.9 47 208.5 38.4 44.2 98 66.3 182.4 67.6 151.5 7.1 203.3-92.6 215.7-123.3 17.4-43.4 26.5-76.2 33.8-102.5 6.4-22.9 11.4-41 19.5-58 8.5-17.9 16.1-29.7 22.2-39.2z"/>
-			</g>
-		</g>
-		<g id="Clip-Path" clip-path="url(#cp2)">
-			<g>
-				<path fill-rule="evenodd" class="s1" d="m1301.9 802.9l0.1 0.9c11.3 104-6.3 184.6-52.2 239.5-44.1 52.8-114.2 81.3-208.3 84.5q-10.2 0.7-19.9 0.8c-159.5 1.6-218.8-108.4-233.5-143.2-19.1-45.1-29.1-78.9-37.2-106-6.2-21-11.1-37.5-18-51.2-7.7-15.5-14.4-25.6-20.4-34.5-10.1-15.1-18.8-28.2-20.3-52.8-1.1-17.7 4.5-33.6 16.2-45.9 13.3-14 33.8-21.8 54.5-20.9 33.3 1.5 57.1 13.7 94.5 74.5-0.3-70.4-6-175.9-30-268.4-9-34.8-0.9-57.7 7.5-70.8 9.7-15.3 25.1-25.5 42.2-28.1 15.7-2.3 36.2-2.9 56.4 13.8 0-0.2 0-0.4 0-0.4-7.4-49.9 8.8-78.8 47.9-86l2.1-0.3c24.1-1.6 45 6.3 62 23.2 9.1 9 17.1 20.7 24.1 35.3 7.4-11.6 16.9-17.6 24.5-20.6 46.4-18.7 72.9 18.9 84.5 42.9 13.5 28 22.1 59.5 27.3 91.6 0.9-0.6 1.8-1.3 2.8-1.9 10.2-6.3 31.5-14.3 59.7 3.8 47.8 30.6 43.4 121.3 35.5 203.9-0.4 4.4-0.8 8.7-1.3 13.1-3.4 33.9-6.6 65.9-0.8 103.3zm-197.6-256.5c2.5 43.7 3.2 94.2 2.1 151.9 18.7 3.5 36.4 7.5 52.7 12 7.2-26.7 37.9-196.3-14-303.7-14.6-30.1-26.5-28.4-37.4-24.1-15.1 6.1-12.2 37.5-10.2 50.7 0.4 1.9 8 50.9 6.8 113.2zm-117.5-199.2c-11.4 2.2-21.6 5.2-16.3 41.6 0.4 2.1 8.8 56.4 11.5 100.8q0 0.5 0 0.9c11.2 43.2 21 105.4 27.2 196.6 20.2 0.9 40.2 2.7 59.7 5.3 3-185.3-13.5-297.2-49.3-332.8-9.1-9.1-19.5-13-32.7-12.4zm278.5 348.5c0.5-4.3 0.9-8.6 1.3-13 3.9-40.1 14.1-146.6-17.9-167-8.6-5.5-15.5-6.7-20.6-3.5-8.7 5.4-17.2 24.5-15.4 53.9 0.1 1.5 0 3.1-0.3 4.6-0.4 68.5-11.2 128.4-17.5 150.9 11.1 4.1 20.9 8.3 29.1 12.6 9.3 4.9 13 16.6 8.3 26.2-3.3 6.7-9.8 10.6-16.6 10.6-2.9 0-5.9-0.6-8.6-2.1-46.2-24.3-155.4-47.7-251-43.2-10.5 0.3-19.2-7.7-19.6-18.5-0.5-10.7 7.5-19.8 17.9-20.2 5.6-0.3 11.4-0.5 17.2-0.6-12.8-183.8-40.2-243.3-61-260.7-8.6-7.1-16.8-7.6-27.9-5.9-3.2 0.5-10.9 2.5-16.4 11.1-5.9 9.3-6.8 22.9-2.5 39.3 37.3 143.5 32.4 313.2 29.2 354.3 0.7 1.4 1.3 2.7 2.1 4.2 8.4 15.6 23.7 31.7 41.7 30.3 10.3-0.7 19.3 7.2 20.1 17.9 0.8 10.6-6.9 20-17.2 20.8-30.3 2.5-59.3-16.3-77.4-50.1-2-3.6-3.6-7-5.1-10.3-0.1-0.2-0.2-0.5-0.3-0.7-3.5-7.5-5.9-14.5-8.2-20.8-3.5-9.6-6.4-18-12.3-28.3-41.6-72.9-57.9-73.7-75.1-74.5-10.2-0.5-20.2 3.1-26.1 9.3-4.1 4.3-5.9 9.7-5.5 16.4 0.8 13.8 4.6 19.4 13.7 33.1 6.3 9.4 14.1 21 23 38.7 8.3 16.9 13.7 34.8 20.5 57.7 7.7 26.2 17.4 58.7 35.6 101.8 12.9 30.5 66.7 129.1 217.3 119.2 84.9-2.9 144.2-26.2 181.7-71.1 38.7-46.3 53.2-116.8 43.3-209.4-6.6-42.3-3-78.2 0.5-113z"/>
-			</g>
-		</g>
-		<path class="s1" d="m739.8 434.2c-3 0-6.1-0.6-9-2-10.5-5-15.2-18-10.3-28.9 15.7-35.6 38.8-68.4 66.6-94.9 8.5-8.1 21.9-7.6 29.7 1.3 7.9 8.9 7.4 22.7-1.2 30.8-23.7 22.7-43.4 50.6-56.8 81-3.6 7.9-11.1 12.6-19 12.7z"/>
-		<path class="s1" d="m668.8 421.6c-10.9 0.1-20.3-8.5-21.2-20-4-51.4-4.2-103.5-0.4-154.8 0.9-12 11.1-21 22.6-20.1 11.6 0.9 20.3 11.3 19.4 23.3-3.6 49.1-3.4 98.9 0.4 148 1 12-7.7 22.5-19.3 23.5-0.5 0-1 0-1.5 0z"/>
-		<path class="s1" d="m596.2 435.1c-9.4 0.1-18.2-6.5-20.6-16.4-8.9-36.3-25.9-70.8-48.9-99.7-7.4-9.3-6.1-23 2.8-30.7 9-7.6 22.3-6.3 29.7 3 26.9 33.8 46.7 74.2 57.2 116.7 2.9 11.6-4 23.5-15.2 26.5-1.8 0.4-3.4 0.6-5.1 0.7z"/>
-	</g>
+<svg width="1365" height="1365" viewBox="0 -24 148 148" fill="none" xmlns="http://www.w3.org/2000/svg">
+<path d="M71.7542 16.863V2.97414C71.7542 1.82355 72.6872 0.890503 73.8378 0.890503C74.9884 0.890503 75.9214 1.82355 75.9214 2.97414V16.863C75.9214 18.0136 74.9884 18.9466 73.8378 18.9466C72.6872 18.9466 71.7542 18.0136 71.7542 16.863Z" fill="black"/>
+<path d="M82.5272 18.9329L89.4716 6.90477C90.0469 5.90832 91.3215 5.5668 92.3179 6.1421C93.3144 6.7174 93.6559 7.99197 93.0806 8.98841L86.1362 21.0165C85.5609 22.0129 84.2863 22.3545 83.2899 21.7792C82.2934 21.2039 81.9519 19.9293 82.5272 18.9329Z" fill="black"/>
+<path d="M65.1481 18.9329L58.2037 6.90477C57.6284 5.90832 56.3538 5.5668 55.3574 6.1421C54.3609 6.7174 54.0194 7.99197 54.5947 8.98841L61.5391 21.0165C62.1144 22.0129 63.389 22.3545 64.3854 21.7792C65.3819 21.2039 65.7234 19.9293 65.1481 18.9329Z" fill="black"/>
+<path d="M140.606 62.0292C140.606 58.409 141.583 47.6748 141.89 44.1323C142.097 41.7374 141.809 40.4247 141.424 39.7542C141.141 39.2626 140.699 38.915 139.634 38.8436C138.865 38.7921 138.027 39.0114 137.401 39.5761C136.814 40.1052 136.159 41.1682 136.159 43.3176L136.155 43.4388L135.198 59.758C135.164 60.3451 134.883 60.8911 134.424 61.2599C133.966 61.6284 133.374 61.7859 132.793 61.6941L122.764 60.1068L111.948 58.6703C110.949 58.5376 110.188 57.7084 110.142 56.7016L109.561 44.1323C109.535 43.621 109.51 43.1141 109.484 42.6146C109.241 37.9294 109.022 33.7805 109.022 32.4282C109.022 28.3859 108.338 26.6806 107.74 25.9634C107.263 25.3915 106.577 25.1402 105.11 25.1402C104.583 25.1402 104.212 25.2481 103.933 25.4111C103.659 25.5714 103.346 25.8587 103.049 26.4208C102.41 27.6257 101.945 29.891 102.118 33.8479C102.342 38.9804 102.692 42.8146 103.035 46.2718C103.377 49.7231 103.718 52.8561 103.908 56.4971C104.204 62.1966 104.178 66.1256 103.945 68.7924C103.828 70.124 103.656 71.1996 103.423 72.0501C103.202 72.8558 102.871 73.6757 102.296 74.2887C101.6 75.0303 100.608 75.3844 99.577 75.136C98.7592 74.9389 98.1847 74.4215 97.8706 74.0916C97.2141 73.4017 96.7501 72.5106 96.568 72.0512C95.5097 69.3812 92.2352 63.1808 87.8023 59.6811C86.5089 58.6599 85.5666 58.3652 84.9736 58.3204C84.4148 58.2783 84.0094 58.4436 83.6909 58.6967C83.34 58.9756 83.0781 59.3811 82.9479 59.7643C82.9019 59.8999 82.8823 59.9968 82.8741 60.0584C84.0759 62.0865 88.8421 69.5222 91.0896 77.069C92.7648 82.6941 96.8038 88.4259 99.8194 90.8809C102.74 93.258 107.988 94.7313 113.9 95.0218C119.756 95.3095 125.788 94.4121 130.033 92.5092C138.233 88.8334 139.903 80.7382 140.651 77.2292C141.232 74.5057 141.243 71.5987 141.087 68.9009C141.01 67.5551 140.894 66.2969 140.793 65.1373C140.695 64.0105 140.606 62.9215 140.606 62.0292ZM120.986 27.0953C120.986 25.8314 120.648 24.7049 120.089 23.9514C119.583 23.27 118.84 22.7987 117.646 22.7984C116.668 22.7982 116.011 22.9187 115.546 23.1167C115.13 23.2943 114.781 23.5699 114.463 24.0831C113.73 25.2671 113.192 27.6455 113.189 32.384L113.721 43.9088C113.91 47.5661 114.106 51.4922 114.235 54.7707L120.986 55.6666V27.0953ZM125.153 56.2652L131.172 57.218L131.992 43.267V32.5083C131.992 31.031 131.39 30.1275 130.678 29.5489C129.884 28.9039 128.957 28.6731 128.519 28.6731C127.722 28.6731 126.899 28.797 126.306 29.2179C125.849 29.5421 125.153 30.3087 125.153 32.5083V56.2652ZM136.159 35.4278C137.406 34.8069 138.74 34.6083 139.912 34.6868C142.037 34.8292 143.91 35.718 145.037 37.6779C146.06 39.4592 146.273 41.8136 146.041 44.4927C145.72 48.1949 144.772 58.6457 144.772 62.0292C144.772 62.708 144.843 63.6116 144.944 64.7758C145.042 65.907 145.165 67.2389 145.247 68.6606C145.411 71.4987 145.422 74.8383 144.727 78.0987C144.002 81.4953 142.041 91.6918 131.738 96.3108C126.731 98.5551 120.002 99.4936 113.696 99.1838C107.445 98.8767 101.128 97.3189 97.1887 94.1122C93.4809 91.0937 88.9938 84.6307 87.0962 78.2589C84.9529 71.0619 80.3109 63.9646 79.1527 61.9533C78.4706 60.7689 78.684 59.3628 79.0019 58.4258C79.3607 57.3688 80.0554 56.2631 81.0993 55.4337C82.1758 54.5784 83.6043 54.0377 85.2876 54.1647C86.9369 54.2893 88.6462 55.0393 90.3834 56.4107C94.8541 59.9401 98.1342 65.5082 99.7424 68.9231C99.759 68.7664 99.779 68.6024 99.7941 68.4298C100.003 66.0435 100.039 62.3344 99.7467 56.7132C99.5635 53.1942 99.2356 50.1809 98.8888 46.6828C98.5425 43.1904 98.184 39.2713 97.955 34.0302C97.7722 29.8481 98.2012 26.6722 99.3672 24.471C99.9716 23.3302 100.79 22.4223 101.83 21.814C102.866 21.2087 103.995 20.974 105.11 20.974C106.759 20.974 108.813 21.2062 110.448 22.7678C110.593 22.4576 110.75 22.1652 110.921 21.8899C111.676 20.6698 112.681 19.8084 113.912 19.2835C115.095 18.7791 116.378 18.6309 117.646 18.6311C120.195 18.6315 122.164 19.7567 123.434 21.4683C124.256 22.576 124.75 23.8775 124.985 25.1982C126.338 24.5876 127.691 24.5068 128.519 24.5068C129.933 24.5068 131.784 25.0791 133.305 26.3154C134.908 27.6179 136.159 29.6733 136.159 32.5083V35.4278Z" fill="black"/>
+<path d="M7.15661 62.0292C7.15661 58.409 6.17994 47.6748 5.87291 44.1323C5.6654 41.7374 5.95357 40.4247 6.33875 39.7542C6.62116 39.2626 7.06336 38.915 8.12834 38.8436C8.89759 38.7921 9.73544 39.0114 10.3616 39.5761C10.9484 40.1052 11.6032 41.1682 11.6032 43.3176L11.6074 43.4388L12.5644 59.758C12.5988 60.3451 12.8798 60.8911 13.338 61.2599C13.7961 61.6284 14.3887 61.7859 14.9695 61.6941L24.9988 60.1068L35.8143 58.6703C36.8135 58.5376 37.5741 57.7084 37.6208 56.7016L38.2015 44.1323C38.2279 43.621 38.2525 43.1141 38.2784 42.6146C38.5218 37.9294 38.7401 33.7805 38.7401 32.4282C38.7401 28.3859 39.4246 26.6806 40.0227 25.9634C40.4996 25.3915 41.185 25.1402 42.6523 25.1402C43.1794 25.1402 43.5505 25.2481 43.8295 25.4111C44.1038 25.5714 44.416 25.8587 44.7138 26.4208C45.3521 27.6257 45.8173 29.891 45.6444 33.8479C45.4201 38.9804 45.0703 42.8146 44.7275 46.2718C44.3853 49.7231 44.0443 52.8561 43.8548 56.4971C43.5582 62.1966 43.5847 66.1256 43.8179 68.7924C43.9344 70.124 44.1069 71.1996 44.3396 72.0501C44.5601 72.8558 44.891 73.6757 45.4663 74.2887C46.1625 75.0303 47.1546 75.3844 48.1855 75.136C49.0033 74.9389 49.5778 74.4215 49.8918 74.0916C50.5484 73.4017 51.0123 72.5106 51.1945 72.0512C52.2527 69.3812 55.5272 63.1808 59.9601 59.6811C61.2536 58.6599 62.1958 58.3652 62.7889 58.3204C63.3476 58.2783 63.753 58.4436 64.0715 58.6967C64.4225 58.9756 64.6844 59.3811 64.8146 59.7643C64.8606 59.8999 64.8801 59.9968 64.8883 60.0584C63.6866 62.0865 58.9204 69.5222 56.6729 77.069C54.9977 82.6941 50.9586 88.4259 47.9431 90.8809C45.0229 93.258 39.7747 94.7313 33.8624 95.0218C28.0068 95.3095 21.9748 94.4121 17.7297 92.5092C9.52988 88.8334 7.85961 80.7382 7.11129 77.2292C6.53054 74.5057 6.5195 71.5987 6.67496 68.9009C6.75251 67.5551 6.86809 66.2969 6.96901 65.1373C7.06707 64.0105 7.1566 62.9215 7.15661 62.0292ZM26.7768 27.0953C26.7768 25.8314 27.1147 24.7049 27.6737 23.9514C28.1792 23.27 28.9221 22.7987 30.1167 22.7984C31.0942 22.7982 31.7518 22.9187 32.2162 23.1167C32.6326 23.2943 32.9817 23.5699 33.2996 24.0831C34.0328 25.2671 34.5705 27.6455 34.5738 32.384L34.0416 43.9088C33.8524 47.5661 33.6565 51.4922 33.5273 54.7707L26.7768 55.6666V27.0953ZM22.6095 56.2652L16.5904 57.218L15.7705 43.267V32.5083C15.7705 31.031 16.3726 30.1275 17.0847 29.5489C17.8785 28.9039 18.8058 28.6731 19.2432 28.6731C20.0404 28.6731 20.8634 28.797 21.4565 29.2179C21.9131 29.5421 22.6095 30.3087 22.6095 32.5083V56.2652ZM11.6032 35.4278C10.3568 34.8069 9.02265 34.6083 7.8501 34.6868C5.72541 34.8292 3.85197 35.718 2.72584 37.6779C1.70247 39.4592 1.48924 41.8136 1.72143 44.4927C2.0423 48.1949 2.99038 58.6457 2.99038 62.0292C2.99037 62.708 2.91991 63.6116 2.81859 64.7758C2.72014 65.907 2.59699 67.2389 2.51505 68.6606C2.3515 71.4987 2.34041 74.8383 3.0357 78.0987C3.76005 81.4953 5.72154 91.6918 16.0245 96.3108C21.0311 98.5551 27.7601 99.4936 34.0669 99.1838C40.3172 98.8767 46.6346 97.3189 50.5737 94.1122C54.2816 91.0937 58.7686 84.6307 60.6662 78.2589C62.8095 71.0619 67.4515 63.9646 68.6098 61.9533C69.2919 60.7689 69.0785 59.3628 68.7605 58.4258C68.4018 57.3688 67.707 56.2631 66.6632 55.4337C65.5867 54.5784 64.1582 54.0377 62.4748 54.1647C60.8256 54.2893 59.1162 55.0393 57.379 56.4107C52.9083 59.9401 49.6283 65.5082 48.02 68.9231C48.0034 68.7664 47.9835 68.6024 47.9684 68.4298C47.7597 66.0435 47.7232 62.3344 48.0158 56.7132C48.1989 53.1942 48.5269 50.1809 48.8737 46.6828C49.22 43.1904 49.5784 39.2713 49.8075 34.0302C49.9903 29.8481 49.5612 26.6722 48.3952 24.471C47.7909 23.3302 46.9729 22.4223 45.9321 21.814C44.8964 21.2087 43.7676 20.974 42.6523 20.974C41.0038 20.974 38.9497 21.2062 37.3141 22.7678C37.1698 22.4576 37.0124 22.1652 36.8419 21.8899C36.0863 20.6698 35.0817 19.8084 33.8508 19.2835C32.6679 18.7791 31.3849 18.6309 30.1167 18.6311C27.5677 18.6315 25.5986 19.7567 24.3285 21.4683C23.5066 22.576 23.0121 23.8775 22.7771 25.1982C21.4247 24.5876 20.0718 24.5068 19.2432 24.5068C17.8298 24.5068 15.9788 25.0791 14.4573 26.3154C12.8542 27.6179 11.6032 29.6733 11.6032 32.5083V35.4278Z" fill="black"/>
 </svg>

frontend/src/api/conversation-service/v1-conversation-service.api.ts

@@ -60,6 +60,8 @@ class V1ConversationService {
     selected_branch?: string,
     conversationInstructions?: string,
     trigger?: ConversationTrigger,
+    parent_conversation_id?: string,
+    agent_type?: "default" | "plan",
   ): Promise<V1AppConversationStartTask> {
     const body: V1AppConversationStartRequest = {
       selected_repository: selectedRepository,
@@ -67,6 +69,8 @@ class V1ConversationService {
       selected_branch,
       title: conversationInstructions,
       trigger,
+      parent_conversation_id: parent_conversation_id || null,
+      agent_type,
     };
 
     // Add initial message if provided
@@ -111,11 +115,11 @@ class V1ConversationService {
    * Search for start tasks (ongoing tasks that haven't completed yet)
    * Use this to find tasks that were started but the user navigated away
    *
-   * Note: Backend only supports filtering by limit. To filter by repository/trigger,
+   * Note: Backend supports filtering by limit and created_at__gte. To filter by repository/trigger,
    * filter the results client-side after fetching.
    *
    * @param limit Maximum number of tasks to return (max 100)
-   * @returns Array of start tasks
+   * @returns Array of start tasks from the last 20 minutes
    */
   static async searchStartTasks(
     limit: number = 100,
@@ -123,6 +127,10 @@ class V1ConversationService {
     const params = new URLSearchParams();
     params.append("limit", limit.toString());
 
+    // Only get tasks from the last 20 minutes
+    const twentyMinutesAgo = new Date(Date.now() - 20 * 60 * 1000);
+    params.append("created_at__gte", twentyMinutesAgo.toISOString());
+
     const { data } = await openHands.get<V1AppConversationStartTaskPage>(
       `/api/v1/app-conversations/start-tasks/search?${params.toString()}`,
     );
@@ -288,6 +296,25 @@ class V1ConversationService {
     const { data } = await openHands.get<{ runtime_id: string }>(url);
     return data;
   }
+
+  /**
+   * Read a file from a specific conversation's sandbox workspace
+   * @param conversationId The conversation ID
+   * @param filePath Path to the file to read within the sandbox workspace (defaults to /workspace/project/PLAN.md)
+   * @returns The content of the file or an empty string if the file doesn't exist
+   */
+  static async readConversationFile(
+    conversationId: string,
+    filePath: string = "/workspace/project/PLAN.md",
+  ): Promise<string> {
+    const params = new URLSearchParams();
+    params.append("file_path", filePath);
+
+    const { data } = await openHands.get<string>(
+      `/api/v1/app-conversations/${conversationId}/file?${params.toString()}`,
+    );
+    return data;
+  }
 }
 
 export default V1ConversationService;

frontend/src/api/conversation-service/v1-conversation-service.types.ts

@@ -3,15 +3,19 @@ import { Provider } from "#/types/settings";
 import { V1SandboxStatus } from "../sandbox-service/sandbox-service.types";
 
 // V1 API Types for requests
-// Note: This represents the serialized API format, not the internal TextContent/ImageContent types
-export interface V1MessageContent {
-  type: "text" | "image_url";
-  text?: string;
-  image_url?: {
-    url: string;
-  };
+// These types match the SDK's TextContent and ImageContent formats
+export interface V1TextContent {
+  type: "text";
+  text: string;
 }
 
+export interface V1ImageContent {
+  type: "image";
+  image_urls: string[];
+}
+
+export type V1MessageContent = V1TextContent | V1ImageContent;
+
 type V1Role = "user" | "system" | "assistant" | "tool";
 
 export interface V1SendMessageRequest {
@@ -30,6 +34,8 @@ export interface V1AppConversationStartRequest {
   title?: string | null;
   trigger?: ConversationTrigger | null;
   pr_number?: number[];
+  parent_conversation_id?: string | null;
+  agent_type?: "default" | "plan";
 }
 
 export type V1AppConversationStartTaskStatus =
@@ -38,6 +44,7 @@ export type V1AppConversationStartTaskStatus =
   | "PREPARING_REPOSITORY"
   | "RUNNING_SETUP_SCRIPT"
   | "SETTING_UP_GIT_HOOKS"
+  | "SETTING_UP_SKILLS"
   | "STARTING_CONVERSATION"
   | "READY"
   | "ERROR";

frontend/src/api/open-hands.types.ts

@@ -77,6 +77,7 @@ export interface Conversation {
   session_api_key: string | null;
   pr_number?: number[] | null;
   conversation_version?: "V0" | "V1";
+  sub_conversation_ids?: string[];
 }
 
 export interface ResultSet<T> {

frontend/src/assets/branding/azure-devops-logo.svg

@@ -0,0 +1 @@
+<svg height="24" viewBox="0 0 24 24" width="24" xmlns="http://www.w3.org/2000/svg"><path fill="currentColor" d="m22 18-5 4-8-3v3l-4.19-5.75 12.91 1.05v-10.96l4.28-.69zm-17.19-1.75v-7.29l12.91-2.62-7.12-4.34v2.84l-6.63 1.92-1.97 2.62v5.69z"/></svg>

frontend/src/assets/branding/openhands-logo.svg

@@ -1,16 +1,9 @@
-<svg xmlns="http://www.w3.org/2000/svg" width="47" height="30" viewBox="0 0 47 30" fill="none">
-  <g clip-path="url(#clip0_10905_18559)">
-    <path d="M44.731 8.9991C43.271 8.13859 42.2956 9.4574 42.4152 11.248L42.4031 11.2616C42.4071 9.39165 42.1435 7.32642 41.2675 5.65567C40.9573 5.06395 40.3287 4.09128 39.0856 4.54957C38.5402 4.75068 38.0454 5.35594 38.3009 6.9184C38.3009 6.9184 38.5848 8.55821 38.532 10.6196V10.6486C38.1772 4.96339 36.8388 3.22883 34.9246 3.34099C34.3122 3.44541 33.4748 3.69873 33.7566 5.44683C33.7566 5.44683 34.0628 7.27034 34.1622 8.72258L34.1683 8.79606H34.1622C33.2618 5.66147 32.0492 5.61893 31.1712 5.74076C30.3743 5.85098 29.5044 6.64381 29.9444 8.20627C31.3253 13.1083 31.0556 19.012 30.9522 19.857C30.6703 19.2789 30.5831 18.8206 30.1918 18.1863C28.6182 15.6396 27.87 15.452 26.9514 15.4133C26.0389 15.3746 25.0534 15.9141 25.1183 16.941C25.1852 17.9678 25.7307 18.1379 26.5053 19.5689C27.1096 20.6827 27.2819 22.1427 28.4986 24.7958C29.5064 26.9925 32.1405 29.402 36.9382 29.1158C40.8255 28.992 46.631 27.6887 45.6212 19.13C45.3697 17.6429 45.5583 16.3976 45.6901 15.1213C45.8949 13.1412 46.195 9.85962 44.733 8.99717L44.731 8.9991Z" fill="#FFE165"/>
-    <path d="M20.458 15.4707C19.5395 15.5268 18.7973 15.7259 17.2724 18.2998C16.8932 18.9398 16.8161 19.4 16.5444 19.9821C16.4248 19.139 16.0415 13.2411 17.3272 8.31587C17.7368 6.74761 16.8526 5.97024 16.0537 5.87356C15.1736 5.7672 13.959 5.83101 13.1195 8.99654H13.1094L13.1215 8.90566C13.1925 7.45149 13.4642 5.62411 13.4642 5.62411C13.7096 3.87021 12.8701 3.63236 12.2557 3.5376C10.3455 3.46025 9.04367 5.20255 8.79222 10.8375H8.78817C8.70097 8.79737 8.95039 7.17303 8.95039 7.17303C9.17547 5.60477 8.66853 5.00918 8.119 4.81774C6.86786 4.38071 6.25749 5.36498 5.95941 5.96251C5.11585 7.64873 4.89077 9.71783 4.93133 11.5878L4.91916 11.5742C5.0023 9.78164 4.0026 8.48023 2.55882 9.36589C1.11504 10.2535 1.47802 13.5292 1.72135 15.5055C1.87952 16.7798 2.09041 18.0213 1.86735 19.5122C1.02379 28.0864 6.85366 29.2872 10.7429 29.3433C15.5447 29.5464 18.1322 27.0886 19.0974 24.8745C20.2613 22.202 20.4074 20.7382 20.9893 19.6147C21.7355 18.1702 22.279 17.9904 22.3256 16.9635C22.3723 15.9367 21.3766 15.4146 20.4641 15.4688L20.458 15.4707Z" fill="#FFE165"/>
-    <path d="M22.3819 15.4845C21.8952 15.0301 21.1632 14.7884 20.419 14.8309C19.2266 14.9025 18.3811 15.3182 17.0813 17.3487C17.0468 15.0262 17.1826 11.5397 17.9816 8.47281C18.2817 7.3203 17.9796 6.56808 17.6713 6.14072C17.3124 5.64182 16.7548 5.31308 16.1383 5.2396C15.5766 5.17192 14.8426 5.16805 14.1268 5.72884C14.1268 5.7211 14.1288 5.71143 14.1288 5.71143C14.36 4.06389 13.7638 3.12023 12.3586 2.90751L12.2815 2.89978C11.4156 2.86304 10.6735 3.13376 10.0753 3.70228C9.75488 4.00588 9.47707 4.39843 9.23577 4.88379C8.96607 4.50672 8.61932 4.31527 8.34557 4.21859C6.67265 3.63267 5.74799 4.88766 5.34649 5.68823C4.8801 6.62029 4.59012 7.66451 4.4279 8.73C4.39343 8.70873 4.36098 8.68746 4.32651 8.66812C3.95746 8.46508 3.18893 8.21756 2.19126 8.83055C0.500091 9.8709 0.715036 12.8605 1.05165 15.5832C1.0699 15.7282 1.08815 15.8713 1.1064 16.0163C1.25037 17.1321 1.38623 18.186 1.19968 19.4255L1.19562 19.4564C0.85698 22.8966 1.53629 25.5438 3.21529 27.3287C4.8294 29.0458 7.35804 29.9392 10.71 29.9876C10.9553 29.9972 11.1946 30.0011 11.4278 29.9992C17.1543 29.9489 19.2084 26.2845 19.7133 25.1242C20.3663 23.6236 20.7049 22.504 20.9746 21.6029C21.1835 20.9067 21.3497 20.3576 21.585 19.9012C21.8526 19.383 22.0878 19.0465 22.2947 18.7487C22.6475 18.2421 22.9517 17.805 22.9882 16.9929C23.0145 16.405 22.8036 15.8829 22.3758 15.4845H22.3819ZM11.0263 4.61114C11.3487 4.30561 11.7198 4.17024 12.1902 4.17991C12.5978 4.24373 12.9669 4.33848 12.7986 5.5374C12.7864 5.61281 12.5228 7.41312 12.4518 8.87889C12.4518 8.88856 12.4518 8.89823 12.4518 8.9079C12.0807 10.3389 11.7705 12.4002 11.6042 15.413C10.8844 15.4555 10.1665 15.529 9.46896 15.6257C9.24388 9.51316 9.76502 5.80619 11.0243 4.61114H11.0263ZM6.56315 6.24128C7.06807 5.23573 7.49188 5.28601 7.88527 5.42331C8.43074 5.61475 8.34557 6.65316 8.28271 7.08439C8.27257 7.154 8.02924 8.77254 8.11441 10.832C8.05155 12.2765 8.05966 13.9414 8.13468 15.8462C7.46754 15.9718 6.83488 16.1169 6.25696 16.2735C5.98321 15.3956 4.77262 9.81869 6.56315 6.24321V6.24128ZM21.1794 18.039C20.9604 18.3523 20.6887 18.7429 20.3825 19.3346C20.0925 19.8935 19.9141 20.4929 19.6849 21.249C19.4233 22.1173 19.0969 23.1982 18.4743 24.6311C18.0323 25.6444 16.1748 28.9356 10.7505 28.7036C7.7271 28.661 5.58982 27.9301 4.21701 26.4701C2.80162 24.9657 2.23587 22.649 2.53395 19.5879C2.74079 18.1879 2.5887 17.0025 2.44068 15.8578C2.42243 15.7147 2.40418 15.5735 2.38593 15.4304C2.2237 14.1097 1.78976 10.5999 2.91923 9.90571C3.2234 9.71814 3.47282 9.6756 3.65735 9.77615C3.97165 9.94825 4.28798 10.5748 4.24337 11.5455C4.24135 11.5977 4.24743 11.648 4.25757 11.6983C4.31435 13.9608 4.73815 15.9293 4.97946 16.668C4.58404 16.8092 4.23526 16.9561 3.94326 17.1031C3.61476 17.2694 3.49107 17.6561 3.66546 17.9694C3.78712 18.1879 4.02235 18.3117 4.26568 18.3097C4.3691 18.3097 4.47454 18.2846 4.5739 18.2343C6.21438 17.4047 10.1057 16.5616 13.5347 16.6525C13.9078 16.6583 14.214 16.3837 14.2241 16.0299C14.2342 15.676 13.9422 15.3821 13.5712 15.3724C13.3664 15.3666 13.1595 15.3666 12.9527 15.3666C13.2954 9.29078 14.2383 7.3087 14.9724 6.72278C15.2765 6.48106 15.5665 6.46172 15.968 6.51007C16.0795 6.5236 16.3594 6.58548 16.5601 6.86394C16.7771 7.16754 16.8176 7.61616 16.6757 8.16148C15.4347 12.9204 15.7145 18.5166 15.8565 19.8741C15.8321 19.9205 15.8098 19.9669 15.7835 20.0153C15.4935 20.5355 14.9541 21.0769 14.3113 21.0402C13.9443 21.0228 13.6219 21.2896 13.5996 21.6416C13.5772 21.9954 13.8591 22.299 14.2302 22.3203C15.3171 22.3822 16.3411 21.746 16.9697 20.6186C17.0366 20.4987 17.0934 20.3846 17.1441 20.2744C17.1482 20.2667 17.1522 20.257 17.1563 20.2493C17.2739 19.9979 17.3591 19.7678 17.4341 19.5609C17.5517 19.2399 17.6531 18.9614 17.8559 18.6172C19.2956 16.1846 19.8796 16.1497 20.4981 16.113C20.861 16.0917 21.222 16.202 21.4349 16.4031C21.587 16.5442 21.6539 16.7202 21.6438 16.9406C21.6235 17.3951 21.4917 17.5846 21.1733 18.0409L21.1794 18.039Z" fill="#0D0F11"/>
-    <path d="M46.2793 19.0284C46.0704 17.7928 46.186 16.7369 46.3077 15.6193C46.3239 15.4742 46.3401 15.3311 46.3543 15.1861C46.6382 12.4595 46.7964 9.46417 45.0829 8.45476C44.073 7.85916 43.3086 8.12022 42.9436 8.32906C42.9091 8.3484 42.8766 8.3716 42.8422 8.39288C42.6576 7.33125 42.3494 6.29284 41.8648 5.36851C41.4491 4.57568 40.5021 3.33615 38.8393 3.95108C38.5676 4.05164 38.2269 4.24888 37.9633 4.63176C37.7119 4.15026 37.426 3.76351 37.0995 3.46571C36.4912 2.9088 35.7429 2.64968 34.8791 2.70189L34.802 2.70962C33.4008 2.94747 32.8229 3.9008 33.0865 5.54835C33.0865 5.54835 33.0865 5.55608 33.0885 5.56188C32.3626 5.0127 31.6285 5.03011 31.0689 5.10746C30.4545 5.19254 29.9029 5.53094 29.5541 6.03565C29.256 6.46881 28.9661 7.2249 29.2885 8.3716C30.1483 11.425 30.351 14.9096 30.3612 17.232C29.0228 15.2248 28.1692 14.8245 26.9768 14.7742C26.2346 14.7433 25.5026 15.0005 25.0261 15.4626C24.6063 15.8687 24.4056 16.3947 24.4441 16.9806C24.4968 17.7908 24.8091 18.224 25.1721 18.7229C25.385 19.0168 25.6263 19.3494 25.9041 19.8619C26.1495 20.3144 26.3259 20.8597 26.549 21.552C26.8369 22.4473 27.1958 23.5611 27.8792 25.0501C28.4064 26.2007 30.5315 29.8303 36.2417 29.7781C36.4729 29.7761 36.7122 29.7684 36.9555 29.7529C40.3257 29.6466 42.8361 28.7068 44.4178 26.9625C46.0603 25.1487 46.6889 22.4898 46.2853 19.0555L46.2813 19.0246L46.2793 19.0284ZM38.961 6.82075C38.89 6.38372 38.7826 5.34724 39.326 5.14806C39.7153 5.00303 40.1412 4.94696 40.6643 5.94283C42.5238 9.48737 41.4227 15.0855 41.1652 15.9673C40.5832 15.8204 39.9485 15.6869 39.2794 15.5728C39.3159 13.6681 39.2915 12.0012 39.2023 10.5587C39.2469 8.49923 38.9732 6.88456 38.961 6.82075ZM34.9967 3.98009C35.4692 3.96075 35.8423 4.09031 36.1687 4.39197C37.4503 5.56575 38.0444 9.26112 37.937 15.3775C37.2374 15.2924 36.5196 15.2325 35.7977 15.2016C35.5746 12.1907 35.2238 10.1371 34.8243 8.71194C34.8243 8.70227 34.8243 8.69261 34.8243 8.68294C34.725 7.21716 34.4249 5.42266 34.4127 5.35304C34.22 4.15219 34.5871 4.05164 34.9947 3.98009H34.9967ZM44.9511 19.2179C45.308 22.2732 44.7868 24.5995 43.4018 26.1291C42.0574 27.6123 39.9343 28.3819 36.8927 28.4786C31.4988 28.8035 29.5724 25.5471 29.1121 24.5415C28.4591 23.1183 28.1124 22.0451 27.8346 21.1807C27.5912 20.4265 27.4006 19.8329 27.1005 19.2779C26.7842 18.692 26.5043 18.3071 26.2793 17.9977C25.9528 17.5472 25.8169 17.3596 25.7865 16.9052C25.7723 16.6847 25.8372 16.5068 25.9852 16.3637C26.1961 16.1588 26.553 16.0408 26.918 16.0582C27.5365 16.0853 28.1205 16.1085 29.6089 18.516C29.8198 18.8563 29.9252 19.1328 30.0489 19.4519C30.13 19.6588 30.2192 19.8889 30.3429 20.1384C30.347 20.1461 30.349 20.1539 30.3531 20.1597C30.4078 20.2699 30.4666 20.382 30.5356 20.5019C31.1865 21.6177 32.2227 22.2365 33.3075 22.1553C33.6766 22.1282 33.9544 21.8188 33.926 21.4669C33.8976 21.1149 33.5752 20.8539 33.2041 20.8771C32.5613 20.9235 32.0118 20.3917 31.7117 19.8773C31.6833 19.829 31.661 19.7845 31.6367 19.7381C31.7522 18.3806 31.9246 12.7786 30.5903 8.04287C30.4362 7.49949 30.4687 7.05086 30.6795 6.7434C30.8762 6.46107 31.154 6.39339 31.2656 6.37792C31.665 6.32184 31.957 6.33731 32.2653 6.57323C33.0115 7.14755 33.9929 9.11223 34.4512 15.1803C34.2444 15.1822 34.0376 15.188 33.8348 15.1977C33.4637 15.2132 33.1778 15.5129 33.194 15.8668C33.2102 16.2206 33.5184 16.4875 33.8956 16.4778C37.3205 16.327 41.2301 17.1005 42.8848 17.903C42.9841 17.9513 43.0896 17.9726 43.195 17.9726C43.4383 17.9707 43.6715 17.843 43.7891 17.6207C43.9575 17.3055 43.8257 16.9187 43.4931 16.7582C43.1991 16.6151 42.8462 16.4759 42.4488 16.3405C42.6759 15.598 43.0632 13.6217 43.0754 11.3592C43.0855 11.309 43.0896 11.2587 43.0855 11.2065C43.0206 10.2377 43.3268 9.60533 43.6371 9.42742C43.8196 9.323 44.069 9.36168 44.3772 9.54345C45.5209 10.2183 45.1559 13.7339 45.018 15.0585C45.0038 15.2016 44.9876 15.3427 44.9713 15.4858C44.8456 16.6345 44.7158 17.8198 44.9511 19.2179Z" fill="#0D0F11"/>
-    <path d="M26.1508 6.85319C26.0434 6.85319 25.9339 6.83386 25.8304 6.78745C25.4512 6.62114 25.285 6.19379 25.4594 5.83218C26.0231 4.6584 26.8484 3.57551 27.844 2.70146C28.1502 2.43267 28.6288 2.45007 28.9106 2.744C29.1925 3.036 29.1742 3.49236 28.866 3.76115C28.0164 4.50757 27.3127 5.4319 26.8301 6.43357C26.7044 6.69463 26.4347 6.85126 26.1508 6.85319Z" fill="#F9F7F2"/>
-    <path d="M23.608 6.43744C23.2166 6.44131 22.8821 6.15511 22.8496 5.7761C22.7056 4.08021 22.6996 2.36112 22.8354 0.665235C22.8679 0.268818 23.2308 -0.0270433 23.6445 0.0019628C24.0602 0.0329026 24.3704 0.377108 24.34 0.773524C24.2103 2.394 24.2163 4.03767 24.3542 5.65814C24.3887 6.05456 24.0784 6.40263 23.6628 6.43357C23.6445 6.43357 23.6263 6.4355 23.608 6.4355V6.43744Z" fill="#F9F7F2"/>
-    <path d="M21.0084 6.88414C20.6697 6.888 20.3575 6.66949 20.2703 6.34269C19.9499 5.14377 19.3436 4.0048 18.5183 3.05147C18.2526 2.74401 18.2993 2.29151 18.6197 2.03819C18.9421 1.78487 19.4166 1.82935 19.6822 2.13488C20.6474 3.25258 21.3572 4.58492 21.7303 5.98688C21.8337 6.3717 21.5883 6.76425 21.1848 6.86287C21.124 6.87834 21.0652 6.88414 21.0043 6.88607L21.0084 6.88414Z" fill="#F9F7F2"/>
-  </g>
-  <defs>
-    <clipPath id="clip0_10905_18559">
-      <rect width="45.7143" height="30" fill="white" transform="translate(0.818359)"/>
-    </clipPath>
-  </defs>
+<svg width="47" height="30" viewBox="0 0 148 100" fill="none" xmlns="http://www.w3.org/2000/svg">
+<path d="M71.754 16.863V2.97414C71.754 1.82355 72.6871 0.890503 73.8377 0.890503C74.9883 0.890503 75.9213 1.82355 75.9213 2.97414V16.863C75.9213 18.0136 74.9883 18.9466 73.8377 18.9466C72.6871 18.9466 71.754 18.0136 71.754 16.863Z" fill="black"/>
+<path d="M82.5273 18.9329L89.4717 6.90477C90.047 5.90832 91.3215 5.5668 92.318 6.1421C93.3144 6.7174 93.6559 7.99197 93.0806 8.98841L86.1362 21.0165C85.5609 22.0129 84.2864 22.3545 83.2899 21.7792C82.2935 21.2039 81.952 19.9293 82.5273 18.9329Z" fill="black"/>
+<path d="M65.1481 18.9329L58.2037 6.90477C57.6284 5.90832 56.3538 5.5668 55.3574 6.1421C54.3609 6.7174 54.0194 7.99197 54.5947 8.98841L61.5391 21.0165C62.1144 22.0129 63.389 22.3545 64.3854 21.7792C65.3819 21.2039 65.7234 19.9293 65.1481 18.9329Z" fill="black"/>
+<path d="M98.5045 92.4969C95.1427 89.7602 90.8793 83.6628 89.0929 77.664C86.8598 70.1655 82.0309 62.777 80.9578 60.9138C80.0056 59.2602 83.0313 53.2605 89.0929 58.0459C93.9421 61.8742 97.3878 68.466 98.5045 71.2833C98.9566 72.4239 103.006 79.2475 101.827 56.6051C101.455 49.445 100.49 44.3126 100.036 33.9388C99.6807 25.7997 101.824 23.057 105.11 23.057C108.396 23.057 111.106 24.1244 111.106 32.4283C111.106 22.8682 113.154 20.7141 117.646 20.7147C121.39 20.7153 123.069 23.9077 123.069 27.0952V32.5084C123.069 27.0952 126.669 26.5896 128.519 26.5896C130.37 26.5896 134.076 28.1957 134.076 32.5084V43.317C134.076 38.116 137.324 36.601 139.773 36.765C142.963 36.9786 144.405 39.2382 143.965 44.3126C143.651 47.9349 142.689 58.5273 142.689 62.0291C142.689 65.1712 143.965 71.6801 142.689 77.664C141.953 81.1168 140.137 90.2628 130.885 94.4102C121.634 98.5575 105.364 98.0807 98.5045 92.4969Z" fill="#FFFF8B"/>
+<path d="M140.606 62.0292C140.606 58.409 141.583 47.6748 141.89 44.1323C142.097 41.7374 141.809 40.4247 141.424 39.7542C141.141 39.2626 140.699 38.915 139.634 38.8436C138.865 38.7921 138.027 39.0114 137.401 39.5761C136.814 40.1052 136.159 41.1682 136.159 43.3176L136.155 43.4388L135.198 59.758C135.164 60.3451 134.883 60.8911 134.424 61.2599C133.966 61.6284 133.374 61.7859 132.793 61.6941L122.764 60.1068L111.948 58.6703C110.949 58.5376 110.188 57.7084 110.142 56.7016L109.545 43.8055L109.54 43.6959C109.274 38.5609 109.022 33.8767 109.022 32.4282C109.022 28.3859 108.338 26.6806 107.74 25.9634C107.263 25.3915 106.577 25.1402 105.11 25.1402C104.583 25.1402 104.212 25.2481 103.933 25.4111C103.659 25.5714 103.346 25.8587 103.049 26.4208C102.41 27.6257 101.945 29.891 102.118 33.8479C102.342 38.9804 102.692 42.8146 103.035 46.2718C103.377 49.7231 103.718 52.8561 103.908 56.4971C104.204 62.1966 104.178 66.1256 103.945 68.7924C103.828 70.124 103.656 71.1996 103.423 72.0501C103.202 72.8558 102.871 73.6757 102.296 74.2887C101.6 75.0303 100.608 75.3844 99.577 75.136C98.7592 74.9389 98.1847 74.4215 97.8706 74.0916C97.2141 73.4017 96.7501 72.5106 96.568 72.0512C95.5097 69.3812 92.2352 63.1808 87.8023 59.6811C86.5089 58.6599 85.5666 58.3652 84.9736 58.3204C84.4148 58.2783 84.0094 58.4436 83.6909 58.6967C83.34 58.9756 83.0781 59.3811 82.9479 59.7643C82.9019 59.8999 82.8823 59.9968 82.8741 60.0584C84.0759 62.0865 88.8421 69.5222 91.0896 77.069C92.7648 82.6941 96.8038 88.4259 99.8194 90.8809C102.74 93.258 107.988 94.7313 113.9 95.0218C119.756 95.3095 125.788 94.4121 130.033 92.5092C138.233 88.8334 139.903 80.7382 140.651 77.2292C141.232 74.5057 141.243 71.5987 141.087 68.9009C141.01 67.5551 140.894 66.2969 140.793 65.1373C140.695 64.0105 140.606 62.9215 140.606 62.0292ZM120.986 27.0953C120.986 25.8314 120.648 24.7049 120.089 23.9514C119.583 23.27 118.84 22.7987 117.646 22.7984C116.668 22.7982 116.011 22.9187 115.546 23.1167C115.13 23.2943 114.781 23.5699 114.463 24.0831C113.73 25.2671 113.192 27.6455 113.189 32.384L113.707 43.6021C113.901 47.3443 114.103 51.3994 114.236 54.7707L120.986 55.6666V27.0953ZM125.153 56.2652L131.172 57.218L131.992 43.267V32.5083C131.992 31.031 131.39 30.1275 130.678 29.5489C129.884 28.9039 128.957 28.6731 128.519 28.6731C127.722 28.6731 126.899 28.797 126.306 29.2179C125.849 29.5421 125.153 30.3087 125.153 32.5083V56.2652ZM136.159 35.4278C137.406 34.8069 138.74 34.6083 139.912 34.6868C142.037 34.8292 143.91 35.718 145.037 37.6779C146.06 39.4592 146.273 41.8136 146.041 44.4927C145.72 48.1949 144.772 58.6457 144.772 62.0292C144.772 62.708 144.843 63.6116 144.944 64.7758C145.042 65.907 145.165 67.2389 145.247 68.6606C145.411 71.4987 145.422 74.8383 144.727 78.0987C144.002 81.4953 142.041 91.6918 131.738 96.3108C126.731 98.5551 120.002 99.4936 113.696 99.1838C107.445 98.8767 101.128 97.3189 97.1887 94.1122C93.4809 91.0936 88.9938 84.6307 87.0962 78.2589C84.9529 71.0619 80.3109 63.9646 79.1527 61.9533C78.4706 60.7689 78.684 59.3628 79.0019 58.4258C79.3607 57.3688 80.0554 56.2631 81.0993 55.4337C82.1758 54.5784 83.6043 54.0377 85.2876 54.1647C86.9369 54.2893 88.6462 55.0393 90.3834 56.4107C94.8541 59.9401 98.1342 65.5082 99.7424 68.9231C99.759 68.7664 99.779 68.6024 99.7941 68.4298C100.003 66.0435 100.039 62.3344 99.7467 56.7132C99.5635 53.1942 99.2356 50.1809 98.8888 46.6828C98.5425 43.1904 98.184 39.2713 97.955 34.0302C97.7722 29.8481 98.2012 26.6722 99.3672 24.471C99.9716 23.3302 100.79 22.4223 101.83 21.814C102.866 21.2087 103.995 20.974 105.11 20.974C106.759 20.974 108.813 21.2062 110.448 22.7678C110.593 22.4576 110.75 22.1652 110.921 21.8899C111.676 20.6698 112.681 19.8084 113.912 19.2835C115.095 18.7791 116.378 18.6309 117.646 18.6311C120.195 18.6315 122.165 19.7565 123.435 21.4683C124.257 22.576 124.75 23.8776 124.985 25.1982C126.338 24.5876 127.691 24.5068 128.519 24.5068C129.933 24.5068 131.784 25.0791 133.305 26.3154C134.908 27.6179 136.159 29.6733 136.159 32.5083V35.4278Z" fill="black"/>
+<path d="M49.258 92.4969C52.6198 89.7602 56.8832 83.6628 58.6696 77.664C60.9027 70.1655 65.7316 62.777 66.8047 60.9138C67.757 59.2602 64.7312 53.2605 58.6696 58.0459C53.8204 61.8742 50.3747 68.466 49.258 71.2833C48.8059 72.4239 44.7566 79.2475 45.935 56.6051C46.3077 49.445 47.2728 44.3126 47.7261 33.9388C48.0818 25.7997 45.9381 23.057 42.6523 23.057C39.3666 23.057 36.6568 24.1244 36.6568 32.4283C36.6568 22.8682 34.6087 20.7141 30.1168 20.7147C26.3728 20.7153 24.6934 23.9077 24.6934 27.0952V32.5084C24.6934 27.0952 21.0938 26.5896 19.243 26.5896C17.3923 26.5896 13.687 28.1957 13.687 32.5084V43.317C13.687 38.116 10.4388 36.601 7.98968 36.765C4.79943 36.9786 3.3574 39.2382 3.79721 44.3126C4.11115 47.9349 5.07331 58.5273 5.07331 62.0291C5.07331 65.1712 3.79721 71.6801 5.07331 77.664C5.80964 81.1168 7.62551 90.2628 16.8772 94.4102C26.129 98.5575 42.3987 98.0807 49.258 92.4969Z" fill="#FFFF8B"/>
+<path d="M7.15667 62.0292C7.15667 58.409 6.18001 47.6748 5.87297 44.1323C5.66546 41.7374 5.95363 40.4247 6.33881 39.7542C6.62122 39.2626 7.06342 38.915 8.1284 38.8436C8.89765 38.7921 9.7355 39.0114 10.3617 39.5761C10.9484 40.1052 11.6032 41.1682 11.6032 43.3176L11.6075 43.4388L12.5644 59.758C12.5989 60.3451 12.8798 60.8911 13.338 61.2599C13.7961 61.6284 14.3888 61.7859 14.9695 61.6941L24.9988 60.1068L35.8144 58.6703C36.8136 58.5376 37.5741 57.7084 37.6208 56.7016L38.2174 43.8055L38.2226 43.6959C38.4887 38.5609 38.7401 33.8767 38.7401 32.4282C38.7401 28.3859 39.4246 26.6806 40.0228 25.9634C40.4997 25.3915 41.1851 25.1402 42.6523 25.1402C43.1795 25.1402 43.5506 25.2481 43.8296 25.4111C44.1038 25.5714 44.416 25.8587 44.7139 26.4208C45.3521 27.6257 45.8174 29.891 45.6445 33.8479C45.4202 38.9804 45.0703 42.8146 44.7276 46.2718C44.3854 49.7231 44.0444 52.8561 43.8549 56.4971C43.5583 62.1966 43.5848 66.1256 43.818 68.7924C43.9345 70.124 44.107 71.1996 44.3397 72.0501C44.5602 72.8558 44.891 73.6757 45.4664 74.2887C46.1626 75.0303 47.1547 75.3844 48.1855 75.136C49.0033 74.9389 49.5779 74.4215 49.8919 74.0916C50.5484 73.4017 51.0124 72.5106 51.1945 72.0512C52.2528 69.3812 55.5273 63.1808 59.9602 59.6811C61.2536 58.6599 62.1959 58.3652 62.7889 58.3204C63.3477 58.2783 63.7531 58.4436 64.0716 58.6967C64.4225 58.9756 64.6844 59.3811 64.8146 59.7643C64.8606 59.8999 64.8802 59.9968 64.8884 60.0584C63.6866 62.0865 58.9205 69.5222 56.6729 77.069C54.9978 82.6941 50.9587 88.4259 47.9431 90.8809C45.0229 93.258 39.7748 94.7313 33.8625 95.0218C28.0069 95.3095 21.9748 94.4121 17.7298 92.5092C9.52994 88.8334 7.85968 80.7382 7.11135 77.2292C6.5306 74.5057 6.51956 71.5987 6.67502 68.9009C6.75257 67.5551 6.86815 66.2969 6.96907 65.1373C7.06713 64.0105 7.15666 62.9215 7.15667 62.0292ZM26.7768 27.0953C26.7768 25.8314 27.1148 24.7049 27.6737 23.9514C28.1793 23.27 28.9221 22.7987 30.1168 22.7984C31.0942 22.7982 31.7519 22.9187 32.2162 23.1167C32.6327 23.2943 32.9818 23.5699 33.2997 24.0831C34.0329 25.2671 34.5706 27.6455 34.5739 32.384L34.0554 43.6021C33.8615 47.3443 33.6592 51.3994 33.5263 54.7707L26.7768 55.6666V27.0953ZM22.6096 56.2652L16.5905 57.218L15.7705 43.267V32.5083C15.7705 31.031 16.3726 30.1275 17.0848 29.5489C17.8786 28.9039 18.8059 28.6731 19.2433 28.6731C20.0405 28.6731 20.8635 28.797 21.4565 29.2179C21.9131 29.5421 22.6095 30.3087 22.6096 32.5083V56.2652ZM11.6032 35.4278C10.3568 34.8069 9.02271 34.6083 7.85016 34.6868C5.72547 34.8292 3.85203 35.718 2.7259 37.6779C1.70253 39.4592 1.4893 41.8136 1.7215 44.4927C2.04237 48.1949 2.99044 58.6457 2.99044 62.0292C2.99043 62.708 2.91997 63.6116 2.81865 64.7758C2.7202 65.907 2.59705 67.2389 2.51511 68.6606C2.35156 71.4987 2.34047 74.8383 3.03576 78.0987C3.76011 81.4953 5.7216 91.6918 16.0245 96.3108C21.0312 98.5551 27.7601 99.4936 34.0669 99.1838C40.3173 98.8767 46.6346 97.3189 50.5738 94.1122C54.2816 91.0936 58.7687 84.6307 60.6663 78.2589C62.8096 71.0619 67.4516 63.9646 68.6099 61.9533C69.292 60.7689 69.0785 59.3628 68.7606 58.4258C68.4018 57.3688 67.7071 56.2631 66.6632 55.4337C65.5867 54.5784 64.1582 54.0377 62.4749 54.1647C60.8256 54.2893 59.1163 55.0393 57.3791 56.4107C52.9084 59.9401 49.6283 65.5082 48.0201 68.9231C48.0035 68.7664 47.9835 68.6024 47.9684 68.4298C47.7597 66.0435 47.7233 62.3344 48.0159 56.7132C48.199 53.1942 48.5269 50.1809 48.8738 46.6828C49.22 43.1904 49.5785 39.2713 49.8076 34.0302C49.9903 29.8481 49.5613 26.6722 48.3953 24.471C47.7909 23.3302 46.9729 22.4223 45.9322 21.814C44.8964 21.2087 43.7676 20.974 42.6523 20.974C41.0038 20.974 38.9498 21.2062 37.3141 22.7678C37.1699 22.4576 37.0125 22.1652 36.842 21.8899C36.0864 20.6698 35.0817 19.8084 33.8509 19.2835C32.668 18.7791 31.3849 18.6309 30.1168 18.6311C27.5676 18.6315 25.5976 19.7565 24.3275 21.4683C23.5057 22.576 23.0121 23.8776 22.7771 25.1982C21.4248 24.5876 20.0718 24.5068 19.2433 24.5068C17.8299 24.5068 15.9789 25.0791 14.4573 26.3154C12.8543 27.6179 11.6033 29.6733 11.6032 32.5083V35.4278Z" fill="black"/>
 </svg>

frontend/src/components/features/chat/change-agent-button.tsx

@@ -1,4 +1,4 @@
-import React, { useMemo, useEffect } from "react";
+import React, { useMemo, useEffect, useState } from "react";
 import { useTranslation } from "react-i18next";
 import { Typography } from "#/ui/typography";
 import { I18nKey } from "#/i18n/declaration";
@@ -11,31 +11,87 @@ import { cn } from "#/utils/utils";
 import { USE_PLANNING_AGENT } from "#/utils/feature-flags";
 import { useAgentState } from "#/hooks/use-agent-state";
 import { AgentState } from "#/types/agent-state";
+import { useActiveConversation } from "#/hooks/query/use-active-conversation";
+import { useUnifiedWebSocketStatus } from "#/hooks/use-unified-websocket-status";
+import { useSubConversationTaskPolling } from "#/hooks/query/use-sub-conversation-task-polling";
+import { useHandlePlanClick } from "#/hooks/use-handle-plan-click";
 
 export function ChangeAgentButton() {
-  const { t } = useTranslation();
-  const [contextMenuOpen, setContextMenuOpen] = React.useState(false);
+  const [contextMenuOpen, setContextMenuOpen] = useState<boolean>(false);
 
-  const conversationMode = useConversationStore(
-    (state) => state.conversationMode,
-  );
+  const { conversationMode, setConversationMode, subConversationTaskId } =
+    useConversationStore();
 
-  const setConversationMode = useConversationStore(
-    (state) => state.setConversationMode,
-  );
+  const webSocketStatus = useUnifiedWebSocketStatus();
+
+  const isWebSocketConnected = webSocketStatus === "CONNECTED";
 
   const shouldUsePlanningAgent = USE_PLANNING_AGENT();
 
   const { curAgentState } = useAgentState();
 
+  const { t } = useTranslation();
+
   const isAgentRunning = curAgentState === AgentState.RUNNING;
 
+  const { data: conversation } = useActiveConversation();
+
+  // Poll sub-conversation task and invalidate parent conversation when ready
+  useSubConversationTaskPolling(
+    subConversationTaskId,
+    conversation?.conversation_id || null,
+  );
+
+  // Get handlePlanClick and isCreatingConversation from custom hook
+  const { handlePlanClick, isCreatingConversation } = useHandlePlanClick();
+
   // Close context menu when agent starts running
   useEffect(() => {
-    if (isAgentRunning && contextMenuOpen) {
+    if ((isAgentRunning || !isWebSocketConnected) && contextMenuOpen) {
       setContextMenuOpen(false);
     }
-  }, [isAgentRunning, contextMenuOpen]);
+  }, [isAgentRunning, contextMenuOpen, isWebSocketConnected]);
+
+  const isButtonDisabled =
+    isAgentRunning ||
+    isCreatingConversation ||
+    !isWebSocketConnected ||
+    !shouldUsePlanningAgent;
+
+  // Handle Shift + Tab keyboard shortcut to cycle through modes
+  useEffect(() => {
+    if (isButtonDisabled) {
+      return undefined;
+    }
+
+    const handleKeyDown = (event: KeyboardEvent) => {
+      // Check for Shift + Tab combination
+      if (event.shiftKey && event.key === "Tab") {
+        // Prevent default tab navigation behavior
+        event.preventDefault();
+        event.stopPropagation();
+
+        // Cycle between modes: code -> plan -> code
+        const nextMode = conversationMode === "code" ? "plan" : "code";
+        if (nextMode === "plan") {
+          handlePlanClick(event);
+        } else {
+          setConversationMode(nextMode);
+        }
+      }
+    };
+
+    document.addEventListener("keydown", handleKeyDown);
+
+    return () => {
+      document.removeEventListener("keydown", handleKeyDown);
+    };
+  }, [
+    isButtonDisabled,
+    conversationMode,
+    setConversationMode,
+    handlePlanClick,
+  ]);
 
   const handleButtonClick = (event: React.MouseEvent<HTMLButtonElement>) => {
     event.preventDefault();
@@ -49,12 +105,6 @@ export function ChangeAgentButton() {
     setConversationMode("code");
   };
 
-  const handlePlanClick = (event: React.MouseEvent<HTMLButtonElement>) => {
-    event.preventDefault();
-    event.stopPropagation();
-    setConversationMode("plan");
-  };
-
   const isExecutionAgent = conversationMode === "code";
 
   const buttonLabel = useMemo(() => {
@@ -80,11 +130,11 @@ export function ChangeAgentButton() {
       <button
         type="button"
         onClick={handleButtonClick}
-        disabled={isAgentRunning}
+        disabled={isButtonDisabled}
         className={cn(
           "flex items-center border border-[#4B505F] rounded-[100px] transition-opacity",
           !isExecutionAgent && "border-[#597FF4] bg-[#4A67BD]",
-          isAgentRunning
+          isButtonDisabled
             ? "opacity-50 cursor-not-allowed"
             : "cursor-pointer hover:opacity-80",
         )}

frontend/src/components/features/chat/change-agent-context-menu.tsx

@@ -5,19 +5,14 @@ import CodeTagIcon from "#/icons/code-tag.svg?react";
 import LessonPlanIcon from "#/icons/lesson-plan.svg?react";
 import { ContextMenu } from "#/ui/context-menu";
 import { ContextMenuListItem } from "../context-menu/context-menu-list-item";
-import { ContextMenuIconText } from "../context-menu/context-menu-icon-text";
+import { ContextMenuIconTextWithDescription } from "../context-menu/context-menu-icon-text-with-description";
 import { useClickOutsideElement } from "#/hooks/use-click-outside-element";
 import { cn } from "#/utils/utils";
-import { CONTEXT_MENU_ICON_TEXT_CLASSNAME } from "#/utils/constants";
 
 const contextMenuListItemClassName = cn(
   "cursor-pointer p-0 h-auto hover:bg-transparent",
-  CONTEXT_MENU_ICON_TEXT_CLASSNAME,
 );
 
-const contextMenuIconTextClassName =
-  "gap-2 p-2 hover:bg-[#5C5D62] rounded h-[30px]";
-
 interface ChangeAgentContextMenuProps {
   onClose: () => void;
   onCodeClick?: (event: React.MouseEvent<HTMLButtonElement>) => void;
@@ -52,17 +47,17 @@ export function ChangeAgentContextMenu({
       testId="change-agent-context-menu"
       position="top"
       alignment="left"
-      className="min-h-fit min-w-[195px] mb-2"
+      className="min-h-fit mb-2 min-w-[195px] max-w-[195px] gap-0"
     >
       <ContextMenuListItem
         testId="code-option"
         onClick={handleCodeClick}
         className={contextMenuListItemClassName}
       >
-        <ContextMenuIconText
+        <ContextMenuIconTextWithDescription
           icon={CodeTagIcon}
-          text={t(I18nKey.COMMON$CODE)}
-          className={contextMenuIconTextClassName}
+          title={t(I18nKey.COMMON$CODE)}
+          description={t(I18nKey.COMMON$CODE_AGENT_DESCRIPTION)}
         />
       </ContextMenuListItem>
       <ContextMenuListItem
@@ -70,10 +65,10 @@ export function ChangeAgentContextMenu({
         onClick={handlePlanClick}
         className={contextMenuListItemClassName}
       >
-        <ContextMenuIconText
+        <ContextMenuIconTextWithDescription
           icon={LessonPlanIcon}
-          text={t(I18nKey.COMMON$PLAN)}
-          className={contextMenuIconTextClassName}
+          title={t(I18nKey.COMMON$PLAN)}
+          description={t(I18nKey.COMMON$PLAN_AGENT_DESCRIPTION)}
         />
       </ContextMenuListItem>
     </ContextMenu>

frontend/src/components/features/chat/chat-message.tsx

@@ -1,15 +1,9 @@
 import React from "react";
-import Markdown from "react-markdown";
-import remarkGfm from "remark-gfm";
-import remarkBreaks from "remark-breaks";
-import { code } from "../markdown/code";
 import { cn } from "#/utils/utils";
-import { ul, ol } from "../markdown/list";
 import { CopyToClipboardButton } from "#/components/shared/buttons/copy-to-clipboard-button";
-import { anchor } from "../markdown/anchor";
 import { OpenHandsSourceType } from "#/types/core/base";
-import { paragraph } from "../markdown/paragraph";
 import { TooltipButton } from "#/components/shared/buttons/tooltip-button";
+import { MarkdownRenderer } from "../markdown/markdown-renderer";
 
 interface ChatMessageProps {
   type: OpenHandsSourceType;
@@ -19,6 +13,7 @@ interface ChatMessageProps {
     onClick: () => void;
     tooltip?: string;
   }>;
+  isFromPlanningAgent?: boolean;
 }
 
 export function ChatMessage({
@@ -26,6 +21,7 @@ export function ChatMessage({
   message,
   children,
   actions,
+  isFromPlanningAgent = false,
 }: React.PropsWithChildren<ChatMessageProps>) {
   const [isHovering, setIsHovering] = React.useState(false);
   const [isCopy, setIsCopy] = React.useState(false);
@@ -59,6 +55,7 @@ export function ChatMessage({
         "flex flex-col gap-2",
         type === "user" && " p-4 bg-tertiary self-end",
         type === "agent" && "mt-6 w-full max-w-full bg-transparent",
+        isFromPlanningAgent && "border border-[#597ff4] bg-tertiary p-4",
       )}
     >
       <div
@@ -113,18 +110,7 @@ export function ChatMessage({
           wordBreak: "break-word",
         }}
       >
-        <Markdown
-          components={{
-            code,
-            ul,
-            ol,
-            a: anchor,
-            p: paragraph,
-          }}
-          remarkPlugins={[remarkGfm, remarkBreaks]}
-        >
-          {message}
-        </Markdown>
+        <MarkdownRenderer includeStandard>{message}</MarkdownRenderer>
       </div>
       {children}
     </article>

frontend/src/components/features/chat/error-message.tsx

@@ -1,13 +1,9 @@
 import React from "react";
-import Markdown from "react-markdown";
-import remarkGfm from "remark-gfm";
-import remarkBreaks from "remark-breaks";
 import { useTranslation } from "react-i18next";
-import { code } from "../markdown/code";
-import { ol, ul } from "../markdown/list";
 import ArrowDown from "#/icons/angle-down-solid.svg?react";
 import ArrowUp from "#/icons/angle-up-solid.svg?react";
 import i18n from "#/i18n";
+import { MarkdownRenderer } from "../markdown/markdown-renderer";
 
 interface ErrorMessageProps {
   errorId?: string;
@@ -40,18 +36,7 @@ export function ErrorMessage({ errorId, defaultMessage }: ErrorMessageProps) {
         </button>
       </div>
 
-      {showDetails && (
-        <Markdown
-          components={{
-            code,
-            ul,
-            ol,
-          }}
-          remarkPlugins={[remarkGfm, remarkBreaks]}
-        >
-          {defaultMessage}
-        </Markdown>
-      )}
+      {showDetails && <MarkdownRenderer>{defaultMessage}</MarkdownRenderer>}
     </div>
   );
 }

frontend/src/components/features/chat/event-message-components/task-tracking-event-message.tsx

@@ -1,11 +1,7 @@
-import React from "react";
-import { useTranslation } from "react-i18next";
 import { OpenHandsObservation } from "#/types/core/observations";
 import { isTaskTrackingObservation } from "#/types/core/guards";
-import { GenericEventMessage } from "../generic-event-message";
 import { TaskTrackingObservationContent } from "../task-tracking-observation-content";
 import { ConfirmationButtons } from "#/components/shared/buttons/confirmation-buttons";
-import { getObservationResult } from "../event-content-helpers/get-observation-result";
 
 interface TaskTrackingEventMessageProps {
   event: OpenHandsObservation;
@@ -16,34 +12,13 @@ export function TaskTrackingEventMessage({
   event,
   shouldShowConfirmationButtons,
 }: TaskTrackingEventMessageProps) {
-  const { t } = useTranslation();
-
   if (!isTaskTrackingObservation(event)) {
     return null;
   }
 
-  const { command } = event.extras;
-  let title: React.ReactNode;
-  let initiallyExpanded = false;
-
-  // Determine title and expansion state based on command
-  if (command === "plan") {
-    title = t("OBSERVATION_MESSAGE$TASK_TRACKING_PLAN");
-    initiallyExpanded = true;
-  } else {
-    // command === "view"
-    title = t("OBSERVATION_MESSAGE$TASK_TRACKING_VIEW");
-    initiallyExpanded = false;
-  }
-
   return (
     <div>
-      <GenericEventMessage
-        title={title}
-        details={<TaskTrackingObservationContent event={event} />}
-        success={getObservationResult(event)}
-        initiallyExpanded={initiallyExpanded}
-      />
+      <TaskTrackingObservationContent event={event} />
       {shouldShowConfirmationButtons && <ConfirmationButtons />}
     </div>
   );

frontend/src/components/features/chat/expandable-message.tsx

@@ -1,9 +1,6 @@
 import { useEffect, useState } from "react";
 import { Trans, useTranslation } from "react-i18next";
-import Markdown from "react-markdown";
 import { Link } from "react-router";
-import remarkGfm from "remark-gfm";
-import remarkBreaks from "remark-breaks";
 import { useConfig } from "#/hooks/query/use-config";
 import { I18nKey } from "#/i18n/declaration";
 import ArrowDown from "#/icons/angle-down-solid.svg?react";
@@ -13,9 +10,7 @@ import XCircle from "#/icons/x-circle-solid.svg?react";
 import { OpenHandsAction } from "#/types/core/actions";
 import { OpenHandsObservation } from "#/types/core/observations";
 import { cn } from "#/utils/utils";
-import { code } from "../markdown/code";
-import { ol, ul } from "../markdown/list";
-import { paragraph } from "../markdown/paragraph";
+import { MarkdownRenderer } from "../markdown/markdown-renderer";
 import { MonoComponent } from "./mono-component";
 import { PathComponent } from "./path-component";
 
@@ -192,17 +187,7 @@ export function ExpandableMessage({
         </div>
         {showDetails && (
           <div className="text-sm">
-            <Markdown
-              components={{
-                code,
-                ul,
-                ol,
-                p: paragraph,
-              }}
-              remarkPlugins={[remarkGfm, remarkBreaks]}
-            >
-              {details}
-            </Markdown>
+            <MarkdownRenderer includeStandard>{details}</MarkdownRenderer>
           </div>
         )}
       </div>

frontend/src/components/features/chat/generic-event-message.tsx

@@ -1,13 +1,9 @@
 import React from "react";
-import Markdown from "react-markdown";
-import remarkGfm from "remark-gfm";
-import remarkBreaks from "remark-breaks";
-import { code } from "../markdown/code";
-import { ol, ul } from "../markdown/list";
 import ArrowDown from "#/icons/angle-down-solid.svg?react";
 import ArrowUp from "#/icons/angle-up-solid.svg?react";
 import { SuccessIndicator } from "./success-indicator";
 import { ObservationResultStatus } from "./event-content-helpers/get-observation-result";
+import { MarkdownRenderer } from "../markdown/markdown-renderer";
 
 interface GenericEventMessageProps {
   title: React.ReactNode;
@@ -49,16 +45,7 @@ export function GenericEventMessage({
 
       {showDetails &&
         (typeof details === "string" ? (
-          <Markdown
-            components={{
-              code,
-              ul,
-              ol,
-            }}
-            remarkPlugins={[remarkGfm, remarkBreaks]}
-          >
-            {details}
-          </Markdown>
+          <MarkdownRenderer>{details}</MarkdownRenderer>
         ) : (
           details
         ))}

frontend/src/components/features/chat/interactive-chat-box.tsx

@@ -8,6 +8,8 @@ import { GitControlBar } from "./git-control-bar";
 import { useConversationStore } from "#/state/conversation-store";
 import { useAgentState } from "#/hooks/use-agent-state";
 import { processFiles, processImages } from "#/utils/file-processing";
+import { useSubConversationTaskPolling } from "#/hooks/query/use-sub-conversation-task-polling";
+import { isTaskPolling } from "#/utils/utils";
 
 interface InteractiveChatBoxProps {
   onSubmit: (message: string, images: File[], files: File[]) => void;
@@ -24,10 +26,18 @@ export function InteractiveChatBox({ onSubmit }: InteractiveChatBoxProps) {
     removeFileLoading,
     addImageLoading,
     removeImageLoading,
+    subConversationTaskId,
   } = useConversationStore();
   const { curAgentState } = useAgentState();
   const { data: conversation } = useActiveConversation();
 
+  // Poll sub-conversation task to check if it's loading
+  const { taskStatus: subConversationTaskStatus } =
+    useSubConversationTaskPolling(
+      subConversationTaskId,
+      conversation?.conversation_id || null,
+    );
+
   // Helper function to validate and filter files
   const validateAndFilterFiles = (selectedFiles: File[]) => {
     const validation = validateFiles(selectedFiles, [...images, ...files]);
@@ -134,7 +144,8 @@ export function InteractiveChatBox({ onSubmit }: InteractiveChatBoxProps) {
 
   const isDisabled =
     curAgentState === AgentState.LOADING ||
-    curAgentState === AgentState.AWAITING_USER_CONFIRMATION;
+    curAgentState === AgentState.AWAITING_USER_CONFIRMATION ||
+    isTaskPolling(subConversationTaskStatus);
 
   return (
     <div data-testid="interactive-chat-box">

frontend/src/components/features/chat/task-tracking-observation-content.tsx

@@ -1,6 +1,5 @@
 import { TaskTrackingObservation } from "#/types/core/observations";
 import { TaskListSection } from "./task-tracking/task-list-section";
-import { ResultSection } from "./task-tracking/result-section";
 
 interface TaskTrackingObservationContentProps {
   event: TaskTrackingObservation;
@@ -16,11 +15,6 @@ export function TaskTrackingObservationContent({
     <div className="flex flex-col gap-4">
       {/* Task List section - only show for 'plan' command */}
       {shouldShowTaskList && <TaskListSection taskList={taskList} />}
-
-      {/* Result message - only show if there's meaningful content */}
-      {event.content && event.content.trim() && (
-        <ResultSection content={event.content} />
-      )}
     </div>
   );
 }

frontend/src/components/features/chat/task-tracking/result-section.tsx

@@ -1,21 +0,0 @@
-import { useTranslation } from "react-i18next";
-import { Typography } from "#/ui/typography";
-
-interface ResultSectionProps {
-  content: string;
-}
-
-export function ResultSection({ content }: ResultSectionProps) {
-  const { t } = useTranslation();
-
-  return (
-    <div className="flex flex-col gap-2">
-      <div className="flex items-center justify-between">
-        <Typography.H3>{t("TASK_TRACKING_OBSERVATION$RESULT")}</Typography.H3>
-      </div>
-      <div className="p-3 bg-gray-900 rounded-md overflow-auto text-gray-300 shadow-inner">
-        <pre className="whitespace-pre-wrap text-sm">{content.trim()}</pre>
-      </div>
-    </div>
-  );
-}

frontend/src/components/features/chat/task-tracking/task-item.tsx

@@ -1,7 +1,11 @@
+import { useMemo } from "react";
 import { useTranslation } from "react-i18next";
+import CircleIcon from "#/icons/u-circle.svg?react";
+import CheckCircleIcon from "#/icons/u-check-circle.svg?react";
+import LoadingIcon from "#/icons/loading.svg?react";
+import { I18nKey } from "#/i18n/declaration";
+import { cn } from "#/utils/utils";
 import { Typography } from "#/ui/typography";
-import { StatusIcon } from "./status-icon";
-import { StatusBadge } from "./status-badge";
 
 interface TaskItemProps {
   task: {
@@ -10,33 +14,47 @@ interface TaskItemProps {
     status: "todo" | "in_progress" | "done";
     notes?: string;
   };
-  index: number;
 }
 
-export function TaskItem({ task, index }: TaskItemProps) {
+export function TaskItem({ task }: TaskItemProps) {
   const { t } = useTranslation();
 
+  const icon = useMemo(() => {
+    switch (task.status) {
+      case "todo":
+        return <CircleIcon className="w-4 h-4 text-[#ffffff]" />;
+      case "in_progress":
+        return <LoadingIcon className="w-4 h-4 text-[#ffffff]" />;
+      case "done":
+        return <CheckCircleIcon className="w-4 h-4 text-[#A3A3A3]" />;
+      default:
+        return <CircleIcon className="w-4 h-4 text-[#ffffff]" />;
+    }
+  }, [task.status]);
+
+  const isDoneStatus = task.status === "done";
+
   return (
-    <div className="border-l-2 border-gray-600 pl-3">
-      <div className="flex items-start gap-2">
-        <StatusIcon status={task.status} />
-        <div className="flex-1">
-          <div className="flex items-center gap-2 mb-1">
-            <Typography.Text className="text-sm text-gray-400">
-              {index + 1}.
-            </Typography.Text>
-            <StatusBadge status={task.status} />
-          </div>
-          <h4 className="font-medium text-white mb-1">{task.title}</h4>
-          <Typography.Text className="text-xs text-gray-400 mb-1">
-            {t("TASK_TRACKING_OBSERVATION$TASK_ID")}: {task.id}
-          </Typography.Text>
-          {task.notes && (
-            <Typography.Text className="text-sm text-gray-300 italic">
-              {t("TASK_TRACKING_OBSERVATION$TASK_NOTES")}: {task.notes}
-            </Typography.Text>
+    <div
+      className="flex gap-[14px] items-center px-4 py-2 w-full"
+      data-name="item"
+    >
+      <div className="shrink-0">{icon}</div>
+      <div className="flex flex-col items-start justify-center leading-[20px] text-nowrap whitespace-pre font-normal">
+        <Typography.Text
+          className={cn(
+            "text-[12px] text-white",
+            isDoneStatus && "text-[#A3A3A3]",
           )}
-        </div>
+        >
+          {task.title}
+        </Typography.Text>
+        <Typography.Text className="text-[10px] text-[#A3A3A3] font-normal">
+          {t(I18nKey.TASK_TRACKING_OBSERVATION$TASK_ID)}: {task.id}
+        </Typography.Text>
+        <Typography.Text className="text-[10px] text-[#A3A3A3]">
+          {t(I18nKey.TASK_TRACKING_OBSERVATION$TASK_NOTES)}: {task.notes}
+        </Typography.Text>
       </div>
     </div>
   );

frontend/src/components/features/chat/task-tracking/task-list-section.tsx

@@ -1,5 +1,7 @@
 import { useTranslation } from "react-i18next";
 import { TaskItem } from "./task-item";
+import LessonPlanIcon from "#/icons/lesson-plan.svg?react";
+import { I18nKey } from "#/i18n/declaration";
 import { Typography } from "#/ui/typography";
 
 interface TaskListSectionProps {
@@ -15,19 +17,20 @@ export function TaskListSection({ taskList }: TaskListSectionProps) {
   const { t } = useTranslation();
 
   return (
-    <div className="flex flex-col gap-2">
-      <div className="flex items-center justify-between">
-        <Typography.H3>
-          {t("TASK_TRACKING_OBSERVATION$TASK_LIST")} ({taskList.length}{" "}
-          {taskList.length === 1 ? "item" : "items"})
-        </Typography.H3>
+    <div className="flex flex-col overflow-clip bg-[#25272d] border border-[#525252] rounded-[12px] w-full">
+      {/* Header Tabs */}
+      <div className="flex gap-1 items-center border-b border-[#525252] h-[41px] px-2 shrink-0">
+        <LessonPlanIcon className="shrink-0 w-4.5 h-4.5 text-[#9299aa]" />
+        <Typography.Text className="text-[11px] text-nowrap text-white tracking-[0.11px] font-medium leading-[16px] whitespace-pre">
+          {t(I18nKey.COMMON$TASKS)}
+        </Typography.Text>
       </div>
-      <div className="p-3 bg-gray-900 rounded-md overflow-auto text-gray-300 max-h-[400px] shadow-inner">
-        <div className="space-y-3">
-          {taskList.map((task, index) => (
-            <TaskItem key={task.id} task={task} index={index} />
-          ))}
-        </div>
+
+      {/* Task Items */}
+      <div>
+        {taskList.map((task) => (
+          <TaskItem key={task.id} task={task} />
+        ))}
       </div>
     </div>
   );

frontend/src/components/features/context-menu/context-menu-icon-text-with-description.tsx

@@ -0,0 +1,39 @@
+import React from "react";
+import { ContextMenuIconText } from "./context-menu-icon-text";
+import { Typography } from "#/ui/typography";
+import { cn } from "#/utils/utils";
+
+interface ContextMenuIconTextWithDescriptionProps {
+  icon: React.ComponentType<{ className?: string }>;
+  title: string;
+  description: string;
+  className?: string;
+  iconClassName?: string;
+}
+
+export function ContextMenuIconTextWithDescription({
+  icon,
+  title,
+  description,
+  className,
+  iconClassName,
+}: ContextMenuIconTextWithDescriptionProps) {
+  return (
+    <div
+      className={cn(
+        "flex flex-col gap-1 justify-center hover:bg-[#5C5D62] rounded p-2",
+        className,
+      )}
+    >
+      <ContextMenuIconText
+        icon={icon}
+        text={title}
+        className="px-0"
+        iconClassName={iconClassName}
+      />
+      <Typography.Text className="text-[#A3A3A3] text-[10px] font-normal whitespace-pre-wrap break-words">
+        {description}
+      </Typography.Text>
+    </div>
+  );
+}

frontend/src/components/features/controls/agent-status.tsx

@@ -7,13 +7,14 @@ import { ChatStopButton } from "../chat/chat-stop-button";
 import { AgentState } from "#/types/agent-state";
 import ClockIcon from "#/icons/u-clock-three.svg?react";
 import { ChatResumeAgentButton } from "../chat/chat-play-button";
-import { cn } from "#/utils/utils";
+import { cn, isTaskPolling } from "#/utils/utils";
 import { AgentLoading } from "./agent-loading";
 import { useConversationStore } from "#/state/conversation-store";
 import CircleErrorIcon from "#/icons/circle-error.svg?react";
 import { useAgentState } from "#/hooks/use-agent-state";
 import { useUnifiedWebSocketStatus } from "#/hooks/use-unified-websocket-status";
 import { useTaskPolling } from "#/hooks/query/use-task-polling";
+import { useSubConversationTaskPolling } from "#/hooks/query/use-sub-conversation-task-polling";
 
 export interface AgentStatusProps {
   className?: string;
@@ -38,6 +39,15 @@ export function AgentStatus({
   const { data: conversation } = useActiveConversation();
   const { taskStatus } = useTaskPolling();
 
+  const { subConversationTaskId } = useConversationStore();
+
+  // Poll sub-conversation task to track its loading state
+  const { taskStatus: subConversationTaskStatus } =
+    useSubConversationTaskPolling(
+      subConversationTaskId,
+      conversation?.conversation_id || null,
+    );
+
   const statusCode = getStatusCode(
     curStatusMessage,
     webSocketStatus,
@@ -45,17 +55,16 @@ export function AgentStatus({
     conversation?.runtime_status || null,
     curAgentState,
     taskStatus,
+    subConversationTaskStatus,
   );
 
-  const isTaskLoading =
-    taskStatus && taskStatus !== "ERROR" && taskStatus !== "READY";
-
   const shouldShownAgentLoading =
     isPausing ||
     curAgentState === AgentState.INIT ||
     curAgentState === AgentState.LOADING ||
     (webSocketStatus === "CONNECTING" && taskStatus !== "ERROR") ||
-    isTaskLoading;
+    isTaskPolling(taskStatus) ||
+    isTaskPolling(subConversationTaskStatus);
 
   const shouldShownAgentError =
     curAgentState === AgentState.ERROR ||

frontend/src/components/features/controls/server-status.tsx

@@ -6,6 +6,7 @@ import { AgentState } from "#/types/agent-state";
 import { useAgentState } from "#/hooks/use-agent-state";
 import { useTaskPolling } from "#/hooks/query/use-task-polling";
 import { getStatusColor } from "#/utils/utils";
+import { useErrorMessageStore } from "#/stores/error-message-store";
 
 export interface ServerStatusProps {
   className?: string;
@@ -21,6 +22,7 @@ export function ServerStatus({
   const { curAgentState } = useAgentState();
   const { t } = useTranslation();
   const { isTask, taskStatus, taskDetail } = useTaskPolling();
+  const { errorMessage } = useErrorMessageStore();
 
   const isStartingStatus =
     curAgentState === AgentState.LOADING || curAgentState === AgentState.INIT;
@@ -69,7 +71,7 @@ export function ServerStatus({
       return t(I18nKey.COMMON$SERVER_STOPPED);
     }
     if (curAgentState === AgentState.ERROR) {
-      return t(I18nKey.COMMON$ERROR);
+      return errorMessage || t(I18nKey.COMMON$ERROR);
     }
     return t(I18nKey.COMMON$RUNNING);
   };
@@ -79,7 +81,7 @@ export function ServerStatus({
   return (
     <div className={className} data-testid="server-status">
       <div className="flex items-center">
-        <DebugStackframeDot className="w-6 h-6" color={statusColor} />
+        <DebugStackframeDot className="w-6 h-6 shrink-0" color={statusColor} />
         <span className="text-[13px] text-white font-normal">{statusText}</span>
       </div>
     </div>

frontend/src/components/features/conversation-panel/conversation-card/conversation-card-footer.tsx

@@ -39,7 +39,7 @@ export function ConversationCardFooter({
       {(createdAt ?? lastUpdatedAt) && (
         <p className="text-xs text-[#A3A3A3] flex-1 text-right">
           <time>
-            {`${formatTimeDelta(new Date(lastUpdatedAt ?? createdAt))} ${t(I18nKey.CONVERSATION$AGO)}`}
+            {`${formatTimeDelta(lastUpdatedAt ?? createdAt)} ${t(I18nKey.CONVERSATION$AGO)}`}
           </time>
         </p>
       )}

frontend/src/components/features/conversation-panel/conversation-card/conversation-repo-link.tsx

@@ -3,12 +3,13 @@ import { FaCodeBranch } from "react-icons/fa";
 import { IconType } from "react-icons/lib";
 import { RepositorySelection } from "#/api/open-hands.types";
 import { Provider } from "#/types/settings";
+import AzureDevOpsLogo from "#/assets/branding/azure-devops-logo.svg?react";
 
 interface ConversationRepoLinkProps {
   selectedRepository: RepositorySelection;
 }
 
-const providerIcon: Record<Provider, IconType> = {
+const providerIcon: Partial<Record<Provider, IconType>> = {
   bitbucket: FaBitbucket,
   github: FaGithub,
   gitlab: FaGitlab,
@@ -26,6 +27,9 @@ export function ConversationRepoLink({
     <div className="flex items-center gap-3 flex-1">
       <div className="flex items-center gap-1">
         {Icon && <Icon size={14} className="text-[#A3A3A3]" />}
+        {selectedRepository.git_provider === "azure_devops" && (
+          <AzureDevOpsLogo className="text-[#A3A3A3] w-[14px] h-[14px]" />
+        )}
         <span
           data-testid="conversation-card-selected-repository"
           className="text-xs text-[#A3A3A3] whitespace-nowrap overflow-hidden text-ellipsis max-w-44"

frontend/src/components/features/conversation-panel/conversation-card/conversation-version-badge.tsx

@@ -25,10 +25,7 @@ export function ConversationVersionBadge({
     <Tooltip content={tooltipText} placement="top">
       <span
         className={cn(
-          "inline-flex items-center px-1.5 py-0.5 rounded text-[10px] font-semibold shrink-0 cursor-help lowercase",
-          version === "V1"
-            ? "bg-green-500/20 text-green-500"
-            : "bg-neutral-500/20 text-neutral-400",
+          "inline-flex items-center px-1.5 py-0.5 rounded text-[10px] font-semibold shrink-0 cursor-help lowercase bg-neutral-500/20 text-neutral-400",
           isConversationArchived && "opacity-60",
         )}
       >

frontend/src/components/features/conversation-panel/start-task-card/start-task-card-footer.tsx

@@ -31,7 +31,7 @@ export function StartTaskCardFooter({
         {createdAt && (
           <p className="text-xs text-[#A3A3A3] flex-1 text-right">
             <time>
-              {`${formatTimeDelta(new Date(createdAt))} ${t(I18nKey.CONVERSATION$AGO)}`}
+              {`${formatTimeDelta(createdAt)} ${t(I18nKey.CONVERSATION$AGO)}`}
             </time>
           </p>
         )}

frontend/src/components/features/conversation-panel/start-task-card/start-task-status-indicator.tsx

@@ -19,6 +19,7 @@ export function StartTaskStatusIndicator({
       case "PREPARING_REPOSITORY":
       case "RUNNING_SETUP_SCRIPT":
       case "SETTING_UP_GIT_HOOKS":
+      case "SETTING_UP_SKILLS":
       case "STARTING_CONVERSATION":
         return "bg-yellow-500 animate-pulse";
       default:

frontend/src/components/features/conversation/metrics-modal/context-window-section.tsx

@@ -12,7 +12,8 @@ export function ContextWindowSection({
 }: ContextWindowSectionProps) {
   const { t } = useTranslation();
 
-  const usagePercentage = (perTurnToken / contextWindow) * 100;
+  const usagePercentage =
+    contextWindow > 0 ? (perTurnToken / contextWindow) * 100 : 0;
   const progressWidth = Math.min(100, usagePercentage);
 
   return (

frontend/src/components/features/home/git-provider-dropdown/git-provider-dropdown.tsx

@@ -51,6 +51,8 @@ export function GitProviderDropdown({
         return "GitLab";
       case "bitbucket":
         return "Bitbucket";
+      case "azure_devops":
+        return "Azure DevOps";
       case "enterprise_sso":
         return "Enterprise SSO";
       default:

frontend/src/components/features/home/recent-conversations/recent-conversation.tsx

@@ -67,12 +67,14 @@ export function RecentConversation({ conversation }: RecentConversationProps) {
               </div>
             ) : null}
           </div>
-          <span>
-            {formatTimeDelta(
-              new Date(conversation.created_at || conversation.last_updated_at),
-            )}{" "}
-            {t(I18nKey.CONVERSATION$AGO)}
-          </span>
+          {(conversation.created_at || conversation.last_updated_at) && (
+            <span>
+              {formatTimeDelta(
+                conversation.created_at || conversation.last_updated_at,
+              )}{" "}
+              {t(I18nKey.CONVERSATION$AGO)}
+            </span>
+          )}
         </div>
       </button>
     </Link>

frontend/src/components/features/home/tasks/task-card.tsx

@@ -56,6 +56,15 @@ export function TaskCard({ task }: TaskCardProps) {
     const issueType =
       task.task_type === "OPEN_ISSUE" ? "issues" : "pull-requests";
     href = `https://bitbucket.org/${task.repo}/${issueType}/${task.issue_number}`;
+  } else if (task.git_provider === "azure_devops") {
+    // Azure DevOps URL format: https://dev.azure.com/{organization}/{project}/_workitems/edit/{id}
+    // or https://dev.azure.com/{organization}/{project}/_git/{repo}/pullrequest/{id}
+    const azureDevOpsBaseUrl = "https://dev.azure.com";
+    if (task.task_type === "OPEN_ISSUE") {
+      href = `${azureDevOpsBaseUrl}/${task.repo}/_workitems/edit/${task.issue_number}`;
+    } else {
+      href = `${azureDevOpsBaseUrl}/${task.repo}/_git/${task.repo.split("/")[1]}/pullrequest/${task.issue_number}`;
+    }
   } else {
     const hrefType = task.task_type === "OPEN_ISSUE" ? "issues" : "pull";
     href = `https://github.com/${task.repo}/${hrefType}/${task.issue_number}`;

frontend/src/components/features/markdown/headings.tsx

@@ -8,7 +8,7 @@ export function h1({
   React.HTMLAttributes<HTMLHeadingElement> &
   ExtraProps) {
   return (
-    <h1 className="text-[32px] text-white font-bold leading-8 mb-4 mt-6 first:mt-0">
+    <h1 className="text-2xl text-white font-bold leading-8 mb-4 mt-6 first:mt-0">
       {children}
     </h1>
   );

frontend/src/components/features/markdown/markdown-renderer.tsx

@@ -0,0 +1,80 @@
+import Markdown, { Components } from "react-markdown";
+import remarkGfm from "remark-gfm";
+import remarkBreaks from "remark-breaks";
+import { code } from "./code";
+import { ul, ol } from "./list";
+import { paragraph } from "./paragraph";
+import { anchor } from "./anchor";
+import { h1, h2, h3, h4, h5, h6 } from "./headings";
+
+interface MarkdownRendererProps {
+  /**
+   * The markdown content to render. Can be passed as children (string) or content prop.
+   */
+  children?: string;
+  content?: string;
+  /**
+   * Additional or override components for markdown elements.
+   * Default components (code, ul, ol) are always included unless overridden.
+   */
+  components?: Partial<Components>;
+  /**
+   * Whether to include standard components (anchor, paragraph).
+   * Defaults to false.
+   */
+  includeStandard?: boolean;
+  /**
+   * Whether to include heading components (h1-h6).
+   * Defaults to false.
+   */
+  includeHeadings?: boolean;
+}
+
+/**
+ * A reusable Markdown renderer component that provides consistent
+ * markdown rendering across the application.
+ *
+ * By default, includes:
+ * - code, ul, ol components
+ * - remarkGfm and remarkBreaks plugins
+ *
+ * Can be extended with:
+ * - includeStandard: adds anchor and paragraph components
+ * - includeHeadings: adds h1-h6 heading components
+ * - components prop: allows custom overrides or additional components
+ */
+export function MarkdownRenderer({
+  children,
+  content,
+  components: customComponents,
+  includeStandard = false,
+  includeHeadings = false,
+}: MarkdownRendererProps) {
+  // Build the components object with defaults and optional additions
+  const components: Components = {
+    code,
+    ul,
+    ol,
+    ...(includeStandard && {
+      a: anchor,
+      p: paragraph,
+    }),
+    ...(includeHeadings && {
+      h1,
+      h2,
+      h3,
+      h4,
+      h5,
+      h6,
+    }),
+    ...customComponents, // Custom components override defaults
+  };
+
+  const markdownContent = content ?? children ?? "";
+
+  return (
+    <Markdown components={components} remarkPlugins={[remarkGfm, remarkBreaks]}>
+      {markdownContent}
+    </Markdown>
+  );
+}

frontend/src/components/features/microagent-management/microagent-management-view-microagent-content.tsx

@@ -1,16 +1,10 @@
 import { useTranslation } from "react-i18next";
 import { Spinner } from "@heroui/react";
-import Markdown from "react-markdown";
-import remarkGfm from "remark-gfm";
-import remarkBreaks from "remark-breaks";
-import { code } from "../markdown/code";
-import { ul, ol } from "../markdown/list";
-import { paragraph } from "../markdown/paragraph";
-import { anchor } from "../markdown/anchor";
 import { useMicroagentManagementStore } from "#/state/microagent-management-store";
 import { useRepositoryMicroagentContent } from "#/hooks/query/use-repository-microagent-content";
 import { I18nKey } from "#/i18n/declaration";
 import { extractRepositoryInfo } from "#/utils/utils";
+import { MarkdownRenderer } from "../markdown/markdown-renderer";
 
 export function MicroagentManagementViewMicroagentContent() {
   const { t } = useTranslation();
@@ -49,18 +43,9 @@ export function MicroagentManagementViewMicroagentContent() {
         </div>
       )}
       {microagentData && !isLoading && !error && (
-        <Markdown
-          components={{
-            code,
-            ul,
-            ol,
-            a: anchor,
-            p: paragraph,
-          }}
-          remarkPlugins={[remarkGfm, remarkBreaks]}
-        >
+        <MarkdownRenderer includeStandard>
           {microagentData.content}
-        </Markdown>
+        </MarkdownRenderer>
       )}
     </div>
   );

frontend/src/components/features/settings/git-settings/azure-devops-token-help-anchor.tsx

@@ -0,0 +1,20 @@
+import { useTranslation } from "react-i18next";
+import { I18nKey } from "#/i18n/declaration";
+
+export function AzureDevOpsTokenHelpAnchor() {
+  const { t } = useTranslation();
+
+  return (
+    <p data-testid="azure-devops-token-help-anchor" className="text-xs">
+      <a
+        href="https://learn.microsoft.com/en-us/azure/devops/organizations/accounts/use-personal-access-tokens-to-authenticate"
+        target="_blank"
+        className="underline underline-offset-2"
+        rel="noopener noreferrer"
+        aria-label={t(I18nKey.GIT$AZURE_DEVOPS_TOKEN_HELP)}
+      >
+        {t(I18nKey.GIT$AZURE_DEVOPS_TOKEN_HELP)}
+      </a>
+    </p>
+  );
+}

frontend/src/components/features/settings/git-settings/azure-devops-token-input.tsx

@@ -0,0 +1,64 @@
+import { useTranslation } from "react-i18next";
+import { I18nKey } from "#/i18n/declaration";
+import { SettingsInput } from "../settings-input";
+import { AzureDevOpsTokenHelpAnchor } from "./azure-devops-token-help-anchor";
+import { KeyStatusIcon } from "../key-status-icon";
+
+interface AzureDevOpsTokenInputProps {
+  onChange: (value: string) => void;
+  onAzureDevOpsHostChange: (value: string) => void;
+  isAzureDevOpsTokenSet: boolean;
+  name: string;
+  azureDevOpsHostSet: string | null | undefined;
+}
+
+export function AzureDevOpsTokenInput({
+  onChange,
+  onAzureDevOpsHostChange,
+  isAzureDevOpsTokenSet,
+  name,
+  azureDevOpsHostSet,
+}: AzureDevOpsTokenInputProps) {
+  const { t } = useTranslation();
+
+  return (
+    <div className="flex flex-col gap-6">
+      <SettingsInput
+        testId={name}
+        name={name}
+        onChange={onChange}
+        label={t(I18nKey.GIT$AZURE_DEVOPS_TOKEN)}
+        type="password"
+        className="w-full max-w-[680px]"
+        placeholder={isAzureDevOpsTokenSet ? "<hidden>" : ""}
+        startContent={
+          isAzureDevOpsTokenSet && (
+            <KeyStatusIcon
+              testId="azure-devops-set-token-indicator"
+              isSet={isAzureDevOpsTokenSet}
+            />
+          )
+        }
+      />
+
+      <SettingsInput
+        onChange={onAzureDevOpsHostChange || (() => {})}
+        name="azure-devops-host-input"
+        testId="azure-devops-host-input"
+        label={t(I18nKey.GIT$AZURE_DEVOPS_HOST)}
+        type="text"
+        className="w-full max-w-[680px]"
+        placeholder={t(I18nKey.GIT$AZURE_DEVOPS_HOST_PLACEHOLDER)}
+        defaultValue={azureDevOpsHostSet || undefined}
+        startContent={
+          azureDevOpsHostSet &&
+          azureDevOpsHostSet.trim() !== "" && (
+            <KeyStatusIcon testId="azure-devops-set-host-indicator" isSet />
+          )
+        }
+      />
+
+      <AzureDevOpsTokenHelpAnchor />
+    </div>
+  );
+}

frontend/src/components/features/settings/git-settings/configure-azure-devops-anchor.tsx

@@ -0,0 +1,37 @@
+import { useTranslation } from "react-i18next";
+import { I18nKey } from "#/i18n/declaration";
+import { useConfig } from "#/hooks/query/use-config";
+import { useAuthUrl } from "#/hooks/use-auth-url";
+import { BrandButton } from "../brand-button";
+
+export function ConfigureAzureDevOpsAnchor() {
+  const { t } = useTranslation();
+  const { data: config } = useConfig();
+
+  const authUrl = useAuthUrl({
+    appMode: config?.APP_MODE ?? null,
+    identityProvider: "azure_devops",
+    authUrl: config?.AUTH_URL,
+  });
+
+  const handleOAuthFlow = () => {
+    if (!authUrl) {
+      return;
+    }
+
+    window.location.href = authUrl;
+  };
+
+  return (
+    <div data-testid="configure-azure-devops-button" className="py-9">
+      <BrandButton
+        type="button"
+        variant="primary"
+        className="w-55"
+        onClick={handleOAuthFlow}
+      >
+        {t(I18nKey.AZURE_DEVOPS$CONNECT_ACCOUNT)}
+      </BrandButton>
+    </div>
+  );
+}

frontend/src/components/features/waitlist/auth-modal.tsx

@@ -8,6 +8,7 @@ import { BrandButton } from "../settings/brand-button";
 import GitHubLogo from "#/assets/branding/github-logo.svg?react";
 import GitLabLogo from "#/assets/branding/gitlab-logo.svg?react";
 import BitbucketLogo from "#/assets/branding/bitbucket-logo.svg?react";
+import AzureDevOpsLogo from "#/assets/branding/azure-devops-logo.svg?react";
 import { useAuthUrl } from "#/hooks/use-auth-url";
 import { GetConfigResponse } from "#/api/option-service/option.types";
 import { Provider } from "#/types/settings";
@@ -41,6 +42,12 @@ export function AuthModal({
     authUrl,
   });
 
+  const azureDevOpsAuthUrl = useAuthUrl({
+    appMode: appMode || null,
+    identityProvider: "azure_devops",
+    authUrl,
+  });
+
   const enterpriseSsoUrl = useAuthUrl({
     appMode: appMode || null,
     identityProvider: "enterprise_sso",
@@ -71,6 +78,13 @@ export function AuthModal({
     }
   };
 
+  const handleAzureDevOpsAuth = () => {
+    if (azureDevOpsAuthUrl) {
+      // Always start the OIDC flow, let the backend handle TOS check
+      window.location.href = azureDevOpsAuthUrl;
+    }
+  };
+
   const handleEnterpriseSsoAuth = () => {
     if (enterpriseSsoUrl) {
       trackLoginButtonClick({ provider: "enterprise_sso" });
@@ -92,6 +106,10 @@ export function AuthModal({
     providersConfigured &&
     providersConfigured.length > 0 &&
     providersConfigured.includes("bitbucket");
+  const showAzureDevOps =
+    providersConfigured &&
+    providersConfigured.length > 0 &&
+    providersConfigured.includes("azure_devops");
   const showEnterpriseSso =
     providersConfigured &&
     providersConfigured.length > 0 &&
@@ -154,6 +172,18 @@ export function AuthModal({
                 </BrandButton>
               )}
 
+              {showAzureDevOps && (
+                <BrandButton
+                  type="button"
+                  variant="primary"
+                  onClick={handleAzureDevOpsAuth}
+                  className="w-full font-semibold"
+                  startContent={<AzureDevOpsLogo width={20} height={20} />}
+                >
+                  {t(I18nKey.AZURE_DEVOPS$CONNECT_ACCOUNT)}
+                </BrandButton>
+              )}
+
               {showEnterpriseSso && (
                 <BrandButton
                   type="button"

frontend/src/components/shared/git-provider-icon.tsx

@@ -1,5 +1,6 @@
 import { FaBitbucket, FaGithub, FaGitlab } from "react-icons/fa6";
 import { Provider } from "#/types/settings";
+import AzureDevOpsLogo from "#/assets/branding/azure-devops-logo.svg?react";
 
 interface GitProviderIconProps {
   gitProvider: Provider;
@@ -13,8 +14,13 @@ export function GitProviderIcon({
   return (
     <>
       {gitProvider === "github" && <FaGithub size={14} className={className} />}
-      {gitProvider === "gitlab" && <FaGitlab className={className} />}
-      {gitProvider === "bitbucket" && <FaBitbucket className={className} />}
+      {gitProvider === "gitlab" && <FaGitlab size={14} className={className} />}
+      {gitProvider === "bitbucket" && (
+        <FaBitbucket size={14} className={className} />
+      )}
+      {gitProvider === "azure_devops" && (
+        <AzureDevOpsLogo className={`${className} w-[14px] h-[14px]`} />
+      )}
     </>
   );
 }

frontend/src/components/shared/modals/settings/model-selector.tsx

@@ -21,7 +21,11 @@ interface ModelSelectorProps {
   isDisabled?: boolean;
   models: Record<string, { separator: string; models: string[] }>;
   currentModel?: string;
-  onChange?: (model: string | null) => void;
+  onChange?: (provider: string | null, model: string | null) => void;
+  onDefaultValuesChanged?: (
+    provider: string | null,
+    model: string | null,
+  ) => void;
   wrapperClassName?: string;
   labelClassName?: string;
 }
@@ -31,6 +35,7 @@ export function ModelSelector({
   models,
   currentModel,
   onChange,
+  onDefaultValuesChanged,
   wrapperClassName,
   labelClassName,
 }: ModelSelectorProps) {
@@ -56,6 +61,7 @@ export function ModelSelector({
       setLitellmId(currentModel);
       setSelectedProvider(provider);
       setSelectedModel(model);
+      onDefaultValuesChanged?.(provider, model);
     }
   }, [currentModel]);
 
@@ -65,6 +71,7 @@ export function ModelSelector({
 
     const separator = models[provider]?.separator || "";
     setLitellmId(provider + separator);
+    onChange?.(provider, null);
   };
 
   const handleChangeModel = (model: string) => {
@@ -76,7 +83,7 @@ export function ModelSelector({
     }
     setLitellmId(fullModel);
     setSelectedModel(model);
-    onChange?.(fullModel);
+    onChange?.(selectedProvider, model);
   };
 
   const clear = () => {

frontend/src/components/v1/chat/event-content-helpers/create-skill-ready-event.ts

@@ -0,0 +1,56 @@
+import { MessageEvent } from "#/types/v1/core";
+import { BaseEvent } from "#/types/v1/core/base/event";
+import { getSkillReadyContent } from "./get-skill-ready-content";
+
+/**
+ * Synthetic event type for Skill Ready events.
+ * This extends BaseEvent and includes a marker to identify it as a skill ready event.
+ */
+export interface SkillReadyEvent extends BaseEvent {
+  _isSkillReadyEvent: true;
+  _skillReadyContent: string;
+}
+
+/**
+ * Type guard for Skill Ready events.
+ */
+export const isSkillReadyEvent = (event: unknown): event is SkillReadyEvent =>
+  typeof event === "object" &&
+  event !== null &&
+  "_isSkillReadyEvent" in event &&
+  event._isSkillReadyEvent === true;
+
+/**
+ * Creates a synthetic "Skill Ready" event from a user MessageEvent.
+ * This event appears as originating from the agent and contains formatted
+ * information about activated skills and extended content.
+ */
+export const createSkillReadyEvent = (
+  userEvent: MessageEvent,
+): SkillReadyEvent => {
+  // Support both activated_skills and activated_microagents field names
+  const activatedSkills =
+    (userEvent as unknown as { activated_skills?: string[] })
+      .activated_skills ||
+    userEvent.activated_microagents ||
+    [];
+
+  const extendedContent = userEvent.extended_content || [];
+
+  // Only create event if we have skills or extended content
+  if (activatedSkills.length === 0 && extendedContent.length === 0) {
+    throw new Error(
+      "Cannot create skill ready event without activated skills or extended content",
+    );
+  }
+
+  const content = getSkillReadyContent(activatedSkills, extendedContent);
+
+  return {
+    id: `${userEvent.id}-skill-ready`,
+    timestamp: userEvent.timestamp,
+    source: "agent",
+    _isSkillReadyEvent: true,
+    _skillReadyContent: content,
+  };
+};

frontend/src/components/v1/chat/event-content-helpers/get-action-content.ts

@@ -4,6 +4,7 @@ import i18n from "#/i18n";
 import { SecurityRisk } from "#/types/v1/core/base/common";
 import {
   ExecuteBashAction,
+  TerminalAction,
   FileEditorAction,
   StrReplaceEditorAction,
   MCPToolAction,
@@ -58,7 +59,7 @@ const getFileEditorActionContent = (
 
 // Command Actions
 const getExecuteBashActionContent = (
-  event: ActionEvent<ExecuteBashAction>,
+  event: ActionEvent<ExecuteBashAction | TerminalAction>,
 ): string => {
   let content = `Command:\n\`${event.action.command}\``;
 
@@ -131,27 +132,61 @@ type BrowserAction =
 
 const getBrowserActionContent = (action: BrowserAction): string => {
   switch (action.kind) {
-    case "BrowserNavigateAction":
-      if ("url" in action) {
-        return `Browsing ${action.url}`;
+    case "BrowserNavigateAction": {
+      let content = `Browsing ${action.url}`;
+      if (action.new_tab) {
+        content += `\n**New Tab:** Yes`;
+      }
+      return content;
+    }
+    case "BrowserClickAction": {
+      let content = `**Element Index:** ${action.index}`;
+      if (action.new_tab) {
+        content += `\n**New Tab:** Yes`;
+      }
+      return content;
+    }
+    case "BrowserTypeAction": {
+      const textPreview =
+        action.text.length > 50
+          ? `${action.text.slice(0, 50)}...`
+          : action.text;
+      return `**Element Index:** ${action.index}\n**Text:** ${textPreview}`;
+    }
+    case "BrowserGetStateAction": {
+      if (action.include_screenshot) {
+        return `**Include Screenshot:** Yes`;
       }
-      break;
-    case "BrowserClickAction":
-    case "BrowserTypeAction":
-    case "BrowserGetStateAction":
-    case "BrowserGetContentAction":
-    case "BrowserScrollAction":
-    case "BrowserGoBackAction":
-    case "BrowserListTabsAction":
-    case "BrowserSwitchTabAction":
-    case "BrowserCloseTabAction":
-      // These browser actions typically don't need detailed content display
       return getNoContentActionContent();
+    }
+    case "BrowserGetContentAction": {
+      const parts: string[] = [];
+      if (action.extract_links) {
+        parts.push(`**Extract Links:** Yes`);
+      }
+      if (action.start_from_char > 0) {
+        parts.push(`**Start From Character:** ${action.start_from_char}`);
+      }
+      return parts.length > 0 ? parts.join("\n") : getNoContentActionContent();
+    }
+    case "BrowserScrollAction": {
+      return `**Direction:** ${action.direction}`;
+    }
+    case "BrowserGoBackAction": {
+      return getNoContentActionContent();
+    }
+    case "BrowserListTabsAction": {
+      return getNoContentActionContent();
+    }
+    case "BrowserSwitchTabAction": {
+      return `**Tab ID:** ${action.tab_id}`;
+    }
+    case "BrowserCloseTabAction": {
+      return `**Tab ID:** ${action.tab_id}`;
+    }
     default:
       return getNoContentActionContent();
   }
-
-  return getNoContentActionContent();
 };
 
 export const getActionContent = (event: ActionEvent): string => {
@@ -164,8 +199,9 @@ export const getActionContent = (event: ActionEvent): string => {
       return getFileEditorActionContent(action);
 
     case "ExecuteBashAction":
+    case "TerminalAction":
       return getExecuteBashActionContent(
-        event as ActionEvent<ExecuteBashAction>,
+        event as ActionEvent<ExecuteBashAction | TerminalAction>,
       );
 
     case "MCPToolAction":

frontend/src/components/v1/chat/event-content-helpers/get-event-content.tsx

@@ -1,10 +1,14 @@
 import { Trans } from "react-i18next";
-import { OpenHandsEvent } from "#/types/v1/core";
+import React from "react";
+import { OpenHandsEvent, ObservationEvent } from "#/types/v1/core";
 import { isActionEvent, isObservationEvent } from "#/types/v1/type-guards";
 import { MonoComponent } from "../../../features/chat/mono-component";
 import { PathComponent } from "../../../features/chat/path-component";
 import { getActionContent } from "./get-action-content";
 import { getObservationContent } from "./get-observation-content";
+import { TaskTrackingObservationContent } from "../task-tracking/task-tracking-observation-content";
+import { TaskTrackerObservation } from "#/types/v1/core/base/observation";
+import { SkillReadyEvent, isSkillReadyEvent } from "./create-skill-ready-event";
 import i18n from "#/i18n";
 
 const trimText = (text: string, maxLength: number): string => {
@@ -46,6 +50,7 @@ const getActionEventTitle = (event: OpenHandsEvent): React.ReactNode => {
 
   switch (actionType) {
     case "ExecuteBashAction":
+    case "TerminalAction":
       actionKey = "ACTION_MESSAGE$RUN";
       actionValues = {
         command: trimText(event.action.command, 80),
@@ -80,11 +85,20 @@ const getActionEventTitle = (event: OpenHandsEvent): React.ReactNode => {
       actionKey = "ACTION_MESSAGE$TASK_TRACKING";
       break;
     case "BrowserNavigateAction":
+    case "BrowserClickAction":
+    case "BrowserTypeAction":
+    case "BrowserGetStateAction":
+    case "BrowserGetContentAction":
+    case "BrowserScrollAction":
+    case "BrowserGoBackAction":
+    case "BrowserListTabsAction":
+    case "BrowserSwitchTabAction":
+    case "BrowserCloseTabAction":
       actionKey = "ACTION_MESSAGE$BROWSE";
       break;
     default:
       // For unknown actions, use the type name
-      return actionType.replace("Action", "").toUpperCase();
+      return String(actionType).replace("Action", "").toUpperCase();
   }
 
   if (actionKey) {
@@ -107,6 +121,7 @@ const getObservationEventTitle = (event: OpenHandsEvent): React.ReactNode => {
 
   switch (observationType) {
     case "ExecuteBashObservation":
+    case "TerminalObservation":
       observationKey = "OBSERVATION_MESSAGE$RUN";
       observationValues = {
         command: event.observation.command
@@ -156,16 +171,36 @@ const getObservationEventTitle = (event: OpenHandsEvent): React.ReactNode => {
   return observationType;
 };
 
-export const getEventContent = (event: OpenHandsEvent) => {
+export const getEventContent = (event: OpenHandsEvent | SkillReadyEvent) => {
   let title: React.ReactNode = "";
-  let details: string = "";
+  let details: string | React.ReactNode = "";
 
-  if (isActionEvent(event)) {
+  // Handle Skill Ready events first
+  if (isSkillReadyEvent(event)) {
+    // Use translation key if available, otherwise use "SKILL READY"
+    const skillReadyKey = "OBSERVATION_MESSAGE$SKILL_READY";
+    if (i18n.exists(skillReadyKey)) {
+      title = createTitleFromKey(skillReadyKey, {});
+    } else {
+      title = "Skill Ready";
+    }
+    details = event._skillReadyContent;
+  } else if (isActionEvent(event)) {
     title = getActionEventTitle(event);
     details = getActionContent(event);
   } else if (isObservationEvent(event)) {
     title = getObservationEventTitle(event);
-    details = getObservationContent(event);
+
+    // For TaskTrackerObservation, use React component instead of markdown
+    if (event.observation.kind === "TaskTrackerObservation") {
+      details = (
+        <TaskTrackingObservationContent
+          event={event as ObservationEvent<TaskTrackerObservation>}
+        />
+      );
+    } else {
+      details = getObservationContent(event);
+    }
   }
 
   return {

frontend/src/components/v1/chat/event-content-helpers/get-observation-content.ts

@@ -8,6 +8,7 @@ import {
   ThinkObservation,
   BrowserObservation,
   ExecuteBashObservation,
+  TerminalObservation,
   FileEditorObservation,
   StrReplaceEditorObservation,
   TaskTrackerObservation,
@@ -23,6 +24,15 @@ const getFileEditorObservationContent = (
     return `**Error:**\n${observation.error}`;
   }
 
+  // Extract text content from the observation if it exists
+  const textContent =
+    "content" in observation && Array.isArray(observation.content)
+      ? observation.content
+          .filter((c) => c.type === "text")
+          .map((c) => c.text)
+          .join("\n")
+      : null;
+
   const successMessage = getObservationResult(event) === "success";
 
   // For view commands or successful edits with content changes, format as code block
@@ -34,30 +44,34 @@ const getFileEditorObservationContent = (
       observation.new_content) ||
     observation.command === "view"
   ) {
-    return `\`\`\`\n${observation.output}\n\`\`\``;
+    // Prefer content over output for view commands, fallback to output if content is not available
+    const displayContent = textContent || observation.output;
+    return `\`\`\`\n${displayContent}\n\`\`\``;
   }
 
-  // For other commands, return the output as-is
-  return observation.output;
+  // For other commands, prefer content if available, otherwise use output
+  return textContent || observation.output;
 };
 
 // Command Observations
-const getExecuteBashObservationContent = (
-  event: ObservationEvent<ExecuteBashObservation>,
+const getTerminalObservationContent = (
+  event: ObservationEvent<ExecuteBashObservation | TerminalObservation>,
 ): string => {
   const { observation } = event;
 
-  let { output } = observation;
+  // Extract text content from the observation
+  const textContent = observation.content
+    .filter((c) => c.type === "text")
+    .map((c) => c.text)
+    .join("\n");
 
-  if (!output) {
-    output = "";
+  let content = textContent || "";
+
+  if (content.length > MAX_CONTENT_LENGTH) {
+    content = `${content.slice(0, MAX_CONTENT_LENGTH)}...`;
   }
 
-  if (output.length > MAX_CONTENT_LENGTH) {
-    output = `${output.slice(0, MAX_CONTENT_LENGTH)}...`;
-  }
-
-  return `Output:\n\`\`\`sh\n${output.trim() || i18n.t("OBSERVATION$COMMAND_NO_OUTPUT")}\n\`\`\``;
+  return `Output:\n\`\`\`sh\n${content.trim() || i18n.t("OBSERVATION$COMMAND_NO_OUTPUT")}\n\`\`\``;
 };
 
 // Tool Observations
@@ -66,14 +80,23 @@ const getBrowserObservationContent = (
 ): string => {
   const { observation } = event;
 
+  // Extract text content from the observation
+  const textContent =
+    "content" in observation && Array.isArray(observation.content)
+      ? observation.content
+          .filter((c) => c.type === "text")
+          .map((c) => c.text)
+          .join("\n")
+      : "";
+
   let contentDetails = "";
 
-  if ("error" in observation && observation.error) {
-    contentDetails += `**Error:**\n${observation.error}\n\n`;
+  if ("is_error" in observation && observation.is_error) {
+    contentDetails += `**Error:**\n${textContent}`;
+  } else {
+    contentDetails += `**Output:**\n${textContent}`;
   }
 
-  contentDetails += `**Output:**\n${observation.output}`;
-
   if (contentDetails.length > MAX_CONTENT_LENGTH) {
     contentDetails = `${contentDetails.slice(0, MAX_CONTENT_LENGTH)}...(truncated)`;
   }
@@ -161,7 +184,22 @@ const getFinishObservationContent = (
   event: ObservationEvent<FinishObservation>,
 ): string => {
   const { observation } = event;
-  return observation.message || "";
+
+  // Extract text content from the observation
+  const textContent = observation.content
+    .filter((c) => c.type === "text")
+    .map((c) => c.text)
+    .join("\n");
+
+  let content = "";
+
+  if (observation.is_error) {
+    content += `**Error:**\n${textContent}`;
+  } else {
+    content += textContent;
+  }
+
+  return content;
 };
 
 export const getObservationContent = (event: ObservationEvent): string => {
@@ -177,8 +215,9 @@ export const getObservationContent = (event: ObservationEvent): string => {
       );
 
     case "ExecuteBashObservation":
-      return getExecuteBashObservationContent(
-        event as ObservationEvent<ExecuteBashObservation>,
+    case "TerminalObservation":
+      return getTerminalObservationContent(
+        event as ObservationEvent<ExecuteBashObservation | TerminalObservation>,
       );
 
     case "BrowserObservation":

frontend/src/components/v1/chat/event-content-helpers/get-observation-result.ts

@@ -17,6 +17,15 @@ export const getObservationResult = (
       if (exitCode === 0 || metadata.exit_code === 0) return "success"; // Command executed successfully
       return "error"; // Command failed
     }
+    case "TerminalObservation": {
+      const exitCode =
+        observation.exit_code ?? observation.metadata.exit_code ?? null;
+
+      if (observation.timeout || exitCode === -1) return "timeout";
+      if (exitCode === 0) return "success";
+      if (observation.is_error) return "error";
+      return "success";
+    }
     case "FileEditorObservation":
     case "StrReplaceEditorObservation":
       // Check if there's an error

frontend/src/components/v1/chat/event-content-helpers/get-skill-ready-content.ts

@@ -0,0 +1,108 @@
+import { TextContent } from "#/types/v1/core/base/common";
+
+/**
+ * Extracts all text content from an array of TextContent items.
+ */
+const extractAllText = (extendedContent: TextContent[]): string =>
+  extendedContent
+    .filter((c) => c.type === "text")
+    .map((c) => c.text)
+    .join("");
+
+/**
+ * Extracts all <EXTRA_INFO> blocks from the given text.
+ * Returns an array of content strings (without the wrapper tags).
+ */
+const extractExtraInfoBlocks = (text: string): string[] => {
+  const blocks: string[] = [];
+  const blockRegex = /<EXTRA_INFO>([\s\S]*?)<\/EXTRA_INFO>/gi;
+  let match = blockRegex.exec(text);
+
+  while (match !== null) {
+    const blockContent = match[1].trim();
+    if (blockContent.length > 0) {
+      blocks.push(blockContent);
+    }
+    match = blockRegex.exec(text);
+  }
+
+  return blocks;
+};
+
+/**
+ * Formats a single skill with its corresponding content block.
+ */
+const formatSkillWithContent = (
+  skill: string,
+  contentBlock: string | undefined,
+): string => {
+  let formatted = `\n\n- **${skill}**`;
+
+  if (contentBlock && contentBlock.trim().length > 0) {
+    formatted += `\n\n${contentBlock}`;
+  }
+
+  return formatted;
+};
+
+/**
+ * Formats skills paired with their corresponding extended content blocks.
+ */
+const formatSkillKnowledge = (
+  activatedSkills: string[],
+  extraInfoBlocks: string[],
+): string => {
+  let content = `\n\n**Triggered Skill Knowledge:**`;
+
+  activatedSkills.forEach((skill, index) => {
+    const contentBlock =
+      index < extraInfoBlocks.length ? extraInfoBlocks[index] : undefined;
+    content += formatSkillWithContent(skill, contentBlock);
+  });
+
+  return content;
+};
+
+/**
+ * Formats extended content blocks when no skills are present.
+ */
+const formatExtendedContentOnly = (extraInfoBlocks: string[]): string => {
+  let content = `\n\n**Extended Content:**`;
+
+  extraInfoBlocks.forEach((block) => {
+    if (block.trim().length > 0) {
+      content += `\n\n${block}`;
+    }
+  });
+
+  return content;
+};
+
+/**
+ * Formats activated skills and extended content into markdown for display.
+ * Similar to how v0 formats microagent knowledge in recall observations.
+ *
+ * Each skill is paired with its corresponding <EXTRA_INFO> block by index.
+ */
+export const getSkillReadyContent = (
+  activatedSkills: string[],
+  extendedContent: TextContent[],
+): string => {
+  // Extract all <EXTRA_INFO> blocks from extended_content
+  const extraInfoBlocks: string[] = [];
+  if (extendedContent && extendedContent.length > 0) {
+    const allText = extractAllText(extendedContent);
+    extraInfoBlocks.push(...extractExtraInfoBlocks(allText));
+  }
+
+  // Format output based on what we have
+  if (activatedSkills && activatedSkills.length > 0) {
+    return formatSkillKnowledge(activatedSkills, extraInfoBlocks);
+  }
+
+  if (extraInfoBlocks.length > 0) {
+    return formatExtendedContentOnly(extraInfoBlocks);
+  }
+
+  return "";
+};

frontend/src/components/v1/chat/event-message-components/finish-event-message.tsx

@@ -27,13 +27,16 @@ export function FinishEventMessage({
   microagentPRUrl,
   actions,
 }: FinishEventMessageProps) {
+  const eventContent = getEventContent(event);
+  // For FinishAction, details is always a string (getActionContent returns string)
+  const message =
+    typeof eventContent.details === "string"
+      ? eventContent.details
+      : String(eventContent.details);
+
   return (
     <>
-      <ChatMessage
-        type="agent"
-        message={getEventContent(event).details}
-        actions={actions}
-      />
+      <ChatMessage type="agent" message={message} actions={actions} />
       <MicroagentStatusWrapper
         microagentStatus={microagentStatus}
         microagentConversationId={microagentConversationId}