Merge branch 'main' into openhands-fix-issue-6060

Co-authored-by: Engel Nyst <enyst@users.noreply.github.com>

.github/workflows/dummy-agent-test.yml

@@ -36,6 +36,8 @@ jobs:
       - name: Set up Docker Buildx
         id: buildx
         uses: docker/setup-buildx-action@v3
+      - name: Install tmux
+        run: sudo apt-get update && sudo apt-get install -y tmux
       - name: Install poetry via pipx
         run: pipx install poetry
       - name: Set up Python

.github/workflows/eval-runner.yml

@@ -29,6 +29,8 @@ jobs:
       - name: Checkout repository
         uses: actions/checkout@v4
 
+      - name: Install tmux
+        run: sudo apt-get update && sudo apt-get install -y tmux
       - name: Install poetry via pipx
         run: pipx install poetry
 

.github/workflows/ghcr-build.yml

@@ -56,7 +56,7 @@ jobs:
           docker-images: false
           swap-storage: true
       - name: Set up QEMU
-        uses: docker/setup-qemu-action@v3.0.0
+        uses: docker/setup-qemu-action@v3.2.0
         with:
           image: tonistiigi/binfmt:latest
       - name: Login to GHCR
@@ -119,7 +119,7 @@ jobs:
           docker-images: false
           swap-storage: true
       - name: Set up QEMU
-        uses: docker/setup-qemu-action@v3.0.0
+        uses: docker/setup-qemu-action@v3.2.0
         with:
           image: tonistiigi/binfmt:latest
       - name: Login to GHCR

.github/workflows/py-unit-tests-mac.yml

@@ -31,6 +31,8 @@ jobs:
           key: ${{ runner.os }}-poetry-${{ hashFiles('**/poetry.lock') }}
           restore-keys: |
             ${{ runner.os }}-poetry-
+      - name: Install tmux
+        run: brew install tmux
       - name: Install poetry via pipx
         run: pipx install poetry
       - name: Install Python dependencies using Poetry

.github/workflows/py-unit-tests.yml

@@ -30,6 +30,8 @@ jobs:
       - name: Set up Docker Buildx
         id: buildx
         uses: docker/setup-buildx-action@v3
+      - name: Install tmux
+        run: sudo apt-get update && sudo apt-get install -y tmux
       - name: Install poetry via pipx
         run: pipx install poetry
       - name: Set up Python

.openhands/microagents/repo.md

@@ -1,6 +1,7 @@
 ---
 name: repo
-agent: CodeAct
+type: repo
+agent: CodeActAgent
 ---
 This repository contains the code for OpenHands, an automated AI software engineer. It has a Python backend
 (in the `openhands` directory) and React frontend (in the `frontend` directory).

Makefile

@@ -106,7 +106,7 @@ check-poetry:
 	@if command -v poetry > /dev/null; then \
 		POETRY_VERSION=$(shell poetry --version 2>&1 | sed -E 's/Poetry \(version ([0-9]+\.[0-9]+\.[0-9]+)\)/\1/'); \
 		IFS='.' read -r -a POETRY_VERSION_ARRAY <<< "$$POETRY_VERSION"; \
-		if [ $${POETRY_VERSION_ARRAY[0]} -ge 1 ] && [ $${POETRY_VERSION_ARRAY[1]} -ge 8 ]; then \
+		if [ $${POETRY_VERSION_ARRAY[0]} -gt 1 ] || ([ $${POETRY_VERSION_ARRAY[0]} -eq 1 ] && [ $${POETRY_VERSION_ARRAY[1]} -ge 8 ]); then \
 			echo "$(BLUE)$(shell poetry --version) is already installed.$(RESET)"; \
 		else \
 			echo "$(RED)Poetry 1.8 or later is required. You can install poetry by running the following command, then adding Poetry to your PATH:"; \
@@ -190,7 +190,7 @@ build-frontend:
 # Start backend
 start-backend:
 	@echo "$(YELLOW)Starting backend...$(RESET)"
-	@poetry run uvicorn openhands.server.listen:app --host $(BACKEND_HOST) --port $(BACKEND_PORT) --reload --reload-exclude "$(shell pwd)/workspace"
+	@poetry run uvicorn openhands.server.listen:app --host $(BACKEND_HOST) --port $(BACKEND_PORT) --reload --reload-exclude "./workspace"
 
 # Start frontend
 start-frontend:

config.template.toml

@@ -180,6 +180,12 @@ model = "gpt-4o"
 # https://docs.litellm.ai/docs/completion/token_usage
 #custom_tokenizer = ""
 
+# Whether to use native tool calling if supported by the model. Can be true, false, or None by default, which chooses the model's default behavior based on the evaluation.
+# ATTENTION: Based on evaluation, enabling native function calling may lead to worse results
+# in some scenarios. Use with caution and consider testing with your specific use case.
+# https://github.com/All-Hands-AI/OpenHands/pull/4711
+#native_tool_calling = None
+
 [llm.gpt4o-mini]
 api_key = "your-api-key"
 model = "gpt-4o"
@@ -192,6 +198,16 @@ model = "gpt-4o"
 # agent.CodeActAgent
 ##############################################################################
 [agent]
+
+# whether the browsing tool is enabled
+codeact_enable_browsing = true
+
+# whether the LLM draft editor is enabled
+codeact_enable_llm_editor = false
+
+# whether the IPython tool is enabled
+codeact_enable_jupyter = true
+
 # Name of the micro agent to use for this agent
 #micro_agent_name = ""
 
@@ -204,6 +220,12 @@ model = "gpt-4o"
 # LLM config group to use
 #llm_config = 'your-llm-config-group'
 
+# Whether to use microagents at all
+#use_microagents = true
+
+# List of microagents to disable
+#disabled_microagents = []
+
 [agent.RepoExplorerAgent]
 # Example: use a cheaper model for RepoExplorerAgent to reduce cost, especially
 # useful when an agent doesn't demand high quality but uses a lot of tokens

containers/app/Dockerfile

@@ -71,6 +71,7 @@ ENV VIRTUAL_ENV=/app/.venv \
 COPY --chown=openhands:app --chmod=770 --from=backend-builder ${VIRTUAL_ENV} ${VIRTUAL_ENV}
 RUN playwright install --with-deps chromium
 
+COPY --chown=openhands:app --chmod=770 ./microagents ./microagents
 COPY --chown=openhands:app --chmod=770 ./openhands ./openhands
 COPY --chown=openhands:app --chmod=777 ./openhands/runtime/plugins ./openhands/runtime/plugins
 COPY --chown=openhands:app --chmod=770 ./openhands/agenthub ./openhands/agenthub

docs/DOC_STYLE_GUIDE.md

@@ -0,0 +1,48 @@
+# Documentation Style Guide
+
+## General Writing Principles
+
+- **Clarity & Conciseness**: Always prioritize clarity and brevity. Avoid unnecessary jargon or overly complex explanations.
+Keep sentences short and to the point.
+- **Gradual Complexity**: Start with the simplest, most basic setup, and then gradually introduce more advanced
+concepts and configurations.
+
+## Formatting Guidelines
+
+### Headers
+
+Use **Title Case** for the first and second level headers.
+
+Example:
+  - **Basic Usage**
+  - **Advanced Configuration Options**
+
+### Lists
+
+When listing items or options, use bullet points to enhance readability.
+
+Example:
+  - Option A
+  - Option B
+  - Option C
+
+### Procedures
+
+For instructions or processes that need to be followed in a specific order, use numbered steps.
+
+Example:
+  1. Step one: Do this.
+  2. Step two: Complete this action.
+  3. Step three: Verify the result.
+
+### Code Blocks
+
+* Use code blocks for multi-line inputs, outputs, commands and code samples.
+
+Example:
+```bash
+docker run -it \
+    -e THIS=this \
+    -e THAT=that
+    ...
+```

docs/modules/usage/about.md

@@ -4,10 +4,9 @@
 
 Achieving full replication of production-grade applications with LLMs is a complex endeavor. Our strategy involves:
 
-1. **Core Technical Research:** Focusing on foundational research to understand and improve the technical aspects of code generation and handling
-2. **Specialist Abilities:** Enhancing the effectiveness of core components through data curation, training methods, and more
-3. **Task Planning:** Developing capabilities for bug detection, codebase management, and optimization
-4. **Evaluation:** Establishing comprehensive evaluation metrics to better understand and improve our models
+- **Core Technical Research:** Focusing on foundational research to understand and improve the technical aspects of code generation and handling.
+- **Task Planning:** Developing capabilities for bug detection, codebase management, and optimization.
+- **Evaluation:** Establishing comprehensive evaluation metrics to better understand and improve our agents.
 
 ## Default Agent
 
@@ -15,11 +14,14 @@ Our default Agent is currently the [CodeActAgent](agents), which is capable of g
 
 ## Built With
 
-OpenHands is built using a combination of powerful frameworks and libraries, providing a robust foundation for its development. Here are the key technologies used in the project:
+OpenHands is built using a combination of powerful frameworks and libraries, providing a robust foundation for its
+development. Here are the key technologies used in the project:
 
 ![FastAPI](https://img.shields.io/badge/FastAPI-black?style=for-the-badge) ![uvicorn](https://img.shields.io/badge/uvicorn-black?style=for-the-badge) ![LiteLLM](https://img.shields.io/badge/LiteLLM-black?style=for-the-badge) ![Docker](https://img.shields.io/badge/Docker-black?style=for-the-badge) ![Ruff](https://img.shields.io/badge/Ruff-black?style=for-the-badge) ![MyPy](https://img.shields.io/badge/MyPy-black?style=for-the-badge) ![LlamaIndex](https://img.shields.io/badge/LlamaIndex-black?style=for-the-badge) ![React](https://img.shields.io/badge/React-black?style=for-the-badge)
 
-Please note that the selection of these technologies is in progress, and additional technologies may be added or existing ones may be removed as the project evolves. We strive to adopt the most suitable and efficient tools to enhance the capabilities of OpenHands.
+Please note that the selection of these technologies is in progress, and additional technologies may be added or
+existing ones may be removed as the project evolves. We strive to adopt the most suitable and efficient tools to
+enhance the capabilities of OpenHands.
 
 ## License
 

docs/modules/usage/configuration-options.md

@@ -11,7 +11,7 @@ take precedence.
 
 # Table of Contents
 
-1. [Core Configuration](#core-configuration)
+- [Core Configuration](#core-configuration)
    - [API Keys](#api-keys)
    - [Workspace](#workspace)
    - [Debugging and Logging](#debugging-and-logging)
@@ -21,7 +21,7 @@ take precedence.
    - [Task Management](#task-management)
    - [Sandbox Configuration](#sandbox-configuration)
    - [Miscellaneous](#miscellaneous)
-2. [LLM Configuration](#llm-configuration)
+- [LLM Configuration](#llm-configuration)
    - [AWS Credentials](#aws-credentials)
    - [API Configuration](#api-configuration)
    - [Custom LLM Provider](#custom-llm-provider)
@@ -30,20 +30,20 @@ take precedence.
    - [Model Selection](#model-selection)
    - [Retrying](#retrying)
    - [Advanced Options](#advanced-options)
-3. [Agent Configuration](#agent-configuration)
+- [Agent Configuration](#agent-configuration)
    - [Microagent Configuration](#microagent-configuration)
    - [Memory Configuration](#memory-configuration)
    - [LLM Configuration](#llm-configuration-2)
    - [ActionSpace Configuration](#actionspace-configuration)
    - [Microagent Usage](#microagent-usage)
-4. [Sandbox Configuration](#sandbox-configuration-2)
+- [Sandbox Configuration](#sandbox-configuration)
    - [Execution](#execution)
    - [Container Image](#container-image)
    - [Networking](#networking)
    - [Linting and Plugins](#linting-and-plugins)
    - [Dependencies and Environment](#dependencies-and-environment)
    - [Evaluation](#evaluation)
-5. [Security Configuration](#security-configuration)
+- [Security Configuration](#security-configuration)
    - [Confirmation Mode](#confirmation-mode)
    - [Security Analyzer](#security-analyzer)
 

docs/modules/usage/feedback.md

@@ -1,10 +1,14 @@
 # ✅ Providing Feedback
 
-When using OpenHands, you will encounter cases where things work well, and others where they don't. We encourage you to provide feedback when you use OpenHands to help give feedback to the development team, and perhaps more importantly, create an open corpus of coding agent training examples -- Share-OpenHands!
+When using OpenHands, you will encounter cases where things work well, and others where they don't. We encourage you to
+provide feedback when you use OpenHands to help give feedback to the development team, and perhaps more importantly,
+create an open corpus of coding agent training examples -- Share-OpenHands!
 
 ## 📝 How to Provide Feedback
 
-Providing feedback is easy! When you are using OpenHands, you can press the thumbs-up or thumbs-down button at any point during your interaction. You will be prompted to provide your email address (e.g. so we can contact you if we want to ask any follow-up questions), and you can choose whether you want to provide feedback publicly or privately.
+Providing feedback is easy! When you are using OpenHands, you can press the thumbs-up or thumbs-down button at any point
+during your interaction. You will be prompted to provide your email address
+(e.g. so we can contact you if we want to ask any follow-up questions), and you can choose whether you want to provide feedback publicly or privately.
 
 <iframe width="560" height="315" src="https://www.youtube.com/embed/5rFx-StMVV0?si=svo7xzp6LhGK_GXr" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe>
 
@@ -14,8 +18,11 @@ Providing feedback is easy! When you are using OpenHands, you can press the thum
 
 When you submit data, you can submit it either publicly or privately.
 
-* **Public** data will be distributed under the MIT License, like OpenHands itself, and can be used by the community to train and test models. Obviously, feedback that you can make public will be more valuable for the community as a whole, so when you are not dealing with sensitive information, we would encourage you to choose this option!
-* **Private** data will only be shared with the OpenHands team for the purpose of improving OpenHands.
+- **Public** data will be distributed under the MIT License, like OpenHands itself, and can be used by the community to
+train and test models. Obviously, feedback that you can make public will be more valuable for the community as a whole,
+so when you are not dealing with sensitive information, we would encourage you to choose this option!
+- **Private** data will be made available to the OpenHands team for the purpose of improving OpenHands.
+However, a link with a unique ID will still be created that you can share publicly with others.
 
 ### Who collects and stores the data?
 
@@ -27,13 +34,17 @@ The public data will be released when we hit fixed milestones, such as 1,000 pub
 At this time, we will follow the following release process:
 
 1. All people who contributed public feedback will receive an email describing the data release and being given an opportunity to opt out.
-2. The person or people in charge of the data release will perform quality control of the data, removing low-quality feedback, removing email submitter email addresses, and attempting to remove any sensitive information.
+2. The person or people in charge of the data release will perform quality control of the data, removing low-quality feedback,
+removing email submitter email addresses, and attempting to remove any sensitive information.
 3. The data will be released publicly under the MIT license through commonly used sites such as github or Hugging Face.
 
 ### What if I want my data deleted?
 
 For data on the All Hands AI servers, we are happy to delete it at request:
 
-**One Piece of Data:** If you want one piece of data deleted, we will shortly be adding a mechanism to delete pieces of data using the link and password that is displayed on the interface when you submit data.
+**One Piece of Data:** If you want one piece of data deleted, we will shortly be adding a mechanism to delete pieces of
+data using the link and password that is displayed on the interface when you submit data.
 
-**All Data:** If you would like all pieces of your data deleted, or you do not have the ID and password that you received when submitting the data, please contact `contact@all-hands.dev` from the email address that you registered when you originally submitted the data.
+**All Data:** If you would like all pieces of your data deleted, or you do not have the ID and password that you
+received when submitting the data, please contact `contact@all-hands.dev` from the email address that you registered
+when you originally submitted the data.

docs/modules/usage/getting-started.mdx

@@ -44,7 +44,7 @@ For example, we might build a TODO app:
 
 We can keep iterating on the app once the skeleton is there:
 
-> Please allow adding an optional due date to every task
+> Please allow adding an optional due date to every task.
 
 Just like with normal development, it's good to commit and push your code frequently.
 This way you can always revert back to an old state if the agent goes off track.
@@ -59,15 +59,15 @@ OpenHands can also do a great job adding new code to an existing code base.
 
 For example, you can ask OpenHands to add a new GitHub action to your project
 which lints your code. OpenHands may take a peek at your codebase to see what language
-it should use, but then it can just drop a new file into `./github/workflows/lint.yml`
+it should use and then drop a new file into `./github/workflows/lint.yml`.
 
-> Please add a GitHub action that lints the code in this repository
+> Please add a GitHub action that lints the code in this repository.
 
 Some tasks might require a bit more context. While OpenHands can use `ls` and `grep`
 to search through your codebase, providing context up front allows it to move faster,
 and more accurately. And it'll cost you fewer tokens!
 
-> Please modify ./backend/api/routes.js to add a new route that returns a list of all tasks
+> Please modify ./backend/api/routes.js to add a new route that returns a list of all tasks.
 
 > Please add a new React component that displays a list of Widgets to the ./frontend/components
 > directory. It should use the existing Widget component.
@@ -78,15 +78,15 @@ OpenHands does great at refactoring existing code, especially in small chunks.
 You probably don't want to try rearchitecting your whole codebase, but breaking up
 long files and functions, renaming variables, etc. tend to work very well.
 
-> Please rename all the single-letter variables in ./app.go
+> Please rename all the single-letter variables in ./app.go.
 
-> Please break the function `build_and_deploy_widgets` into two functions, `build_widgets` and `deploy_widgets` in widget.php
+> Please break the function `build_and_deploy_widgets` into two functions, `build_widgets` and `deploy_widgets` in widget.php.
 
-> Please break ./api/routes.js into separate files for each route
+> Please break ./api/routes.js into separate files for each route.
 
 ## Bug Fixes
 
-OpenHands can also help you track down and fix bugs in your code. But, as any
+OpenHands can also help you track down and fix bugs in your code. But as any
 developer knows, bug fixing can be extremely tricky, and often OpenHands will need more context.
 It helps if you've diagnosed the bug, but want OpenHands to figure out the logic.
 
@@ -94,18 +94,18 @@ It helps if you've diagnosed the bug, but want OpenHands to figure out the logic
 
 > The `search_widgets` function in ./app.py is doing a case-sensitive search. Please make it case-insensitive.
 
-It often helps to do test-driven development when bugfixing with an agent.
+It often helps to do test-driven development when bug fixing with an agent.
 You can ask the agent to write a new test, and then iterate until it fixes the bug:
 
 > The `hello` function crashes on the empty string. Please write a test that reproduces this bug, then fix the code so it passes.
 
 ## More
 
-OpenHands is capable of helping out on just about any coding task. But it takes some practice
+OpenHands is capable of helping out on just about any coding task but it takes some practice
 to get the most out of it. Remember to:
-* Keep your tasks small
-* Be as specific as possible
-* Provide as much context as possible
-* Commit and push frequently
+* Keep your tasks small.
+* Be as specific as possible.
+* Provide as much context as possible.
+* Commit and push frequently.
 
 See [Prompting Best Practices](./prompting/prompting-best-practices) for more tips on how to get the most out of OpenHands.

docs/modules/usage/how-to/cli-mode.md

@@ -26,9 +26,9 @@ To run OpenHands in CLI mode with Docker:
 
 1. Set the following environmental variables in your terminal:
 
-* `WORKSPACE_BASE` to the directory you want OpenHands to edit (Ex: `export WORKSPACE_BASE=$(pwd)/workspace`).
-* `LLM_MODEL` to the model to use (Ex: `export LLM_MODEL="anthropic/claude-3-5-sonnet-20241022"`).
-* `LLM_API_KEY` to the API key (Ex: `export LLM_API_KEY="sk_test_12345"`).
+- `WORKSPACE_BASE` to the directory you want OpenHands to edit (Ex: `export WORKSPACE_BASE=$(pwd)/workspace`).
+- `LLM_MODEL` to the model to use (Ex: `export LLM_MODEL="anthropic/claude-3-5-sonnet-20241022"`).
+- `LLM_API_KEY` to the API key (Ex: `export LLM_API_KEY="sk_test_12345"`).
 
 2. Run the following Docker command:
 

docs/modules/usage/how-to/custom-sandbox-guide.md

@@ -9,8 +9,8 @@ as python and Node.js but may need other software installed by default.
 
 You have two options for customization:
 
-1. Use an existing image with the required software.
-2. Create your own custom Docker image.
+- Use an existing image with the required software.
+- Create your own custom Docker image.
 
 If you choose the first option, you can skip the `Create Your Docker Image` section.
 
@@ -58,7 +58,3 @@ sandbox_base_container_image="custom-image"
 ### Run
 
 Run OpenHands by running ```make run``` in the top level directory.
-
-## Technical Explanation
-
-Please refer to [custom docker image section of the runtime documentation](https://docs.all-hands.dev/modules/usage/architecture/runtime#advanced-how-openhands-builds-and-maintains-od-runtime-images) for more details.

docs/modules/usage/how-to/github-action.md

@@ -21,10 +21,10 @@ the [README for the OpenHands Resolver](https://github.com/All-Hands-AI/OpenHand
 ### Iterative resolution
 
 1. Create an issue in the repository.
-2. Add the `fix-me` label to the issue, or leave a comment starting with `@openhands-agent`
-3. Review the attempt to resolve the issue by checking the pull request
-4. Follow up with feedback through general comments, review comments, or inline thread comments
-5. Add the `fix-me` label to the pull request, or address a specific comment by starting with `@openhands-agent`
+2. Add the `fix-me` label to the issue, or leave a comment starting with `@openhands-agent`.
+3. Review the attempt to resolve the issue by checking the pull request.
+4. Follow up with feedback through general comments, review comments, or inline thread comments.
+5. Add the `fix-me` label to the pull request, or address a specific comment by starting with `@openhands-agent`.
 
 ### Label versus Macro
 

docs/modules/usage/how-to/gui-mode.md

@@ -2,12 +2,12 @@
 
 ## Introduction
 
-OpenHands provides a user-friendly Graphical User Interface (GUI) mode for interacting with the AI assistant. This mode offers an intuitive way to set up the environment, manage settings, and communicate with the AI.
+OpenHands provides a user-friendly Graphical User Interface (GUI) mode for interacting with the AI assistant.
+This mode offers an intuitive way to set up the environment, manage settings, and communicate with the AI.
 
 ## Installation and Setup
 
 1. Follow the instructions in the [Installation](../installation) guide to install OpenHands.
-
 2. After running the command, access OpenHands at [http://localhost:3000](http://localhost:3000).
 
 ## Interacting with the GUI
@@ -23,39 +23,39 @@ OpenHands provides a user-friendly Graphical User Interface (GUI) mode for inter
 
 OpenHands automatically exports a `GITHUB_TOKEN` to the shell environment if it is available. This can happen in two ways:
 
-1. **Locally (OSS)**: The user directly inputs their GitHub token
-2. **Online (SaaS)**: The token is obtained through GitHub OAuth authentication
+- **Locally (OSS)**: The user directly inputs their GitHub token.
+- **Online (SaaS)**: The token is obtained through GitHub OAuth authentication.
 
 #### Setting Up a Local GitHub Token
 
 1. **Generate a Personal Access Token (PAT)**:
-   - Go to GitHub Settings > Developer Settings > Personal Access Tokens > Tokens (classic)
-   - Click "Generate new token (classic)"
+   - Go to GitHub Settings > Developer Settings > Personal Access Tokens > Tokens (classic).
+   - Click "Generate new token (classic)".
    - Required scopes:
      - `repo` (Full control of private repositories)
      - `workflow` (Update GitHub Action workflows)
      - `read:org` (Read organization data)
 
 2. **Enter Token in OpenHands**:
-   - Click the Settings button (gear icon) in the top right
-   - Navigate to the "GitHub" section
-   - Paste your token in the "GitHub Token" field
-   - Click "Save" to apply the changes
+   - Click the Settings button (gear icon) in the top right.
+   - Navigate to the "GitHub" section.
+   - Paste your token in the "GitHub Token" field.
+   - Click "Save" to apply the changes.
 
 #### Organizational Token Policies
 
 If you're working with organizational repositories, additional setup may be required:
 
 1. **Check Organization Requirements**:
-   - Organization admins may enforce specific token policies
-   - Some organizations require tokens to be created with SSO enabled
-   - Review your organization's [token policy settings](https://docs.github.com/en/organizations/managing-programmatic-access-to-your-organization/setting-a-personal-access-token-policy-for-your-organization)
+   - Organization admins may enforce specific token policies.
+   - Some organizations require tokens to be created with SSO enabled.
+   - Review your organization's [token policy settings](https://docs.github.com/en/organizations/managing-programmatic-access-to-your-organization/setting-a-personal-access-token-policy-for-your-organization).
 
 2. **Verify Organization Access**:
-   - Go to your token settings on GitHub
-   - Look for the organization under "Organization access"
-   - If required, click "Enable SSO" next to your organization
-   - Complete the SSO authorization process
+   - Go to your token settings on GitHub.
+   - Look for the organization under "Organization access".
+   - If required, click "Enable SSO" next to your organization.
+   - Complete the SSO authorization process.
 
 #### OAuth Authentication (Online Mode)
 
@@ -67,31 +67,31 @@ When using OpenHands in online mode, the GitHub OAuth flow:
    - Organization read access
 
 2. Authentication steps:
-   - Click "Sign in with GitHub" when prompted
-   - Review the requested permissions
-   - Authorize OpenHands to access your GitHub account
-   - If using an organization, authorize organization access if prompted
+   - Click "Sign in with GitHub" when prompted.
+   - Review the requested permissions.
+   - Authorize OpenHands to access your GitHub account.
+   - If using an organization, authorize organization access if prompted.
 
 #### Troubleshooting
 
 Common issues and solutions:
 
 1. **Token Not Recognized**:
-   - Ensure the token is properly saved in settings
-   - Check that the token hasn't expired
-   - Verify the token has the required scopes
-   - Try regenerating the token
+   - Ensure the token is properly saved in settings.
+   - Check that the token hasn't expired.
+   - Verify the token has the required scopes.
+   - Try regenerating the token.
 
 2. **Organization Access Denied**:
-   - Check if SSO is required but not enabled
-   - Verify organization membership
-   - Contact organization admin if token policies are blocking access
+   - Check if SSO is required but not enabled.
+   - Verify organization membership.
+   - Contact organization admin if token policies are blocking access.
 
 3. **Verifying Token Works**:
-   - The app will show a green checkmark if the token is valid
-   - Try accessing a repository to confirm permissions
-   - Check the browser console for any error messages
-   - Use the "Test Connection" button in settings if available
+   - The app will show a green checkmark if the token is valid.
+   - Try accessing a repository to confirm permissions.
+   - Check the browser console for any error messages.
+   - Use the "Test Connection" button in settings if available.
 
 ### Advanced Settings
 
@@ -103,11 +103,11 @@ Common issues and solutions:
 
 The main interface consists of several key components:
 
-1. **Chat Window**: The central area where you can view the conversation history with the AI assistant.
-2. **Input Box**: Located at the bottom of the screen, use this to type your messages or commands to the AI.
-3. **Send Button**: Click this to send your message to the AI.
-4. **Settings Button**: A gear icon that opens the settings modal, allowing you to adjust your configuration at any time.
-5. **Workspace Panel**: Displays the files and folders in your workspace, allowing you to navigate and view files, or the agent's past commands or web browsing history.
+- **Chat Window**: The central area where you can view the conversation history with the AI assistant.
+- **Input Box**: Located at the bottom of the screen, use this to type your messages or commands to the AI.
+- **Send Button**: Click this to send your message to the AI.
+- **Settings Button**: A gear icon that opens the settings modal, allowing you to adjust your configuration at any time.
+- **Workspace Panel**: Displays the files and folders in your workspace, allowing you to navigate and view files, or the agent's past commands or web browsing history.
 
 ### Interacting with the AI
 
@@ -118,8 +118,9 @@ The main interface consists of several key components:
 
 ## Tips for Effective Use
 
-1. Be specific in your requests to get the most accurate and helpful responses, as described in the [prompting best practices](../prompting/prompting-best-practices).
-2. Use the workspace panel to explore your project structure.
-3. Use one of the recommended models, as described in the [LLMs section](usage/llms/llms.md).
+- Be specific in your requests to get the most accurate and helpful responses, as described in the [prompting best practices](../prompting/prompting-best-practices).
+- Use the workspace panel to explore your project structure.
+- Use one of the recommended models, as described in the [LLMs section](usage/llms/llms.md).
 
-Remember, the GUI mode of OpenHands is designed to make your interaction with the AI assistant as smooth and intuitive as possible. Don't hesitate to explore its features to maximize your productivity.
+Remember, the GUI mode of OpenHands is designed to make your interaction with the AI assistant as smooth and intuitive
+as possible. Don't hesitate to explore its features to maximize your productivity.

docs/modules/usage/how-to/headless-mode.md

@@ -23,9 +23,9 @@ To run OpenHands in Headless mode with Docker:
 
 1. Set the following environmental variables in your terminal:
 
-* `WORKSPACE_BASE` to the directory you want OpenHands to edit (Ex: `export WORKSPACE_BASE=$(pwd)/workspace`).
-* `LLM_MODEL` to the model to use (Ex: `export LLM_MODEL="anthropic/claude-3-5-sonnet-20241022"`).
-* `LLM_API_KEY` to the API key (Ex: `export LLM_API_KEY="sk_test_12345"`).
+- `WORKSPACE_BASE` to the directory you want OpenHands to edit (Ex: `export WORKSPACE_BASE=$(pwd)/workspace`).
+- `LLM_MODEL` to the model to use (Ex: `export LLM_MODEL="anthropic/claude-3-5-sonnet-20241022"`).
+- `LLM_API_KEY` to the API key (Ex: `export LLM_API_KEY="sk_test_12345"`).
 
 2. Run the following Docker command:
 
@@ -53,4 +53,4 @@ To view all available configuration options for headless mode, run the Python co
 
 ### Additional Logs
 
-For the headless mode to log all the agent actions, in your terminal run: `export LOG_ALL_EVENTS=true`
+For the headless mode to log all the agent actions, in the terminal run: `export LOG_ALL_EVENTS=true`

docs/modules/usage/how-to/persist-session-data.md

@@ -1,6 +1,6 @@
 # Persisting Session Data
 
-Using the standard installation, the session data is stored in memory. Currently, if OpenHands' service is restarted,
+Using the standard Development Workflow, the session data is stored in memory. Currently, if OpenHands' service is restarted,
 previous sessions become invalid (a new secret is generated) and thus not recoverable.
 
 ## How to Persist Session Data

docs/modules/usage/installation.mdx

@@ -2,9 +2,9 @@
 
 ## System Requirements
 
-* Docker version 26.0.0+ or Docker Desktop 4.31.0+.
-* You must be using Linux or Mac OS.
-  * If you are on Windows, you must use [WSL](https://learn.microsoft.com/en-us/windows/wsl/install).
+- Docker version 26.0.0+ or Docker Desktop 4.31.0+.
+- You must be using Linux or Mac OS.
+  - If you are on Windows, you must use [WSL](https://learn.microsoft.com/en-us/windows/wsl/install).
 
 ## Start the app
 
@@ -33,8 +33,6 @@ or run it on tagged issues with [a github action](https://docs.all-hands.dev/mod
 
 ## Setup
 
-After running the command above, you'll find OpenHands running at [http://localhost:3000](http://localhost:3000).
-
 Upon launching OpenHands, you'll see a settings modal. You **must** select an `LLM Provider` and `LLM Model` and enter a corresponding `API Key`.
 These can be changed at any time by selecting the `Settings` button (gear icon) in the UI.
 

docs/modules/usage/llms/azure-llms.md

@@ -18,17 +18,18 @@ docker run -it --pull=always \
     ...
 ```
 
-Then set the following in the OpenHands UI through the Settings:
+Then in the OpenHands UI Settings:
 
 :::note
 You will need your ChatGPT deployment name which can be found on the deployments page in Azure. This is referenced as
 <deployment-name> below.
 :::
 
-* Enable `Advanced Options`
-* `Custom Model` to azure/<deployment-name>
-* `Base URL` to your Azure API Base URL (e.g. `https://example-endpoint.openai.azure.com`)
-* `API Key` to your Azure API key
+1. Enable `Advanced Options`
+2. Set the following:
+   - `Custom Model` to azure/<deployment-name>
+   - `Base URL` to your Azure API Base URL (e.g. `https://example-endpoint.openai.azure.com`)
+   - `API Key` to your Azure API key
 
 ## Embeddings
 

docs/modules/usage/llms/google-llms.md

@@ -8,10 +8,10 @@ OpenHands uses LiteLLM to make calls to Google's chat models. You can find their
 ## Gemini - Google AI Studio Configs
 
 When running OpenHands, you'll need to set the following in the OpenHands UI through the Settings:
-* `LLM Provider` to `Gemini`
-* `LLM Model` to the model you will be using.
+- `LLM Provider` to `Gemini`
+- `LLM Model` to the model you will be using.
 If the model is not in the list, toggle `Advanced Options`, and enter it in `Custom Model` (e.g. gemini/<model-name> like `gemini/gemini-1.5-pro`).
-* `API Key` to your Gemini API key
+- `API Key` to your Gemini API key
 
 ## VertexAI - Google Cloud Platform Configs
 
@@ -25,6 +25,6 @@ VERTEXAI_LOCATION="<your-gcp-location>"
 ```
 
 Then set the following in the OpenHands UI through the Settings:
-* `LLM Provider` to `VertexAI`
-* `LLM Model` to the model you will be using.
+- `LLM Provider` to `VertexAI`
+- `LLM Model` to the model you will be using.
 If the model is not in the list, toggle `Advanced Options`, and enter it in `Custom Model` (e.g. vertex_ai/&lt;model-name&gt;).

docs/modules/usage/llms/groq.md

@@ -5,19 +5,20 @@ OpenHands uses LiteLLM to make calls to chat models on Groq. You can find their
 ## Configuration
 
 When running OpenHands, you'll need to set the following in the OpenHands UI through the Settings:
-* `LLM Provider` to `Groq`
-* `LLM Model` to the model you will be using. [Visit here to see the list of
+- `LLM Provider` to `Groq`
+- `LLM Model` to the model you will be using. [Visit here to see the list of
 models that Groq hosts](https://console.groq.com/docs/models). If the model is not in the list, toggle
 `Advanced Options`, and enter it in `Custom Model` (e.g. groq/<model-name> like `groq/llama3-70b-8192`).
-* `API key` to your Groq API key. To find or create your Groq API Key, [see here](https://console.groq.com/keys).
+- `API key` to your Groq API key. To find or create your Groq API Key, [see here](https://console.groq.com/keys).
 
 
 
 ## Using Groq as an OpenAI-Compatible Endpoint
 
 The Groq endpoint for chat completion is [mostly OpenAI-compatible](https://console.groq.com/docs/openai). Therefore, you can access Groq models as you
-would access any OpenAI-compatible endpoint. You can set the following in the OpenHands UI through the Settings:
-* Enable `Advanced Options`
-* `Custom Model` to the prefix `openai/` + the model you will be using (e.g. `openai/llama3-70b-8192`)
-* `Base URL` to `https://api.groq.com/openai/v1`
-* `API Key` to your Groq API key
+would access any OpenAI-compatible endpoint. In the OpenHands UI through the Settings:
+1. Enable `Advanced Options`
+2. Set the following:
+   - `Custom Model` to the prefix `openai/` + the model you will be using (e.g. `openai/llama3-70b-8192`)
+   - `Base URL` to `https://api.groq.com/openai/v1`
+   - `API Key` to your Groq API key

docs/modules/usage/llms/llms.md

@@ -4,7 +4,9 @@ OpenHands can connect to any LLM supported by LiteLLM. However, it requires a po
 
 ## Model Recommendations
 
-Based on our evaluations of language models for coding tasks (using the SWE-bench dataset), we can provide some recommendations for model selection. Some analyses can be found in [this blog article comparing LLMs](https://www.all-hands.dev/blog/evaluation-of-llms-as-coding-agents-on-swe-bench-at-30x-speed) and [this blog article with some more recent results](https://www.all-hands.dev/blog/openhands-codeact-21-an-open-state-of-the-art-software-development-agent).
+Based on our evaluations of language models for coding tasks (using the SWE-bench dataset), we can provide some
+recommendations for model selection. Some analyses can be found in [this blog article comparing LLMs](https://www.all-hands.dev/blog/evaluation-of-llms-as-coding-agents-on-swe-bench-at-30x-speed) and
+[this blog article with some more recent results](https://www.all-hands.dev/blog/openhands-codeact-21-an-open-state-of-the-art-software-development-agent).
 
 When choosing a model, consider both the quality of outputs and the associated costs. Here's a summary of the findings:
 
@@ -69,9 +71,11 @@ We have a few guides for running OpenHands with specific model providers:
 
 ### API retries and rate limits
 
-LLM providers typically have rate limits, sometimes very low, and may require retries. OpenHands will automatically retry requests if it receives a Rate Limit Error (429 error code), API connection error, or other transient errors.
+LLM providers typically have rate limits, sometimes very low, and may require retries. OpenHands will automatically
+retry requests if it receives a Rate Limit Error (429 error code), API connection error, or other transient errors.
 
-You can customize these options as you need for the provider you're using. Check their documentation, and set the following environment variables to control the number of retries and the time between retries:
+You can customize these options as you need for the provider you're using. Check their documentation, and set the
+following environment variables to control the number of retries and the time between retries:
 
 - `LLM_NUM_RETRIES` (Default of 8)
 - `LLM_RETRY_MIN_WAIT` (Default of 15 seconds)

docs/modules/usage/llms/openai-llms.md

@@ -17,8 +17,9 @@ Just as for OpenAI Chat completions, we use LiteLLM for OpenAI-compatible endpoi
 
 ## Using an OpenAI Proxy
 
-If you're using an OpenAI proxy, you'll need to set the following in the OpenHands UI through the Settings:
-* Enable `Advanced Options`
-* `Custom Model` to openai/<model-name> (e.g. `openai/gpt-4o` or openai/<proxy-prefix>/<model-name>)
-* `Base URL` to the URL of your OpenAI proxy
-* `API Key` to your OpenAI API key
+If you're using an OpenAI proxy, in the OpenHands UI through the Settings:
+1. Enable `Advanced Options`
+2. Set the following:
+   - `Custom Model` to openai/<model-name> (e.g. `openai/gpt-4o` or openai/<proxy-prefix>/<model-name>)
+   - `Base URL` to the URL of your OpenAI proxy
+   - `API Key` to your OpenAI API key

docs/modules/usage/prompting/customization.md

@@ -9,11 +9,11 @@ You can customize OpenHands' behavior for your repository by creating a `.openha
 be given to the agent every time it works with this repository.
 
 We suggest including the following information:
-1. **Repository Overview**: A brief description of your project's purpose and architecture
-2. **Directory Structure**: Key directories and their purposes
-3. **Development Guidelines**: Project-specific coding standards and practices
-4. **Testing Requirements**: How to run tests and what types of tests are required
-5. **Setup Instructions**: Steps needed to build and run the project
+- **Repository Overview**: A brief description of your project's purpose and architecture.
+- **Directory Structure**: Key directories and their purposes.
+- **Development Guidelines**: Project-specific coding standards and practices.
+- **Testing Requirements**: How to run tests and what types of tests are required.
+- **Setup Instructions**: Steps needed to build and run the project.
 
 ### Example Repository Configuration
 Example `.openhands/microagents/repo.md` file:
@@ -39,11 +39,11 @@ Guidelines:
 
 ### Customizing Prompts
 
-When working with a customized repository:
+When working with a repository:
 
-1. **Reference Project Standards**: Mention specific coding standards or patterns used in your project
-2. **Include Context**: Reference relevant documentation or existing implementations
-3. **Specify Testing Requirements**: Include project-specific testing requirements in your prompts
+- **Reference Project Standards**: Mention specific coding standards or patterns used in your project.
+- **Include Context**: Reference relevant documentation or existing implementations.
+- **Specify Testing Requirements**: Include project-specific testing requirements in your prompts.
 
 Example customized prompt:
 ```
@@ -54,14 +54,14 @@ The component should use our shared styling from src/styles/components.
 
 ### Best Practices for Repository Customization
 
-1. **Keep Instructions Updated**: Regularly update your `.openhands` directory as your project evolves
-2. **Be Specific**: Include specific paths, patterns, and requirements unique to your project
-3. **Document Dependencies**: List all tools and dependencies required for development
-4. **Include Examples**: Provide examples of good code patterns from your project
-5. **Specify Conventions**: Document naming conventions, file organization, and code style preferences
+- **Keep Instructions Updated**: Regularly update your `.openhands` directory as your project evolves.
+- **Be Specific**: Include specific paths, patterns, and requirements unique to your project.
+- **Document Dependencies**: List all tools and dependencies required for development.
+- **Include Examples**: Provide examples of good code patterns from your project.
+- **Specify Conventions**: Document naming conventions, file organization, and code style preferences.
 
 By customizing OpenHands for your repository, you'll get more accurate and consistent results that align with your project's standards and requirements.
 
 ## Other Microagents
 You can create other instructions in the `.openhands/microagents/` directory
-that will be sent to the agent if a particular keyword is found, like `test`, `frontend`, or `migration`. See [Microagents](microagents.md) for more information.
+that will be sent to the agent if a particular keyword is found, like `test`, `frontend`, or `migration`. See [Micro-Agents](microagents.md) for more information.

docs/modules/usage/prompting/microagents.md

@@ -6,10 +6,10 @@ OpenHands uses specialized micro-agents to handle specific tasks and contexts ef
 
 Micro-agents are defined in markdown files under the `openhands/agenthub/codeact_agent/micro/` directory. Each micro-agent is configured with:
 
-- A unique name
-- The agent type (typically CodeActAgent)
-- Trigger keywords that activate the agent
-- Specific instructions and capabilities
+- A unique name.
+- The agent type (typically CodeActAgent).
+- Trigger keywords that activate the agent.
+- Specific instructions and capabilities.
 
 ## Available Micro-Agents
 
@@ -18,10 +18,10 @@ Micro-agents are defined in markdown files under the `openhands/agenthub/codeact
 **Triggers**: `github`, `git`
 
 The GitHub agent specializes in GitHub API interactions and repository management. It:
-- Has access to a `GITHUB_TOKEN` for API authentication
-- Follows strict guidelines for repository interactions
-- Handles branch management and pull requests
-- Uses the GitHub API instead of web browser interactions
+- Has access to a `GITHUB_TOKEN` for API authentication.
+- Follows strict guidelines for repository interactions.
+- Handles branch management and pull requests.
+- Uses the GitHub API instead of web browser interactions.
 
 Key features:
 - Branch protection (prevents direct pushes to main/master)
@@ -34,13 +34,14 @@ Key features:
 **Triggers**: `npm`
 
 Specializes in handling npm package management with specific focus on:
-- Non-interactive shell operations
-- Automated confirmation handling using Unix 'yes' command
-- Package installation automation
+- Non-interactive shell operations.
+- Automated confirmation handling using Unix 'yes' command.
+- Package installation automation.
 
 ### Custom Micro-Agents
 
-You can create your own micro-agents by adding new markdown files to the micro-agents directory. Each file should follow this structure:
+You can create your own micro-agents by adding new markdown files to the micro-agents directory.
+Each file should follow this structure:
 
 ```markdown
 ---
@@ -57,19 +58,18 @@ Instructions and capabilities for the micro-agent...
 ## Best Practices
 
 When working with micro-agents:
-
-1. **Use Appropriate Triggers**: Ensure your commands include the relevant trigger words to activate the correct micro-agent
-2. **Follow Agent Guidelines**: Each agent has specific instructions and limitations - respect these for optimal results
-3. **API-First Approach**: When available, use API endpoints rather than web interfaces
-4. **Automation Friendly**: Design commands that work well in non-interactive environments
+- **Use Appropriate Triggers**: Ensure your commands include the relevant trigger words to activate the correct micro-agent.
+- **Follow Agent Guidelines**: Each agent has specific instructions and limitations. Respect these for optimal results.
+- **API-First Approach**: When available, use API endpoints rather than web interfaces.
+- **Automation Friendly**: Design commands that work well in non-interactive environments.
 
 ## Integration
 
 Micro-agents are automatically integrated into OpenHands' workflow. They:
-- Monitor incoming commands for their trigger words
-- Activate when relevant triggers are detected
-- Apply their specialized knowledge and capabilities
-- Follow their specific guidelines and restrictions
+- Monitor incoming commands for their trigger words.
+- Activate when relevant triggers are detected.
+- Apply their specialized knowledge and capabilities.
+- Follow their specific guidelines and restrictions.
 
 ## Example Usage
 
@@ -105,7 +105,7 @@ Create a new markdown file in `openhands/agenthub/codeact_agent/micro/` with a d
 
 Your micro-agent file must include:
 
-1. **Front Matter**: YAML metadata at the start of the file:
+- **Front Matter**: YAML metadata at the start of the file:
 ```markdown
 ---
 name: your_agent_name
@@ -116,7 +116,7 @@ triggers:
 ---
 ```
 
-2. **Instructions**: Clear, specific guidelines for the agent's behavior:
+- **Instructions**: Clear, specific guidelines for the agent's behavior:
 ```markdown
 You are responsible for [specific task/domain].
 
@@ -135,19 +135,19 @@ Examples of usage:
 
 ### 4. Best Practices for Micro-Agent Development
 
-1. **Clear Scope**: Keep the agent focused on a specific domain or task
-2. **Explicit Instructions**: Provide clear, unambiguous guidelines
-3. **Useful Examples**: Include practical examples of common use cases
-4. **Safety First**: Include necessary warnings and constraints
-5. **Integration Awareness**: Consider how the agent interacts with other components
+- **Clear Scope**: Keep the agent focused on a specific domain or task.
+- **Explicit Instructions**: Provide clear, unambiguous guidelines.
+- **Useful Examples**: Include practical examples of common use cases.
+- **Safety First**: Include necessary warnings and constraints.
+- **Integration Awareness**: Consider how the agent interacts with other components.
 
 ### 5. Testing Your Micro-Agent
 
 Before submitting:
-1. Test the agent with various prompts
-2. Verify trigger words activate the agent correctly
-3. Ensure instructions are clear and comprehensive
-4. Check for potential conflicts with existing agents
+- Test the agent with various prompts.
+- Verify trigger words activate the agent correctly.
+- Ensure instructions are clear and comprehensive.
+- Check for potential conflicts with existing agents.
 
 ### 6. Example Implementation
 
@@ -199,11 +199,12 @@ Remember to:
 
 ### 7. Submission Process
 
-1. Create your micro-agent file in the correct directory
-2. Test thoroughly
+1. Create your micro-agent file in the correct directory.
+2. Test thoroughly.
 3. Submit a pull request with:
-   - The new micro-agent file
-   - Updated documentation if needed
-   - Description of the agent's purpose and capabilities
+   - The new micro-agent file.
+   - Updated documentation if needed.
+   - Description of the agent's purpose and capabilities.
 
-Remember that micro-agents are a powerful way to extend OpenHands' capabilities in specific domains. Well-designed agents can significantly improve the system's ability to handle specialized tasks.
+Remember that micro-agents are a powerful way to extend OpenHands' capabilities in specific domains. Well-designed
+agents can significantly improve the system's ability to handle specialized tasks.

docs/modules/usage/prompting/prompting-best-practices.md

@@ -6,35 +6,31 @@ When working with OpenHands AI software developer, it's crucial to provide clear
 
 Good prompts are:
 
-1. **Concrete**: They explain exactly what functionality should be added or what error needs to be fixed.
-2. **Location-specific**: If known, they explain the locations in the code base that should be modified.
-3. **Appropriately scoped**: They should be the size of a single feature, typically not exceeding 100 lines of code.
+- **Concrete**: They explain exactly what functionality should be added or what error needs to be fixed.
+- **Location-specific**: If known, they explain the locations in the code base that should be modified.
+- **Appropriately scoped**: They should be the size of a single feature, typically not exceeding 100 lines of code.
 
 ## Examples
 
 ### Good Prompt Examples
 
-1. "Add a function `calculate_average` in `utils/math_operations.py` that takes a list of numbers as input and returns their average."
-
-2. "Fix the TypeError in `frontend/src/components/UserProfile.tsx` occurring on line 42. The error suggests we're trying to access a property of undefined."
-
-3. "Implement input validation for the email field in the registration form. Update `frontend/src/components/RegistrationForm.tsx` to check if the email is in a valid format before submission."
+- "Add a function `calculate_average` in `utils/math_operations.py` that takes a list of numbers as input and returns their average."
+- "Fix the TypeError in `frontend/src/components/UserProfile.tsx` occurring on line 42. The error suggests we're trying to access a property of undefined."
+- "Implement input validation for the email field in the registration form. Update `frontend/src/components/RegistrationForm.tsx` to check if the email is in a valid format before submission."
 
 ### Bad Prompt Examples
 
-1. "Make the code better." (Too vague, not concrete)
-
-2. "Rewrite the entire backend to use a different framework." (Not appropriately scoped)
-
-3. "There's a bug somewhere in the user authentication. Can you find and fix it?" (Lacks specificity and location information)
+- "Make the code better." (Too vague, not concrete)
+- "Rewrite the entire backend to use a different framework." (Not appropriately scoped)
+- "There's a bug somewhere in the user authentication. Can you find and fix it?" (Lacks specificity and location information)
 
 ## Tips for Effective Prompting
 
-1. Be as specific as possible about the desired outcome or the problem to be solved.
-2. Provide context, including relevant file paths and line numbers if available.
-3. Break down large tasks into smaller, manageable prompts.
-4. Include any relevant error messages or logs.
-5. Specify the programming language or framework if it's not obvious from the context.
+- Be as specific as possible about the desired outcome or the problem to be solved.
+- Provide context, including relevant file paths and line numbers if available.
+- Break down large tasks into smaller, manageable prompts.
+- Include any relevant error messages or logs.
+- Specify the programming language or framework if it's not obvious from the context.
 
 Remember, the more precise and informative your prompt is, the better the AI can assist you in developing or modifying the OpenHands software.
 

docs/modules/usage/runtimes.md

@@ -26,30 +26,29 @@ that contains our Runtime server, as well as some basic utilities for Python and
 You can also [build your own runtime image](how-to/custom-sandbox-guide).
 
 ### Connecting to Your filesystem
-One useful feature here is the ability to connect to your local filesystem.
+One useful feature here is the ability to connect to your local filesystem. To mount your filesystem into the runtime:
+1. Set `WORKSPACE_BASE`:
 
-To mount your filesystem into the runtime, first set WORKSPACE_BASE:
-```bash
-export WORKSPACE_BASE=/path/to/your/code
+    ```bash
+    export WORKSPACE_BASE=/path/to/your/code
 
-# Linux and Mac Example
-# export WORKSPACE_BASE=$HOME/OpenHands
-# Will set $WORKSPACE_BASE to /home/<username>/OpenHands
-#
-# WSL on Windows Example
-# export WORKSPACE_BASE=/mnt/c/dev/OpenHands
-# Will set $WORKSPACE_BASE to C:\dev\OpenHands
-```
+    # Linux and Mac Example
+    # export WORKSPACE_BASE=$HOME/OpenHands
+    # Will set $WORKSPACE_BASE to /home/<username>/OpenHands
+    #
+    # WSL on Windows Example
+    # export WORKSPACE_BASE=/mnt/c/dev/OpenHands
+    # Will set $WORKSPACE_BASE to C:\dev\OpenHands
+    ```
+2. Add the following options to the `docker run` command:
 
-then add the following options to the `docker run` command:
-
-```bash
-docker run # ...
-    -e SANDBOX_USER_ID=$(id -u) \
-    -e WORKSPACE_MOUNT_PATH=$WORKSPACE_BASE \
-    -v $WORKSPACE_BASE:/opt/workspace_base \
-    # ...
-```
+    ```bash
+    docker run # ...
+        -e SANDBOX_USER_ID=$(id -u) \
+        -e WORKSPACE_MOUNT_PATH=$WORKSPACE_BASE \
+        -v $WORKSPACE_BASE:/opt/workspace_base \
+        # ...
+    ```
 
 Be careful! There's nothing stopping the OpenHands agent from deleting or modifying
 any files that are mounted into its workspace.
@@ -59,7 +58,7 @@ but seems to work well on most systems.
 
 ## All Hands Runtime
 The All Hands Runtime is currently in beta. You can request access by joining
-the #remote-runtime-limited-beta channel on Slack ([see the README](https://github.com/All-Hands-AI/OpenHands?tab=readme-ov-file#-join-our-community) for an invite).
+the #remote-runtime-limited-beta channel on Slack ([see the README](https://github.com/All-Hands-AI/OpenHands?tab=readme-ov-file#-how-to-join-the-community) for an invite).
 
 To use the All Hands Runtime, set the following environment variables when
 starting OpenHands:

docs/static/img/backend_architecture.puml

@@ -123,7 +123,6 @@ class openhands.state.State {
   updated_info: List[Tuple[Action, Observation]]
 }
 class openhands.observation.CmdOutputObservation {
-  command_id: int
   command: str
   exit_code: int
   observation: str

evaluation/benchmarks/EDA/run_infer.py

@@ -75,6 +75,8 @@ def get_config(
         workspace_mount_path=None,
     )
     config.set_llm_config(metadata.llm_config)
+    agent_config = config.get_agent_config(metadata.agent_class)
+    agent_config.use_microagents = False
     return config
 
 

evaluation/benchmarks/agent_bench/run_infer.py

@@ -59,6 +59,8 @@ def get_config(
         workspace_mount_path=None,
     )
     config.set_llm_config(metadata.llm_config)
+    agent_config = config.get_agent_config(metadata.agent_class)
+    agent_config.use_microagents = False
     return config
 
 
@@ -135,7 +137,6 @@ def complete_runtime(
 
         action = CmdRunAction(
             command=f'chmod +x ./{script_name} && ./{script_name}',
-            keep_prompt=False,
         )
         logger.info(action, extra={'msg_type': 'ACTION'})
         obs = runtime.run_action(action)
@@ -162,8 +163,7 @@ def complete_runtime(
             logger.info(f'Running get ground truth cmd: {script_name}')
 
             action = CmdRunAction(
-                command=f'chmod +x ./{script_name} && ./{script_name}',
-                keep_prompt=False,
+                command=f'chmod +x ./{script_name} && ./{script_name}'
             )
             logger.info(action, extra={'msg_type': 'ACTION'})
             obs = runtime.run_action(action)

evaluation/benchmarks/aider_bench/run_infer.py

@@ -67,6 +67,8 @@ def get_config(
         workspace_mount_path=None,
     )
     config.set_llm_config(metadata.llm_config)
+    agent_config = config.get_agent_config(metadata.agent_class)
+    agent_config.use_microagents = False
 
     # copy 'draft_editor' config if exists
     config_copy = copy.deepcopy(config)
@@ -143,10 +145,7 @@ def complete_runtime(
         )
         logger.info(f'Running test file: {script_name}')
 
-    action = CmdRunAction(
-        command=f'python3 -m unittest {script_name}',
-        keep_prompt=False,
-    )
+    action = CmdRunAction(command=f'python3 -m unittest {script_name}')
     logger.info(action, extra={'msg_type': 'ACTION'})
     obs = runtime.run_action(action)
     logger.info(obs, extra={'msg_type': 'OBSERVATION'})

evaluation/benchmarks/biocoder/run_infer.py

@@ -73,6 +73,8 @@ def get_config(
         workspace_mount_path=None,
     )
     config.set_llm_config(metadata.llm_config)
+    agent_config = config.get_agent_config(metadata.agent_class)
+    agent_config.use_microagents = False
     return config
 
 
@@ -197,7 +199,7 @@ def complete_runtime(
     if obs.exit_code == 0:
         test_result['metadata']['1_copy_change_success'] = True
 
-        action = CmdRunAction(command=f'cat {generated_path}', keep_prompt=False)
+        action = CmdRunAction(command=f'cat {generated_path}')
         logger.info(action, extra={'msg_type': 'ACTION'})
         obs = runtime.run_action(action)
         assert obs.exit_code == 0
@@ -221,9 +223,7 @@ def complete_runtime(
     logger.info(obs, extra={'msg_type': 'OBSERVATION'})
     assert obs.exit_code == 0
 
-    action = CmdRunAction(
-        command='cat /testing_files/results_biocoder.json', keep_prompt=False
-    )
+    action = CmdRunAction(command='cat /testing_files/results_biocoder.json')
     logger.info(action, extra={'msg_type': 'ACTION'})
     obs = runtime.run_action(action)
     if obs.exit_code == 0:

evaluation/benchmarks/bird/README.md

@@ -127,7 +127,6 @@ For each problem, OpenHands is given a set number of iterations to fix the faili
         "observation": "run",
         "content": "california_schools/california_schools.sqlite\r\n[(1.0,)]",
         "extras": {
-          "command_id": -1,
           "command": "python3 0.py",
           "exit_code": 0
         }

evaluation/benchmarks/bird/run_infer.py

@@ -86,6 +86,8 @@ def get_config(
         workspace_mount_path=None,
     )
     config.set_llm_config(metadata.llm_config)
+    agent_config = config.get_agent_config(metadata.agent_class)
+    agent_config.use_microagents = False
     return config
 
 
@@ -266,10 +268,7 @@ def initialize_runtime(
     runtime.copy_to(db_file, '/workspace')
 
     # Check the database is copied
-    action = CmdRunAction(
-        command='cd /workspace && ls -l',
-        keep_prompt=False,
-    )
+    action = CmdRunAction(command='cd /workspace && ls -l')
     obs = runtime.run_action(action)
     logger.info(obs, extra={'msg_type': 'OBSERVATION'})
     assert obs.exit_code == 0
@@ -298,10 +297,7 @@ def complete_runtime(
     instance_id = instance.instance_id.replace('/', '__')
     path = os.path.join('/workspace', f'{instance_id}.py')
 
-    action = CmdRunAction(
-        command=f'cat {path}',
-        keep_prompt=False,
-    )
+    action = CmdRunAction(command=f'cat {path}')
     obs = runtime.run_action(action)
     logger.info(obs, extra={'msg_type': 'OBSERVATION'})
 

evaluation/benchmarks/browsing_delegation/run_infer.py

@@ -50,6 +50,8 @@ def get_config(
         workspace_mount_path=None,
     )
     config.set_llm_config(metadata.llm_config)
+    agent_config = config.get_agent_config(metadata.agent_class)
+    agent_config.use_microagents = False
     return config
 
 

evaluation/benchmarks/discoverybench/run_infer.py

@@ -77,6 +77,8 @@ def get_config(
         workspace_mount_path=None,
     )
     config.set_llm_config(metadata.llm_config)
+    agent_config = config.get_agent_config(metadata.agent_class)
+    agent_config.use_microagents = False
     agent_config = AgentConfig(
         function_calling=False,
         codeact_enable_jupyter=True,

evaluation/benchmarks/gaia/run_infer.py

@@ -62,6 +62,8 @@ def get_config(
         workspace_mount_path=None,
     )
     config.set_llm_config(metadata.llm_config)
+    agent_config = config.get_agent_config(metadata.agent_class)
+    agent_config.use_microagents = False
     return config
 
 

evaluation/benchmarks/gorilla/run_infer.py

@@ -55,6 +55,8 @@ def get_config(
         workspace_mount_path=None,
     )
     config.set_llm_config(metadata.llm_config)
+    agent_config = config.get_agent_config(metadata.agent_class)
+    agent_config.use_microagents = False
     return config
 
 

evaluation/benchmarks/gpqa/run_infer.py

@@ -76,6 +76,8 @@ def get_config(
         workspace_mount_path=None,
     )
     config.set_llm_config(metadata.llm_config)
+    agent_config = config.get_agent_config(metadata.agent_class)
+    agent_config.use_microagents = False
     return config
 
 

evaluation/benchmarks/humanevalfix/README.md

@@ -71,7 +71,6 @@ For each problem, OpenHands is given a set number of iterations to fix the faili
                 "observation": "run",
                 "content": "[File: /workspace/Python__2.py (14 lines total)]\r\n1:def truncate_number(number: float) -> float:\r\n2:    return number % 1.0 + 1.0\r\n3:\r\n4:\r\n5:\r\n6:\r\n7:\r\n8:\r\n9:def check(truncate_number):\r\n10:    assert truncate_number(3.5) == 0.5\r\n11:    assert abs(truncate_number(1.33) - 0.33) < 1e-6\r\n12:    assert abs(truncate_number(123.456) - 0.456) < 1e-6\r\n13:\r\n14:check(truncate_number)",
                 "extras": {
-                    "command_id": -1,
                     "command": "open Python__2.py",
                     "exit_code": 0
                 }
@@ -98,7 +97,6 @@ For each problem, OpenHands is given a set number of iterations to fix the faili
                 "observation": "run",
                 "content": "> > [File: /workspace/Python__2.py (14 lines total)]\r\n1:def truncate_number(number: float) -> float:\r\n2:    return number % 1.0\r\n3:\r\n4:\r\n5:\r\n6:\r\n7:\r\n8:\r\n9:def check(truncate_number):\r\n10:    assert truncate_number(3.5) == 0.5\r\n11:    assert abs(truncate_number(1.33) - 0.33) < 1e-6\r\n12:    assert abs(truncate_number(123.456) - 0.456) < 1e-6\r\n13:\r\n14:check(truncate_number)\r\nFile updated. Please review the changes and make sure they are correct (correct indentation, no duplicate lines, etc). Edit the file again if necessary.",
                 "extras": {
-                    "command_id": -1,
                     "command": "edit 2:2 <<EOF\n    return number % 1.0\nEOF",
                     "exit_code": 0
                 }
@@ -125,7 +123,6 @@ For each problem, OpenHands is given a set number of iterations to fix the faili
                 "observation": "run",
                 "content": "",
                 "extras": {
-                    "command_id": -1,
                     "command": "python3 Python__2.py",
                     "exit_code": 0
                 }

evaluation/benchmarks/humanevalfix/run_infer.py

@@ -97,6 +97,8 @@ def get_config(
         workspace_mount_path=None,
     )
     config.set_llm_config(metadata.llm_config)
+    agent_config = config.get_agent_config(metadata.agent_class)
+    agent_config.use_microagents = False
     return config
 
 
@@ -169,9 +171,7 @@ def complete_runtime(
     num_workers = LANGUAGE_TO_NUM_WORKERS[language]
     python_imports = '\n'.join(IMPORT_HELPER[language])
 
-    action = CmdRunAction(
-        command=f'cat /workspace/{_get_instance_id(instance)}.py', keep_prompt=False
-    )
+    action = CmdRunAction(command=f'cat /workspace/{_get_instance_id(instance)}.py')
     obs = runtime.run_action(action)
     assert obs.exit_code == 0
 

evaluation/benchmarks/logic_reasoning/run_infer.py

@@ -61,6 +61,8 @@ def get_config(
         workspace_mount_path=None,
     )
     config.set_llm_config(metadata.llm_config)
+    agent_config = config.get_agent_config(metadata.agent_class)
+    agent_config.use_microagents = False
     return config
 
 

evaluation/benchmarks/mint/run_infer.py

@@ -119,6 +119,8 @@ def get_config(
         workspace_mount_path=None,
     )
     config.set_llm_config(metadata.llm_config)
+    agent_config = config.get_agent_config(metadata.agent_class)
+    agent_config.use_microagents = False
     return config
 
 

evaluation/benchmarks/ml_bench/run_infer.py

@@ -92,6 +92,8 @@ def get_config(
         workspace_mount_path=None,
     )
     config.set_llm_config(metadata.llm_config)
+    agent_config = config.get_agent_config(metadata.agent_class)
+    agent_config.use_microagents = False
     return config
 
 
@@ -161,7 +163,7 @@ def complete_runtime(
     eval_script = os.path.join(task_path, 'run.sh')
     logger.info(f'Running evaluation script: {eval_script}')
 
-    action = CmdRunAction(command=f'cat {eval_script}', keep_prompt=False)
+    action = CmdRunAction(command=f'cat {eval_script}')
     logger.info(action, extra={'msg_type': 'ACTION'})
     obs = runtime.run_action(action)
     if obs.exit_code == 0:

evaluation/benchmarks/scienceagentbench/run_infer.py

@@ -121,10 +121,7 @@ def initialize_runtime(
     runtime.copy_to(dataset_dir, '/workspace/benchmark/datasets', recursive=True)
 
     # Check the dataset exists
-    action = CmdRunAction(
-        command='cd /workspace/benchmark/datasets && ls',
-        keep_prompt=False,
-    )
+    action = CmdRunAction(command='cd /workspace/benchmark/datasets && ls')
     obs = runtime.run_action(action)
     logger.info(obs, extra={'msg_type': 'OBSERVATION'})
     assert obs.exit_code == 0
@@ -154,10 +151,7 @@ def complete_runtime(
 
     assert obs.exit_code == 0
 
-    action = CmdRunAction(
-        command=f'cat pred_programs/{instance.pred_program_name}',
-        keep_prompt=False,
-    )
+    action = CmdRunAction(command=f'cat pred_programs/{instance.pred_program_name}')
     logger.info(action, extra={'msg_type': 'ACTION'})
     obs = runtime.run_action(action)
 

evaluation/benchmarks/swe_bench/eval_infer.py

@@ -98,6 +98,7 @@ def process_instance(
     metadata: EvalMetadata,
     reset_logger: bool = True,
     log_dir: str | None = None,
+    runtime_failure_count: int = 0,
 ) -> EvalOutput:
     """
     Evaluate agent performance on a SWE-bench problem instance.
@@ -146,6 +147,16 @@ def process_instance(
             metadata=metadata,
         )
 
+    # Increase resource_factor with increasing attempt_id
+    if runtime_failure_count > 0:
+        config.sandbox.remote_runtime_resource_factor = min(
+            config.sandbox.remote_runtime_resource_factor * (2**runtime_failure_count),
+            4,  # hardcode maximum resource factor to 4
+        )
+        logger.warning(
+            f'This is the second attempt for instance {instance.instance_id}, setting resource factor to {config.sandbox.remote_runtime_resource_factor}'
+        )
+
     runtime = create_runtime(config)
     call_async_from_sync(runtime.connect)
     # Get patch and save it to /tmp/patch.diff
@@ -177,7 +188,7 @@ def process_instance(
         "(patch --batch --fuzz=5 -p1 -i /tmp/patch.diff && echo 'APPLY_PATCH_PASS' || "
         "echo 'APPLY_PATCH_FAIL')))"
     )
-    action = CmdRunAction(command=exec_command, keep_prompt=False)
+    action = CmdRunAction(command=exec_command)
     action.timeout = 600
     obs = runtime.run_action(action)
     assert isinstance(obs, CmdOutputObservation)
@@ -200,9 +211,7 @@ def process_instance(
 
             # Run eval script in background and save output to log file
             log_file = '/tmp/eval_output.log'
-            action = CmdRunAction(
-                command=f'/tmp/eval.sh > {log_file} 2>&1 & echo $!', keep_prompt=False
-            )
+            action = CmdRunAction(command=f'/tmp/eval.sh > {log_file} 2>&1 & echo $!')
             action.timeout = 60  # Short timeout just to get the process ID
             obs = runtime.run_action(action)
 
@@ -224,7 +233,7 @@ def process_instance(
                         instance['test_result']['report']['test_timeout'] = True
                         break
                     check_action = CmdRunAction(
-                        command=f'ps -p {pid} > /dev/null; echo $?', keep_prompt=False
+                        command=f'ps -p {pid} > /dev/null; echo $?'
                     )
                     check_action.timeout = 60
                     check_obs = runtime.run_action(check_action)
@@ -242,7 +251,7 @@ def process_instance(
                     time.sleep(30)  # Wait for 30 seconds before checking again
 
                 # Read the log file
-                cat_action = CmdRunAction(command=f'cat {log_file}', keep_prompt=False)
+                cat_action = CmdRunAction(command=f'cat {log_file}')
                 cat_action.timeout = 300
                 cat_obs = runtime.run_action(cat_action)
 

evaluation/benchmarks/swe_bench/run_infer.py

@@ -282,6 +282,16 @@ def initialize_runtime(
     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.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)
@@ -337,8 +347,7 @@ def complete_runtime(
     git_patch = None
     while n_retries < 5:
         action = CmdRunAction(
-            command=f'git diff --no-color --cached {instance["base_commit"]}',
-            keep_prompt=False,
+            command=f'git diff --no-color --cached {instance["base_commit"]}'
         )
         action.timeout = 600 + 100 * n_retries
         logger.info(action, extra={'msg_type': 'ACTION'})
@@ -385,7 +394,7 @@ def process_instance(
     if runtime_failure_count > 0:
         config.sandbox.remote_runtime_resource_factor = min(
             config.sandbox.remote_runtime_resource_factor * (2**runtime_failure_count),
-            2,  # hardcode maximum resource factor to 2
+            8,
         )
         logger.warning(
             f'This is the second attempt for instance {instance.instance_id}, setting resource factor to {config.sandbox.remote_runtime_resource_factor}'
@@ -535,4 +544,5 @@ if __name__ == '__main__':
         args.eval_num_workers,
         process_instance,
         timeout_seconds=120 * 60,  # 2 hour PER instance should be more than enough
+        max_retries=5,
     )

evaluation/benchmarks/swe_bench/scripts/eval/compare_outputs.py

@@ -1,13 +1,20 @@
 #!/usr/bin/env python3
 import argparse
+import os
 
 import pandas as pd
+from termcolor import colored
 
 parser = argparse.ArgumentParser(
     description='Compare two swe_bench output JSONL files and print the resolved diff'
 )
 parser.add_argument('input_file_1', type=str)
 parser.add_argument('input_file_2', type=str)
+parser.add_argument(
+    '--show-paths',
+    action='store_true',
+    help='Show visualization paths for failed instances',
+)
 args = parser.parse_args()
 
 df1 = pd.read_json(args.input_file_1, orient='records', lines=True)
@@ -58,10 +65,60 @@ df_diff_y_only = df_diff[~df_diff['resolved_x'] & df_diff['resolved_y']].sort_va
 print(f'# y resolved but x not={df_diff_y_only.shape[0]}')
 print(df_diff_y_only[['instance_id', 'report_x', 'report_y']])
 # get instance_id from df_diff_y_only
-print('-' * 100)
-print('Instances that x resolved but y not:')
-print(df_diff_x_only['instance_id'].tolist())
+
+x_only_by_repo = {}
+for instance_id in df_diff_x_only['instance_id'].tolist():
+    repo = instance_id.split('__')[0]
+    x_only_by_repo.setdefault(repo, []).append(instance_id)
+y_only_by_repo = {}
+for instance_id in df_diff_y_only['instance_id'].tolist():
+    repo = instance_id.split('__')[0]
+    y_only_by_repo.setdefault(repo, []).append(instance_id)
 
 print('-' * 100)
-print('Instances that y resolved but x not:')
-print(df_diff_y_only['instance_id'].tolist())
+print(
+    colored('Repository comparison (x resolved vs y resolved):', 'cyan', attrs=['bold'])
+)
+all_repos = sorted(set(list(x_only_by_repo.keys()) + list(y_only_by_repo.keys())))
+
+# Calculate diffs and sort repos by diff magnitude
+repo_diffs = []
+for repo in all_repos:
+    x_count = len(x_only_by_repo.get(repo, []))
+    y_count = len(y_only_by_repo.get(repo, []))
+    diff = abs(x_count - y_count)
+    repo_diffs.append((repo, diff))
+
+# Sort by diff (descending) and then by repo name
+repo_diffs.sort(key=lambda x: (-x[1], x[0]))
+threshold = max(
+    3, sum(d[1] for d in repo_diffs) / len(repo_diffs) * 1.5 if repo_diffs else 0
+)
+
+x_input_file_folder = os.path.join(os.path.dirname(args.input_file_1), 'output.viz')
+
+for repo, diff in repo_diffs:
+    x_instances = x_only_by_repo.get(repo, [])
+    y_instances = y_only_by_repo.get(repo, [])
+
+    # Determine if this repo has a significant diff
+    is_significant = diff >= threshold
+    repo_color = 'red' if is_significant else 'yellow'
+
+    print(f"\n{colored(repo, repo_color, attrs=['bold'])}:")
+    print(colored(f'Difference: {diff} instances!', repo_color, attrs=['bold']))
+    print(colored(f'X resolved but Y failed: ({len(x_instances)} instances)', 'green'))
+    if x_instances:
+        print('  ' + str(x_instances))
+    print(colored(f'Y resolved but X failed: ({len(y_instances)} instances)', 'red'))
+    if y_instances:
+        print('  ' + str(y_instances))
+        if args.show_paths:
+            print(
+                colored('    Visualization path for X failed:', 'cyan', attrs=['bold'])
+            )
+            for instance_id in y_instances:
+                instance_file = os.path.join(
+                    x_input_file_folder, f'false.{instance_id}.md'
+                )
+                print(f'    {instance_file}')

evaluation/benchmarks/swe_bench/scripts/eval/convert_oh_output_to_md.py

@@ -20,6 +20,13 @@ output_md_folder = args.oh_output_file.replace('.jsonl', '.viz')
 print(f'Converting {args.oh_output_file} to markdown files in {output_md_folder}')
 
 oh_format = pd.read_json(args.oh_output_file, orient='records', lines=True)
+
+swebench_eval_file = args.oh_output_file.replace('.jsonl', '.swebench_eval.jsonl')
+if os.path.exists(swebench_eval_file):
+    eval_output_df = pd.read_json(swebench_eval_file, orient='records', lines=True)
+else:
+    eval_output_df = None
+
 # model name is the folder name of oh_output_file
 model_name = os.path.basename(os.path.dirname(args.oh_output_file))
 
@@ -50,7 +57,7 @@ def convert_history_to_str(history):
     return ret
 
 
-def write_row_to_md_file(row):
+def write_row_to_md_file(row, instance_id_to_test_result):
     if 'git_patch' in row:
         model_patch = row['git_patch']
     elif 'test_result' in row and 'git_patch' in row['test_result']:
@@ -58,8 +65,21 @@ def write_row_to_md_file(row):
     else:
         raise ValueError(f'Row {row} does not have a git_patch')
 
-    if 'report' in row:
-        resolved = row['report'].get('resolved', False)
+    test_output = None
+    if row['instance_id'] in instance_id_to_test_result:
+        report = instance_id_to_test_result[row['instance_id']].get('report', {})
+        resolved = report.get('resolved', False)
+        test_output = instance_id_to_test_result[row['instance_id']].get(
+            'test_output', None
+        )
+    elif 'report' in row and row['report'] is not None:
+        if not isinstance(row['report'], dict):
+            resolved = None
+            print(
+                f'ERROR: Report is not a dict, but a {type(row["report"])}. Row: {row}'
+            )
+        else:
+            resolved = row['report'].get('resolved', False)
     else:
         resolved = None
 
@@ -84,5 +104,18 @@ def write_row_to_md_file(row):
         f.write('## Model Patch\n')
         f.write(f'{process_git_patch(model_patch)}\n')
 
+        f.write('## Test Output\n')
+        f.write(str(test_output))
 
-oh_format.progress_apply(write_row_to_md_file, axis=1)
+
+instance_id_to_test_result = {}
+if eval_output_df is not None:
+    instance_id_to_test_result = (
+        eval_output_df[['instance_id', 'test_result']]
+        .set_index('instance_id')['test_result']
+        .to_dict()
+    )
+
+oh_format.progress_apply(
+    write_row_to_md_file, axis=1, instance_id_to_test_result=instance_id_to_test_result
+)

evaluation/benchmarks/swe_bench/scripts/eval/update_output_with_eval.py

@@ -111,6 +111,11 @@ elif os.path.exists(openhands_remote_report_jsonl):
         instance_id_to_status[row['instance_id']] = row['test_result']['report']
     df['report'] = df.apply(apply_report, axis=1)
 
+    report_is_dict = df['report'].apply(lambda x: isinstance(x, dict))
+    if not report_is_dict.all():
+        print(df[~report_is_dict])
+        raise ValueError(f'Report is not a dict, but a {type(row["report"])}')
+
     _n_instances = len(df)
     _n_resolved = len(df[df['report'].apply(lambda x: x.get('resolved', False))])
     _n_unresolved = _n_instances - _n_resolved

evaluation/benchmarks/toolqa/run_infer.py

@@ -56,6 +56,8 @@ def get_config(
         workspace_mount_path=None,
     )
     config.set_llm_config(metadata.llm_config)
+    agent_config = config.get_agent_config(metadata.agent_class)
+    agent_config.use_microagents = False
     return config
 
 

evaluation/benchmarks/webarena/run_infer.py

@@ -77,6 +77,8 @@ def get_config(
         workspace_mount_path=None,
     )
     config.set_llm_config(metadata.llm_config)
+    agent_config = config.get_agent_config(metadata.agent_class)
+    agent_config.use_microagents = False
     return config
 
 

evaluation/integration_tests/tests/t01_fix_simple_typo.py

@@ -24,7 +24,7 @@ class Test(BaseIntegrationTest):
     @classmethod
     def verify_result(cls, runtime: Runtime, histories: list[Event]) -> TestResult:
         # check if the file /workspace/bad.txt has been fixed
-        action = CmdRunAction(command='cat /workspace/bad.txt', keep_prompt=False)
+        action = CmdRunAction(command='cat /workspace/bad.txt')
         obs = runtime.run_action(action)
         if obs.exit_code != 0:
             return TestResult(

evaluation/integration_tests/tests/t02_add_bash_hello.py

@@ -10,14 +10,14 @@ class Test(BaseIntegrationTest):
 
     @classmethod
     def initialize_runtime(cls, runtime: Runtime) -> None:
-        action = CmdRunAction(command='mkdir -p /workspace', keep_prompt=False)
+        action = CmdRunAction(command='mkdir -p /workspace')
         obs = runtime.run_action(action)
         assert_and_raise(obs.exit_code == 0, f'Failed to run command: {obs.content}')
 
     @classmethod
     def verify_result(cls, runtime: Runtime, histories: list[Event]) -> TestResult:
         # check if the file /workspace/hello.sh exists
-        action = CmdRunAction(command='cat /workspace/hello.sh', keep_prompt=False)
+        action = CmdRunAction(command='cat /workspace/hello.sh')
         obs = runtime.run_action(action)
         if obs.exit_code != 0:
             return TestResult(
@@ -26,7 +26,7 @@ class Test(BaseIntegrationTest):
             )
 
         # execute the script
-        action = CmdRunAction(command='bash /workspace/hello.sh', keep_prompt=False)
+        action = CmdRunAction(command='bash /workspace/hello.sh')
         obs = runtime.run_action(action)
         if obs.exit_code != 0:
             return TestResult(

evaluation/integration_tests/tests/t03_jupyter_write_file.py

@@ -10,14 +10,14 @@ class Test(BaseIntegrationTest):
 
     @classmethod
     def initialize_runtime(cls, runtime: Runtime) -> None:
-        action = CmdRunAction(command='mkdir -p /workspace', keep_prompt=False)
+        action = CmdRunAction(command='mkdir -p /workspace')
         obs = runtime.run_action(action)
         assert_and_raise(obs.exit_code == 0, f'Failed to run command: {obs.content}')
 
     @classmethod
     def verify_result(cls, runtime: Runtime, histories: list[Event]) -> TestResult:
         # check if the file /workspace/hello.sh exists
-        action = CmdRunAction(command='cat /workspace/test.txt', keep_prompt=False)
+        action = CmdRunAction(command='cat /workspace/test.txt')
         obs = runtime.run_action(action)
         if obs.exit_code != 0:
             return TestResult(
@@ -26,7 +26,7 @@ class Test(BaseIntegrationTest):
             )
 
         # execute the script
-        action = CmdRunAction(command='cat /workspace/test.txt', keep_prompt=False)
+        action = CmdRunAction(command='cat /workspace/test.txt')
         obs = runtime.run_action(action)
 
         if obs.exit_code != 0:

evaluation/integration_tests/tests/t04_git_staging.py

@@ -10,31 +10,29 @@ class Test(BaseIntegrationTest):
 
     @classmethod
     def initialize_runtime(cls, runtime: Runtime) -> None:
-        action = CmdRunAction(command='mkdir -p /workspace', keep_prompt=False)
+        action = CmdRunAction(command='mkdir -p /workspace')
         obs = runtime.run_action(action)
         assert_and_raise(obs.exit_code == 0, f'Failed to run command: {obs.content}')
 
         # git init
-        action = CmdRunAction(command='git init', keep_prompt=False)
+        action = CmdRunAction(command='git init')
         obs = runtime.run_action(action)
         assert_and_raise(obs.exit_code == 0, f'Failed to run command: {obs.content}')
 
         # create README.md
-        action = CmdRunAction(
-            command='echo \'print("hello world")\' > hello.py', keep_prompt=False
-        )
+        action = CmdRunAction(command='echo \'print("hello world")\' > hello.py')
         obs = runtime.run_action(action)
         assert_and_raise(obs.exit_code == 0, f'Failed to run command: {obs.content}')
 
         # git add README.md
-        action = CmdRunAction(command='git add hello.py', keep_prompt=False)
+        action = CmdRunAction(command='git add hello.py')
         obs = runtime.run_action(action)
         assert_and_raise(obs.exit_code == 0, f'Failed to run command: {obs.content}')
 
     @classmethod
     def verify_result(cls, runtime: Runtime, histories: list[Event]) -> TestResult:
         # check if the file /workspace/hello.py exists
-        action = CmdRunAction(command='cat /workspace/hello.py', keep_prompt=False)
+        action = CmdRunAction(command='cat /workspace/hello.py')
         obs = runtime.run_action(action)
         if obs.exit_code != 0:
             return TestResult(
@@ -43,7 +41,7 @@ class Test(BaseIntegrationTest):
             )
 
         # check if the staging area is empty
-        action = CmdRunAction(command='git status', keep_prompt=False)
+        action = CmdRunAction(command='git status')
         obs = runtime.run_action(action)
         if obs.exit_code != 0:
             return TestResult(

evaluation/integration_tests/tests/t05_simple_browsing.py

@@ -83,11 +83,11 @@ class Test(BaseIntegrationTest):
 
     @classmethod
     def initialize_runtime(cls, runtime: Runtime) -> None:
-        action = CmdRunAction(command='mkdir -p /workspace', keep_prompt=False)
+        action = CmdRunAction(command='mkdir -p /workspace')
         obs = runtime.run_action(action)
         assert_and_raise(obs.exit_code == 0, f'Failed to run command: {obs.content}')
 
-        action = CmdRunAction(command='mkdir -p /tmp/server', keep_prompt=False)
+        action = CmdRunAction(command='mkdir -p /tmp/server')
         obs = runtime.run_action(action)
         assert_and_raise(obs.exit_code == 0, f'Failed to run command: {obs.content}')
 
@@ -101,8 +101,7 @@ class Test(BaseIntegrationTest):
 
         # create README.md
         action = CmdRunAction(
-            command='cd /tmp/server && nohup python3 -m http.server 8000 &',
-            keep_prompt=False,
+            command='cd /tmp/server && nohup python3 -m http.server 8000 &'
         )
         obs = runtime.run_action(action)
 

evaluation/utils/shared.py

@@ -376,7 +376,12 @@ def _process_instance_wrapper(
                 + '\n'
             )
             if isinstance(
-                e, (AgentRuntimeDisconnectedError, AgentRuntimeUnavailableError)
+                e,
+                (
+                    AgentRuntimeDisconnectedError,
+                    AgentRuntimeUnavailableError,
+                    AgentRuntimeNotFoundError,
+                ),
             ):
                 runtime_failure_count += 1
                 msg += f'Runtime disconnected error detected for instance {instance.instance_id}, runtime failure count: {runtime_failure_count}'

frontend/.eslintrc

@@ -73,6 +73,7 @@
             }
           }
         ],
+        "react/prop-types": "off",
         "react/no-array-index-key": "off",
         "react-hooks/exhaustive-deps": "off",
         "import/no-extraneous-dependencies": "off",

frontend/__tests__/components/chat/chat-input.test.tsx

@@ -218,4 +218,30 @@ describe("ChatInput", () => {
     // Verify image paste was handled
     expect(onImagePaste).toHaveBeenCalledWith([file]);
   });
+
+  it("should not submit when Enter is pressed during IME composition", async () => {
+    const user = userEvent.setup();
+    render(<ChatInput onSubmit={onSubmitMock} />);
+    const textarea = screen.getByRole("textbox");
+
+    await user.type(textarea, "こんにちは");
+
+    // Simulate Enter during IME composition
+    fireEvent.keyDown(textarea, {
+      key: "Enter",
+      isComposing: true,
+      nativeEvent: { isComposing: true },
+    });
+
+    expect(onSubmitMock).not.toHaveBeenCalled();
+
+    // Simulate normal Enter after composition is done
+    fireEvent.keyDown(textarea, {
+      key: "Enter",
+      isComposing: false,
+      nativeEvent: { isComposing: false },
+    });
+
+    expect(onSubmitMock).toHaveBeenCalledWith("こんにちは");
+  });
 });

frontend/__tests__/components/features/github/github-repo-selector.test.tsx

@@ -0,0 +1,79 @@
+import { screen } from "@testing-library/react";
+import { describe, expect, it, vi } from "vitest";
+import { renderWithProviders } from "test-utils";
+import { GitHubRepositorySelector } from "#/components/features/github/github-repo-selector";
+import OpenHands from "#/api/open-hands";
+import * as GitHubAPI from "#/api/github";
+
+describe("GitHubRepositorySelector", () => {
+  const onInputChangeMock = vi.fn();
+  const onSelectMock = vi.fn();
+
+  it("should render the search input", () => {
+    renderWithProviders(
+      <GitHubRepositorySelector
+        onInputChange={onInputChangeMock}
+        onSelect={onSelectMock}
+        publicRepositories={[]}
+        userRepositories={[]}
+      />,
+    );
+
+    expect(
+      screen.getByPlaceholderText("Select a GitHub project"),
+    ).toBeInTheDocument();
+  });
+
+  it("should show the GitHub login button in OSS mode", () => {
+    const getConfigSpy = vi.spyOn(OpenHands, "getConfig");
+    getConfigSpy.mockResolvedValue({
+      APP_MODE: "oss",
+      APP_SLUG: "openhands",
+      GITHUB_CLIENT_ID: "test-client-id",
+      POSTHOG_CLIENT_KEY: "test-posthog-key",
+    });
+
+    renderWithProviders(
+      <GitHubRepositorySelector
+        onInputChange={onInputChangeMock}
+        onSelect={onSelectMock}
+        publicRepositories={[]}
+        userRepositories={[]}
+      />,
+    );
+
+    expect(screen.getByTestId("github-repo-selector")).toBeInTheDocument();
+  });
+
+  it("should show the search results", () => {
+    const mockSearchedRepos = [
+      {
+        id: 1,
+        full_name: "test/repo1",
+        stargazers_count: 100,
+      },
+      {
+        id: 2,
+        full_name: "test/repo2",
+        stargazers_count: 200,
+      },
+    ];
+
+    const searchPublicRepositoriesSpy = vi.spyOn(
+      GitHubAPI,
+      "searchPublicRepositories",
+    );
+    searchPublicRepositoriesSpy.mockResolvedValue(mockSearchedRepos);
+
+    renderWithProviders(
+      <GitHubRepositorySelector
+        onInputChange={onInputChangeMock}
+        onSelect={onSelectMock}
+        publicRepositories={[]}
+        userRepositories={[]}
+      />,
+    );
+
+    expect(screen.getByTestId("github-repo-selector")).toBeInTheDocument();
+  });
+});

frontend/package-lock.json

@@ -14,7 +14,7 @@
         "@react-router/serve": "^7.1.1",
         "@react-types/shared": "^3.25.0",
         "@reduxjs/toolkit": "^2.5.0",
-        "@tanstack/react-query": "^5.62.11",
+        "@tanstack/react-query": "^5.62.12",
         "@vitejs/plugin-react": "^4.3.2",
         "@xterm/addon-fit": "^0.10.0",
         "@xterm/xterm": "^5.4.0",
@@ -27,11 +27,11 @@
         "isbot": "^5.1.19",
         "jose": "^5.9.4",
         "monaco-editor": "^0.52.2",
-        "posthog-js": "^1.203.2",
+        "posthog-js": "^1.203.3",
         "react": "^19.0.0",
         "react-dom": "^19.0.0",
         "react-highlight": "^0.15.0",
-        "react-hot-toast": "^2.4.1",
+        "react-hot-toast": "^2.5.1",
         "react-i18next": "^15.4.0",
         "react-icons": "^5.4.0",
         "react-markdown": "^9.0.1",
@@ -56,7 +56,7 @@
         "@testing-library/jest-dom": "^6.6.1",
         "@testing-library/react": "^16.1.0",
         "@testing-library/user-event": "^14.5.2",
-        "@types/node": "^22.10.2",
+        "@types/node": "^22.10.5",
         "@types/react": "^19.0.2",
         "@types/react-dom": "^19.0.2",
         "@types/react-highlight": "^0.12.8",
@@ -83,7 +83,7 @@
         "postcss": "^8.4.47",
         "prettier": "^3.4.2",
         "tailwindcss": "^3.4.17",
-        "typescript": "^5.6.3",
+        "typescript": "^5.7.2",
         "vite-plugin-svgr": "^4.2.0",
         "vite-tsconfig-paths": "^5.1.4",
         "vitest": "^1.6.0"
@@ -5361,22 +5361,20 @@
       }
     },
     "node_modules/@tanstack/query-core": {
-      "version": "5.62.9",
-      "resolved": "https://registry.npmjs.org/@tanstack/query-core/-/query-core-5.62.9.tgz",
-      "integrity": "sha512-lwePd8hNYhyQ4nM/iRQ+Wz2cDtspGeZZHFZmCzHJ7mfKXt+9S301fULiY2IR2byJYY6Z03T427E5PoVfMexHjw==",
-      "license": "MIT",
+      "version": "5.62.12",
+      "resolved": "https://registry.npmjs.org/@tanstack/query-core/-/query-core-5.62.12.tgz",
+      "integrity": "sha512-6igFeBgymHkCxVgaEk+yiLwkMf9haui/EQLmI3o9CatOyDThEoFKe8toLWvWliZC/Jf+h7NwHi/zjfyLArr1ow==",
       "funding": {
         "type": "github",
         "url": "https://github.com/sponsors/tannerlinsley"
       }
     },
     "node_modules/@tanstack/react-query": {
-      "version": "5.62.11",
-      "resolved": "https://registry.npmjs.org/@tanstack/react-query/-/react-query-5.62.11.tgz",
-      "integrity": "sha512-Xb1nw0cYMdtFmwkvH9+y5yYFhXvLRCnXoqlzSw7UkqtCVFq3cG8q+rHZ2Yz1XrC+/ysUaTqbLKJqk95mCgC1oQ==",
-      "license": "MIT",
+      "version": "5.62.12",
+      "resolved": "https://registry.npmjs.org/@tanstack/react-query/-/react-query-5.62.12.tgz",
+      "integrity": "sha512-yt8p7l5MlHA3QCt6xF1Cu9dw1Anf93yTK+DMDJQ64h/mshAymVAtcwj8TpsyyBrZNWAAZvza/m76bnTSR79ZtQ==",
       "dependencies": {
-        "@tanstack/query-core": "5.62.9"
+        "@tanstack/query-core": "5.62.12"
       },
       "funding": {
         "type": "github",
@@ -5627,11 +5625,10 @@
       "license": "MIT"
     },
     "node_modules/@types/node": {
-      "version": "22.10.2",
-      "resolved": "https://registry.npmjs.org/@types/node/-/node-22.10.2.tgz",
-      "integrity": "sha512-Xxr6BBRCAOQixvonOye19wnzyDiUtTeqldOOmj3CkeblonbccA12PFwlufvRdrpjXxqnmUaeiU5EOA+7s5diUQ==",
+      "version": "22.10.5",
+      "resolved": "https://registry.npmjs.org/@types/node/-/node-22.10.5.tgz",
+      "integrity": "sha512-F8Q+SeGimwOo86fiovQh8qiXfFEh2/ocYv7tU5pJ3EXMSSxk1Joj5wefpFK2fHTf/N6HKGSxIDBT9f3gCxXPkQ==",
       "devOptional": true,
-      "license": "MIT",
       "dependencies": {
         "undici-types": "~6.20.0"
       }
@@ -13810,10 +13807,9 @@
       "license": "MIT"
     },
     "node_modules/posthog-js": {
-      "version": "1.203.2",
-      "resolved": "https://registry.npmjs.org/posthog-js/-/posthog-js-1.203.2.tgz",
-      "integrity": "sha512-3aLpEhM4i9sQQtobRmDttJ3rTW1+gwQ9HL7QiOeDueE2T7CguYibYS7weY1UhXMerx5lh1A7+szlOJTTibifLQ==",
-      "license": "MIT",
+      "version": "1.203.3",
+      "resolved": "https://registry.npmjs.org/posthog-js/-/posthog-js-1.203.3.tgz",
+      "integrity": "sha512-DTK6LfL87xC7PPleKDParEIfkXl7hXtuDeSOPfhcyCXLuVspq0z7YyRB5dQE9Pbalf3yoGqUKvomYFp/BGVfQg==",
       "dependencies": {
         "core-js": "^3.38.1",
         "fflate": "^0.4.8",
@@ -14144,12 +14140,12 @@
       }
     },
     "node_modules/react-hot-toast": {
-      "version": "2.4.1",
-      "resolved": "https://registry.npmjs.org/react-hot-toast/-/react-hot-toast-2.4.1.tgz",
-      "integrity": "sha512-j8z+cQbWIM5LY37pR6uZR6D4LfseplqnuAO4co4u8917hBUvXlEqyP1ZzqVLcqoyUesZZv/ImreoCeHVDpE5pQ==",
-      "license": "MIT",
+      "version": "2.5.1",
+      "resolved": "https://registry.npmjs.org/react-hot-toast/-/react-hot-toast-2.5.1.tgz",
+      "integrity": "sha512-54Gq1ZD1JbmAb4psp9bvFHjS7lje+8ubboUmvKZkCsQBLH6AOpZ9JemfRvIdHcfb9AZXRaFLrb3qUobGYDJhFQ==",
       "dependencies": {
-        "goober": "^2.1.10"
+        "csstype": "^3.1.3",
+        "goober": "^2.1.16"
       },
       "engines": {
         "node": ">=10"

frontend/package.json

@@ -13,7 +13,7 @@
     "@react-router/serve": "^7.1.1",
     "@react-types/shared": "^3.25.0",
     "@reduxjs/toolkit": "^2.5.0",
-    "@tanstack/react-query": "^5.62.11",
+    "@tanstack/react-query": "^5.62.12",
     "@vitejs/plugin-react": "^4.3.2",
     "@xterm/addon-fit": "^0.10.0",
     "@xterm/xterm": "^5.4.0",
@@ -26,11 +26,11 @@
     "isbot": "^5.1.19",
     "jose": "^5.9.4",
     "monaco-editor": "^0.52.2",
-    "posthog-js": "^1.203.2",
+    "posthog-js": "^1.203.3",
     "react": "^19.0.0",
     "react-dom": "^19.0.0",
     "react-highlight": "^0.15.0",
-    "react-hot-toast": "^2.4.1",
+    "react-hot-toast": "^2.5.1",
     "react-i18next": "^15.4.0",
     "react-icons": "^5.4.0",
     "react-markdown": "^9.0.1",
@@ -83,7 +83,7 @@
     "@testing-library/jest-dom": "^6.6.1",
     "@testing-library/react": "^16.1.0",
     "@testing-library/user-event": "^14.5.2",
-    "@types/node": "^22.10.2",
+    "@types/node": "^22.10.5",
     "@types/react": "^19.0.2",
     "@types/react-dom": "^19.0.2",
     "@types/react-highlight": "^0.12.8",
@@ -110,7 +110,7 @@
     "postcss": "^8.4.47",
     "prettier": "^3.4.2",
     "tailwindcss": "^3.4.17",
-    "typescript": "^5.6.3",
+    "typescript": "^5.7.2",
     "vite-plugin-svgr": "^4.2.0",
     "vite-tsconfig-paths": "^5.1.4",
     "vitest": "^1.6.0"

frontend/src/api/github.ts

@@ -104,6 +104,26 @@ export const retrieveGitHubUser = async () => {
   return user;
 };
 
+export const searchPublicRepositories = async (
+  query: string,
+  per_page = 5,
+  sort: "" | "updated" | "stars" | "forks" = "stars",
+  order: "desc" | "asc" = "desc",
+): Promise<GitHubRepository[]> => {
+  const response = await github.get<{ items: GitHubRepository[] }>(
+    "/search/repositories",
+    {
+      params: {
+        q: query,
+        per_page,
+        sort,
+        order,
+      },
+    },
+  );
+  return response.data.items;
+};
+
 export const retrieveLatestGitHubCommit = async (
   repository: string,
 ): Promise<GitHubCommit | null> => {

frontend/src/components/agent-status-map.constant.ts

@@ -61,4 +61,8 @@ export const AGENT_STATUS_MAP: {
     message: I18nKey.CHAT_INTERFACE$AGENT_ACTION_USER_REJECTED_MESSAGE,
     indicator: IndicatorColor.RED,
   },
+  [AgentState.RATE_LIMITED]: {
+    message: I18nKey.CHAT_INTERFACE$AGENT_RATE_LIMITED_MESSAGE,
+    indicator: IndicatorColor.YELLOW,
+  },
 };

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

@@ -94,7 +94,12 @@ export function ChatInput({
   };
 
   const handleKeyPress = (event: React.KeyboardEvent<HTMLTextAreaElement>) => {
-    if (event.key === "Enter" && !event.shiftKey && !disabled) {
+    if (
+      event.key === "Enter" &&
+      !event.shiftKey &&
+      !disabled &&
+      !event.nativeEvent.isComposing
+    ) {
       event.preventDefault();
       handleSubmitMessage();
     }

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

@@ -154,7 +154,8 @@ export function ChatInterface() {
           onStop={handleStop}
           isDisabled={
             curAgentState === AgentState.LOADING ||
-            curAgentState === AgentState.AWAITING_USER_CONFIRMATION
+            curAgentState === AgentState.AWAITING_USER_CONFIRMATION ||
+            curAgentState === AgentState.RATE_LIMITED
           }
           mode={curAgentState === AgentState.RUNNING ? "stop" : "submit"}
           value={messageToSend ?? undefined}

frontend/src/components/features/chat/messages.tsx

@@ -1,3 +1,4 @@
+import React from "react";
 import { ChatMessage } from "#/components/features/chat/chat-message";
 import { ConfirmationButtons } from "#/components/shared/buttons/confirmation-buttons";
 import { ImageCarousel } from "../images/image-carousel";
@@ -8,32 +9,36 @@ interface MessagesProps {
   isAwaitingUserConfirmation: boolean;
 }
 
-export function Messages({
-  messages,
-  isAwaitingUserConfirmation,
-}: MessagesProps) {
-  return messages.map((message, index) => {
-    if (message.type === "error" || message.type === "action") {
-      return (
-        <ExpandableMessage
-          key={index}
-          type={message.type}
-          id={message.translationID}
-          message={message.content}
-          success={message.success}
-        />
-      );
-    }
+export const Messages: React.FC<MessagesProps> = React.memo(
+  ({ messages, isAwaitingUserConfirmation }) =>
+    messages.map((message, index) => {
+      if (message.type === "error" || message.type === "action") {
+        return (
+          <ExpandableMessage
+            key={index}
+            type={message.type}
+            id={message.translationID}
+            message={message.content}
+            success={message.success}
+          />
+        );
+      }
 
-    return (
-      <ChatMessage key={index} type={message.sender} message={message.content}>
-        {message.imageUrls && message.imageUrls.length > 0 && (
-          <ImageCarousel size="small" images={message.imageUrls} />
-        )}
-        {messages.length - 1 === index &&
-          message.sender === "assistant" &&
-          isAwaitingUserConfirmation && <ConfirmationButtons />}
-      </ChatMessage>
-    );
-  });
-}
+      return (
+        <ChatMessage
+          key={index}
+          type={message.sender}
+          message={message.content}
+        >
+          {message.imageUrls && message.imageUrls.length > 0 && (
+            <ImageCarousel size="small" images={message.imageUrls} />
+          )}
+          {messages.length - 1 === index &&
+            message.sender === "assistant" &&
+            isAwaitingUserConfirmation && <ConfirmationButtons />}
+        </ChatMessage>
+      );
+    }),
+);
+
+Messages.displayName = "Messages";

frontend/src/components/features/github/github-repo-selector.tsx

@@ -1,40 +1,43 @@
 import React from "react";
-import { Autocomplete, AutocompleteItem } from "@nextui-org/react";
+import {
+  Autocomplete,
+  AutocompleteItem,
+  AutocompleteSection,
+} from "@nextui-org/react";
 import { useDispatch } from "react-redux";
 import posthog from "posthog-js";
 import { setSelectedRepository } from "#/state/initial-query-slice";
 import { useConfig } from "#/hooks/query/use-config";
+import { sanitizeQuery } from "#/utils/sanitize-query";
 
 interface GitHubRepositorySelectorProps {
+  onInputChange: (value: string) => void;
   onSelect: () => void;
-  repositories: GitHubRepository[];
+  userRepositories: GitHubRepository[];
+  publicRepositories: GitHubRepository[];
 }
 
 export function GitHubRepositorySelector({
+  onInputChange,
   onSelect,
-  repositories,
+  userRepositories,
+  publicRepositories,
 }: GitHubRepositorySelectorProps) {
   const { data: config } = useConfig();
   const [selectedKey, setSelectedKey] = React.useState<string | null>(null);
 
-  // Add option to install app onto more repos
-  const finalRepositories =
-    config?.APP_MODE === "saas"
-      ? [{ id: -1000, full_name: "Add more repositories..." }, ...repositories]
-      : repositories;
+  const allRepositories: GitHubRepository[] = [
+    ...publicRepositories.filter(
+      (repo) => !publicRepositories.find((r) => r.id === repo.id),
+    ),
+    ...userRepositories,
+  ];
 
   const dispatch = useDispatch();
 
   const handleRepoSelection = (id: string | null) => {
-    const repo = finalRepositories.find((r) => r.id.toString() === id);
-    if (id === "-1000") {
-      if (config?.APP_SLUG)
-        window.open(
-          `https://github.com/apps/${config.APP_SLUG}/installations/new`,
-          "_blank",
-        );
-    } else if (repo) {
-      // set query param
+    const repo = allRepositories.find((r) => r.id.toString() === id);
+    if (repo) {
       dispatch(setSelectedRepository(repo.full_name));
       posthog.capture("repository_selected");
       onSelect();
@@ -43,22 +46,10 @@ export function GitHubRepositorySelector({
   };
 
   const handleClearSelection = () => {
-    // clear query param
     dispatch(setSelectedRepository(null));
   };
 
-  const emptyContent = config?.APP_SLUG ? (
-    <a
-      href={`https://github.com/apps/${config.APP_SLUG}/installations/new`}
-      target="_blank"
-      rel="noreferrer noopener"
-      className="underline"
-    >
-      Add more repositories...
-    </a>
-  ) : (
-    "No results found."
-  );
+  const emptyContent = "No results found.";
 
   return (
     <Autocomplete
@@ -66,6 +57,7 @@ export function GitHubRepositorySelector({
       name="repo"
       aria-label="GitHub Repository"
       placeholder="Select a GitHub project"
+      isVirtualized={false}
       selectedKey={selectedKey}
       inputProps={{
         classNames: {
@@ -74,20 +66,63 @@ export function GitHubRepositorySelector({
         },
       }}
       onSelectionChange={(id) => handleRepoSelection(id?.toString() ?? null)}
+      onInputChange={onInputChange}
       clearButtonProps={{ onClick: handleClearSelection }}
       listboxProps={{
         emptyContent,
       }}
+      defaultFilter={(textValue, inputValue) =>
+        !inputValue ||
+        sanitizeQuery(textValue).includes(sanitizeQuery(inputValue))
+      }
     >
-      {finalRepositories.map((repo) => (
-        <AutocompleteItem
-          data-testid="github-repo-item"
-          key={repo.id}
-          value={repo.id}
-        >
-          {repo.full_name}
-        </AutocompleteItem>
-      ))}
+      {config?.APP_MODE === "saas" &&
+        config?.APP_SLUG &&
+        ((
+          <AutocompleteItem key="install">
+            <a
+              href={`https://github.com/apps/${config.APP_SLUG}/installations/new`}
+              target="_blank"
+              rel="noreferrer noopener"
+              onClick={(e) => e.stopPropagation()}
+            >
+              Add more repositories...
+            </a>
+          </AutocompleteItem> // eslint-disable-next-line @typescript-eslint/no-explicit-any
+        ) as any)}
+      {userRepositories.length > 0 && (
+        <AutocompleteSection showDivider title="Your Repos">
+          {userRepositories.map((repo) => (
+            <AutocompleteItem
+              data-testid="github-repo-item"
+              key={repo.id}
+              value={repo.id}
+              className="data-[selected=true]:bg-default-100"
+              textValue={repo.full_name}
+            >
+              {repo.full_name}
+            </AutocompleteItem>
+          ))}
+        </AutocompleteSection>
+      )}
+      {publicRepositories.length > 0 && (
+        <AutocompleteSection showDivider title="Public Repos">
+          {publicRepositories.map((repo) => (
+            <AutocompleteItem
+              data-testid="github-repo-item"
+              key={repo.id}
+              value={repo.id}
+              className="data-[selected=true]:bg-default-100"
+              textValue={repo.full_name}
+            >
+              {repo.full_name}
+              <span className="ml-1 text-gray-400">
+                ({repo.stargazers_count || 0}⭐)
+              </span>
+            </AutocompleteItem>
+          ))}
+        </AutocompleteSection>
+      )}
     </Autocomplete>
   );
 }

frontend/src/components/features/github/github-repositories-suggestion-box.tsx

@@ -6,22 +6,39 @@ import { ModalButton } from "#/components/shared/buttons/modal-button";
 import { ConnectToGitHubModal } from "#/components/shared/modals/connect-to-github-modal";
 import { ModalBackdrop } from "#/components/shared/modals/modal-backdrop";
 import { isGitHubErrorReponse } from "#/api/github-axios-instance";
+import { useAppRepositories } from "#/hooks/query/use-app-repositories";
+import { useSearchRepositories } from "#/hooks/query/use-search-repositories";
+import { useUserRepositories } from "#/hooks/query/use-user-repositories";
+import { sanitizeQuery } from "#/utils/sanitize-query";
+import { useDebounce } from "#/hooks/use-debounce";
 
 interface GitHubRepositoriesSuggestionBoxProps {
   handleSubmit: () => void;
-  repositories: GitHubRepository[];
   gitHubAuthUrl: string | null;
   user: GitHubErrorReponse | GitHubUser | null;
 }
 
 export function GitHubRepositoriesSuggestionBox({
   handleSubmit,
-  repositories,
   gitHubAuthUrl,
   user,
 }: GitHubRepositoriesSuggestionBoxProps) {
   const [connectToGitHubModalOpen, setConnectToGitHubModalOpen] =
     React.useState(false);
+  const [searchQuery, setSearchQuery] = React.useState<string>("");
+  const debouncedSearchQuery = useDebounce(searchQuery, 300);
+
+  // TODO: Use `useQueries` to fetch all repositories in parallel
+  const { data: appRepositories } = useAppRepositories();
+  const { data: userRepositories } = useUserRepositories();
+  const { data: searchedRepos } = useSearchRepositories(
+    sanitizeQuery(debouncedSearchQuery),
+  );
+
+  const repositories =
+    userRepositories?.pages.flatMap((page) => page.data) ||
+    appRepositories?.pages.flatMap((page) => page.data) ||
+    [];
 
   const handleConnectToGitHub = () => {
     if (gitHubAuthUrl) {
@@ -40,8 +57,10 @@ export function GitHubRepositoriesSuggestionBox({
         content={
           isLoggedIn ? (
             <GitHubRepositorySelector
+              onInputChange={setSearchQuery}
               onSelect={handleSubmit}
-              repositories={repositories}
+              publicRepositories={searchedRepos}
+              userRepositories={repositories}
             />
           ) : (
             <ModalButton

frontend/src/components/shared/modals/settings/settings-modal.tsx

@@ -16,7 +16,7 @@ export function SettingsModal({ onClose, settings }: SettingsModalProps) {
     <ModalBackdrop onClose={onClose}>
       <div
         data-testid="ai-config-modal"
-        className="bg-root-primary w-[384px] p-6 rounded-xl flex flex-col gap-2"
+        className="bg-root-primary min-w-[384px] max-w-[700px] p-6 rounded-xl flex flex-col gap-2"
       >
         {aiConfigOptions.error && (
           <p className="text-danger text-xs">{aiConfigOptions.error.message}</p>

frontend/src/context/settings-up-to-date-context.tsx

@@ -25,9 +25,7 @@ export function SettingsUpToDateProvider({
   );
 
   return (
-    <SettingsUpToDateContext.Provider value={value}>
-      {children}
-    </SettingsUpToDateContext.Provider>
+    <SettingsUpToDateContext value={value}>{children}</SettingsUpToDateContext>
   );
 }
 

frontend/src/context/ws-client-provider.tsx

@@ -1,13 +1,24 @@
 import posthog from "posthog-js";
 import React from "react";
 import { io, Socket } from "socket.io-client";
-
 import EventLogger from "#/utils/event-logger";
 import { handleAssistantMessage } from "#/services/actions";
 import { useRate } from "#/hooks/use-rate";
+import { OpenHandsParsedEvent } from "#/types/core";
+import { AgentStateChangeObservation } from "#/types/core/observations";
 
-const isOpenHandsMessage = (event: Record<string, unknown>) =>
-  event.action === "message";
+const isOpenHandsMessage = (event: unknown): event is OpenHandsParsedEvent =>
+  typeof event === "object" &&
+  event !== null &&
+  "id" in event &&
+  "source" in event &&
+  "message" in event &&
+  "timestamp" in event;
+
+const isAgentStateChangeObservation = (
+  event: OpenHandsParsedEvent,
+): event is AgentStateChangeObservation =>
+  "observation" in event && event.observation === "agent_state_changed";
 
 export enum WsClientProviderStatus {
   CONNECTED,
@@ -63,7 +74,7 @@ export function WsClientProvider({
   }
 
   function handleMessage(event: Record<string, unknown>) {
-    if (isOpenHandsMessage(event)) {
+    if (isOpenHandsMessage(event) && !isAgentStateChangeObservation(event)) {
       messageRateHandler.record(new Date().getTime());
     }
     setEvents((prevEvents) => [...prevEvents, event]);

frontend/src/hooks/query/use-app-repositories.ts

@@ -27,7 +27,7 @@ export const useAppRepositories = () => {
         installationIndex || 0,
         installations,
         repoPage || 1,
-        30,
+        1000,
       );
     },
     initialPageParam: { installationIndex: 0, repoPage: 1 },

frontend/src/hooks/query/use-conversation-config.ts

@@ -17,7 +17,7 @@ export const useConversationConfig = () => {
       if (!conversationId) throw new Error("No conversation ID");
       return OpenHands.getRuntimeId(conversationId);
     },
-    enabled: status === WsClientProviderStatus.CONNECTED && !!conversationId,
+    enabled: status !== WsClientProviderStatus.DISCONNECTED && !!conversationId,
   });
 
   React.useEffect(() => {

frontend/src/hooks/query/use-list-files.ts

@@ -1,10 +1,9 @@
 import { useQuery } from "@tanstack/react-query";
-import {
-  useWsClient,
-  WsClientProviderStatus,
-} from "#/context/ws-client-provider";
+import { useSelector } from "react-redux";
 import OpenHands from "#/api/open-hands";
 import { useConversation } from "#/context/conversation-context";
+import { RootState } from "#/store";
+import { RUNTIME_INACTIVE_STATES } from "#/types/agent-state";
 
 interface UseListFilesConfig {
   path?: string;
@@ -17,8 +16,8 @@ const DEFAULT_CONFIG: UseListFilesConfig = {
 
 export const useListFiles = (config: UseListFilesConfig = DEFAULT_CONFIG) => {
   const { conversationId } = useConversation();
-  const { status } = useWsClient();
-  const isActive = status === WsClientProviderStatus.CONNECTED;
+  const { curAgentState } = useSelector((state: RootState) => state.agent);
+  const isActive = !RUNTIME_INACTIVE_STATES.includes(curAgentState);
 
   return useQuery({
     queryKey: ["files", conversationId, config?.path],

frontend/src/hooks/query/use-search-repositories.ts

@@ -0,0 +1,12 @@
+import { useQuery } from "@tanstack/react-query";
+import { searchPublicRepositories } from "#/api/github";
+
+export function useSearchRepositories(query: string) {
+  return useQuery({
+    queryKey: ["repositories", query],
+    queryFn: () => searchPublicRepositories(query, 3),
+    enabled: !!query,
+    select: (data) => data.map((repo) => ({ ...repo, is_public: true })),
+    initialData: [],
+  });
+}

frontend/src/hooks/query/use-user-repositories.ts

@@ -11,7 +11,7 @@ export const useUserRepositories = () => {
   const repos = useInfiniteQuery({
     queryKey: ["repositories", gitHubToken],
     queryFn: async ({ pageParam }) =>
-      retrieveGitHubUserRepositories(pageParam, 100),
+      retrieveGitHubUserRepositories(pageParam, 1000),
     initialPageParam: 1,
     getNextPageParam: (lastPage) => lastPage.nextPage,
     enabled: !!gitHubToken && config?.APP_MODE === "oss",

frontend/src/hooks/use-click-outside-element.ts

@@ -17,6 +17,7 @@ export const useClickOutsideElement = <T extends HTMLElement>(
     };
 
     document.addEventListener("click", handleClickOutside);
+
     return () => document.removeEventListener("click", handleClickOutside);
   }, []);
 

frontend/src/hooks/use-debounce.ts

@@ -0,0 +1,12 @@
+import { useEffect, useState } from "react";
+
+export function useDebounce<T>(value: T, delay: number): T {
+  const [debouncedValue, setDebouncedValue] = useState<T>(value);
+
+  useEffect(() => {
+    const timer = setTimeout(() => setDebouncedValue(value), delay);
+    return () => clearTimeout(timer);
+  }, [value, delay]);
+
+  return debouncedValue;
+}

frontend/src/i18n/translation.json

@@ -1196,6 +1196,20 @@
     "fr": "L'agent attend l'entrée de l'utilisateur...",
     "tr": "Ajan kullanıcı girdisini bekliyor..."
   },
+  "CHAT_INTERFACE$AGENT_RATE_LIMITED_MESSAGE": {
+  "en": "Agent is Rate Limited",
+  "zh-CN": "智能体已达到速率限制",
+  "zh-TW": "智能體已達到速率限制",
+  "de": "Agent ist ratenbegrenzt",
+  "ko-KR": "에이전트가 속도 제한되었습니다",
+  "no": "Agenten er hastighetsbegrenset",
+  "it": "L'agente è limitato dalla frequenza",
+  "pt": "O agente está com limite de taxa",
+  "es": "El agente está limitado por tasa",
+  "ar": "الوكيل مقيد بحد السرعة",
+  "fr": "L'agent est limité en fréquence",
+  "tr": "Ajan hız sınırına ulaştı"
+  },
   "CHAT_INTERFACE$AGENT_PAUSED_MESSAGE": {
     "en": "Agent has paused.",
     "de": "Agent pausiert.",

frontend/src/routes/_oh._index/route.tsx

@@ -3,9 +3,6 @@ import { useDispatch } from "react-redux";
 import posthog from "posthog-js";
 import { setImportedProjectZip } from "#/state/initial-query-slice";
 import { convertZipToBase64 } from "#/utils/convert-zip-to-base64";
-import { useUserRepositories } from "#/hooks/query/use-user-repositories";
-import { useAppRepositories } from "#/hooks/query/use-app-repositories";
-
 import { useGitHubUser } from "#/hooks/query/use-github-user";
 import { useGitHubAuthUrl } from "#/hooks/use-github-auth-url";
 import { useConfig } from "#/hooks/query/use-config";
@@ -22,8 +19,6 @@ function Home() {
 
   const { data: config } = useConfig();
   const { data: user } = useGitHubUser();
-  const { data: appRepositories } = useAppRepositories();
-  const { data: userRepositories } = useUserRepositories();
 
   const gitHubAuthUrl = useGitHubAuthUrl({
     gitHubToken,
@@ -47,11 +42,6 @@ function Home() {
         <div className="flex gap-4 w-full flex-col md:flex-row">
           <GitHubRepositoriesSuggestionBox
             handleSubmit={() => formRef.current?.requestSubmit()}
-            repositories={
-              userRepositories?.pages.flatMap((page) => page.data) ||
-              appRepositories?.pages.flatMap((page) => page.data) ||
-              []
-            }
             gitHubAuthUrl={gitHubAuthUrl}
             user={user || null}
           />

frontend/src/routes/_oh.app/hooks/use-ws-status-change.ts

@@ -71,7 +71,7 @@ export const useWSStatusChange = () => {
     }
     statusRef.current = status;
 
-    if (status === WsClientProviderStatus.CONNECTED && initialQuery) {
+    if (status !== WsClientProviderStatus.DISCONNECTED && initialQuery) {
       dispatch(
         addUserMessage({
           content: initialQuery,

frontend/src/services/actions.ts

@@ -16,6 +16,7 @@ import {
   StatusMessage,
 } from "#/types/message";
 import { handleObservationMessage } from "./observations";
+import { appendInput } from "#/state/command-slice";
 
 const messageActions = {
   [ActionType.BROWSE]: (message: ActionMessage) => {
@@ -62,6 +63,10 @@ export function handleActionMessage(message: ActionMessage) {
     return;
   }
 
+  if (message.action === ActionType.RUN) {
+    store.dispatch(appendInput(message.args.command));
+  }
+
   if ("args" in message && "security_risk" in message.args) {
     store.dispatch(appendSecurityAnalyzerInput(message));
   }

frontend/src/services/observations.ts

@@ -80,8 +80,7 @@ export function handleObservationMessage(message: ObservationMessage) {
             observation: "run" as const,
             extras: {
               command: String(message.extras.command || ""),
-              command_id: Number(message.extras.command_id || 0),
-              exit_code: Number(message.extras.exit_code || 0),
+              metadata: message.extras.metadata,
               hidden: Boolean(message.extras.hidden),
             },
           }),

frontend/src/state/chat-slice.ts

@@ -93,7 +93,7 @@ export const chatSlice = createSlice({
       const translationID = `ACTION_MESSAGE$${actionID.toUpperCase()}`;
       let text = "";
       if (actionID === "run") {
-        text = `\`${action.payload.args.command}\``;
+        text = `Command:\n\`${action.payload.args.command}\``;
       } else if (actionID === "run_ipython") {
         text = `\`\`\`\n${action.payload.args.code}\n\`\`\``;
       } else if (actionID === "write") {
@@ -144,7 +144,7 @@ export const chatSlice = createSlice({
       // Set success property based on observation type
       if (observationID === "run") {
         const commandObs = observation.payload as CommandObservation;
-        causeMessage.success = commandObs.extras.exit_code === 0;
+        causeMessage.success = commandObs.extras.metadata.exit_code === 0;
       } else if (observationID === "run_ipython") {
         // For IPython, we consider it successful if there's no error message
         const ipythonObs = observation.payload as IPythonObservation;
@@ -158,7 +158,9 @@ export const chatSlice = createSlice({
         if (content.length > MAX_CONTENT_LENGTH) {
           content = `${content.slice(0, MAX_CONTENT_LENGTH)}...`;
         }
-        content = `\`\`\`\n${content}\n\`\`\``;
+        content = `${
+          causeMessage.content
+        }\n\nOutput:\n\`\`\`\n${content.trim() || "[Command finished execution with no output]"}\n\`\`\``;
         causeMessage.content = content; // Observation content includes the action
       } else if (observationID === "read" || observationID === "edit") {
         const { content } = observation.payload;

frontend/src/types/agent-state.tsx

@@ -8,6 +8,7 @@ export enum AgentState {
   FINISHED = "finished",
   REJECTED = "rejected",
   ERROR = "error",
+  RATE_LIMITED = "rate_limited",
   AWAITING_USER_CONFIRMATION = "awaiting_user_confirmation",
   USER_CONFIRMED = "user_confirmed",
   USER_REJECTED = "user_rejected",

frontend/src/types/core/observations.ts

@@ -13,9 +13,8 @@ export interface CommandObservation extends OpenHandsObservationEvent<"run"> {
   source: "agent";
   extras: {
     command: string;
-    command_id: number;
-    exit_code: number;
     hidden?: boolean;
+    metadata: Record<string, unknown>;
   };
 }
 

frontend/src/types/github.d.ts

@@ -16,6 +16,7 @@ interface GitHubUser {
 interface GitHubRepository {
   id: number;
   full_name: string;
+  stargazers_count?: number;
 }
 
 interface GitHubAppRepository {

frontend/src/types/message.tsx

@@ -27,8 +27,11 @@ export interface ObservationMessage {
   // The observed data
   content: string;
 
-  // Additional structured data
-  extras: Record<string, string>;
+  extras: {
+    metadata: Record<string, unknown>;
+    error_id: string;
+    [key: string]: string | Record<string, unknown>;
+  };
 
   // A friendly message that can be put in the chat log
   message: string;

frontend/src/utils/sanitize-query.ts

@@ -0,0 +1,7 @@
+export const sanitizeQuery = (query: string) =>
+  query
+    .trim()
+    .replace(/https?:\/\//, "")
+    .replace(/github.com\//, "")
+    .replace(/\.git$/, "")
+    .toLowerCase();

frontend/test-utils.tsx

@@ -68,7 +68,13 @@ export function renderWithProviders(
       <Provider store={store}>
         <AuthProvider>
           <SettingsUpToDateProvider>
-            <QueryClientProvider client={new QueryClient()}>
+            <QueryClientProvider
+              client={
+                new QueryClient({
+                  defaultOptions: { queries: { retry: false } },
+                })
+              }
+            >
               <ConversationProvider>
                 <I18nextProvider i18n={i18n}>{children}</I18nextProvider>
               </ConversationProvider>

microagents/README.md

@@ -0,0 +1,163 @@
+# OpenHands MicroAgents
+
+MicroAgents are specialized prompts that enhance OpenHands with domain-specific knowledge and task-specific workflows. They help developers by providing expert guidance, automating common tasks, and ensuring consistent practices across projects. Each microagent is designed to excel in a specific area, from Git operations to code review processes.
+
+## Sources of Microagents
+
+OpenHands loads microagents from two sources:
+
+### 1. Shareable Microagents (Public)
+This directory (`OpenHands/microagents/`) contains shareable microagents that are:
+- Available to all OpenHands users
+- Maintained in the OpenHands repository
+- Perfect for reusable knowledge and common workflows
+
+Directory structure:
+```
+OpenHands/microagents/
+├── knowledge/     # Keyword-triggered expertise
+│   ├── git.md      # Git operations
+│   ├── testing.md  # Testing practices
+│   └── docker.md   # Docker guidelines
+└── tasks/        # Interactive workflows
+    ├── pr_review.md   # PR review process
+    ├── bug_fix.md     # Bug fixing workflow
+    └── feature.md     # Feature implementation
+```
+
+### 2. Repository Instructions (Private)
+Each repository can have its own instructions in `.openhands/microagents/repo.md`. These instructions are:
+- Private to that repository
+- Automatically loaded when working with that repository
+- Perfect for repository-specific guidelines and team practices
+
+Example repository structure:
+```
+your-repository/
+└── .openhands/
+    └── microagents/
+        └── repo.md    # Repository-specific instructions
+        └── knowledges/  # Private micro-agents that are only available inside this repo
+        └── tasks/       # Private micro-agents that are only available inside this repo
+```
+
+
+## Loading Order
+
+When OpenHands works with a repository, it:
+1. Loads repository-specific instructions from `.openhands/microagents/repo.md` if present
+2. Loads relevant knowledge agents based on keywords in conversations
+3. Enable task agent if user select one of them
+
+## Types of MicroAgents
+
+All microagents use markdown files with YAML frontmatter.
+
+
+### 1. Knowledge Agents
+
+Knowledge agents provide specialized expertise that's triggered by keywords in conversations. They help with:
+- Language best practices
+- Framework guidelines
+- Common patterns
+- Tool usage
+
+Key characteristics:
+- **Trigger-based**: Activated by specific keywords in conversations
+- **Context-aware**: Provide relevant advice based on file types and content
+- **Reusable**: Knowledge can be applied across multiple projects
+- **Versioned**: Support multiple versions of tools/frameworks
+
+You can see an example of a knowledge-based agent in [OpenHands's github microagent](https://github.com/All-Hands-AI/OpenHands/tree/main/microagents/knowledge/github.md).
+
+### 2. Repository Agents
+
+Repository agents provide repository-specific knowledge and guidelines. They are:
+- Loaded from `.openhands/microagents/repo.md`
+- Specific to individual repositories
+- Automatically activated for their repository
+- Perfect for team practices and project conventions
+
+Key features:
+- **Project-specific**: Contains guidelines unique to the repository
+- **Team-focused**: Enforces team conventions and practices
+- **Always active**: Automatically loaded for the repository
+- **Locally maintained**: Updated with the project
+
+You can see an example of a repo agent in [the agent for the OpenHands repo itself](https://github.com/All-Hands-AI/OpenHands/blob/main/.openhands/microagents/repo.md).
+
+### 3. Task Agents
+
+Task agents provide interactive workflows that guide users through common development tasks. They:
+- Accept user inputs
+- Follow predefined steps
+- Adapt to context
+- Provide consistent results
+
+Key capabilities:
+- **Interactive**: Guide users through complex processes
+- **Validating**: Check inputs and conditions
+- **Flexible**: Adapt to different scenarios
+- **Reproducible**: Ensure consistent outcomes
+
+Example workflow:
+You can see an example of a task-based agent in [OpenHands's pull request updating microagent](https://github.com/All-Hands-AI/OpenHands/tree/main/microagents/tasks/update_pr_description.md).
+
+## Contributing
+
+### When to Contribute
+
+1. **Knowledge Agents** - When you have:
+   - Language/framework best practices
+   - Tool usage patterns
+   - Common problem solutions
+   - General development guidelines
+
+2. **Task Agents** - When you have:
+   - Repeatable workflows
+   - Multi-step processes
+   - Common development tasks
+   - Standard procedures
+
+3. **Repository Agents** - When you need:
+   - Project-specific guidelines
+   - Team conventions and practices
+   - Custom workflow documentation
+   - Repository-specific setup instructions
+
+### Best Practices
+
+1. **For Knowledge Agents**:
+   - Choose distinctive triggers
+   - Focus on one area of expertise
+   - Include practical examples
+   - Use file patterns when relevant
+   - Keep knowledge general and reusable
+
+2. **For Task Agents**:
+   - Break workflows into clear steps
+   - Validate user inputs
+   - Provide helpful defaults
+   - Include usage examples
+   - Make steps adaptable
+
+3. **For Repository Agents**:
+   - Document clear setup instructions
+   - Include repository structure details
+   - Specify testing and build procedures
+   - List environment requirements
+   - Maintain up-to-date team practices
+
+### Submission Process
+
+1. Create your agent file in the appropriate directory:
+   - `knowledge/` for expertise (public, shareable)
+   - `tasks/` for workflows (public, shareable)
+   - Note: Repository agents should remain in their respective repositories' `.openhands/microagents/` directory
+2. Test thoroughly
+3. Submit a pull request to OpenHands
+
+
+## License
+
+All microagents are subject to the same license as OpenHands. See the root LICENSE file for details.

microagents/knowledge/flarglebargle.md

@@ -1,5 +1,7 @@
 ---
 name: flarglebargle
+type: knowledge
+version: 1.0.0
 agent: CodeActAgent
 triggers:
 - flarglebargle

microagents/knowledge/github.md

@@ -1,5 +1,7 @@
 ---
 name: github
+type: knowledge
+version: 1.0.0
 agent: CodeActAgent
 triggers:
 - github
@@ -17,8 +19,10 @@ Here are some instructions for pushing, but ONLY do this if the user asks you to
 * Git config (username and email) is pre-set. Do not modify.
 * You may already be on a branch starting with `openhands-workspace`. Create a new branch with a better name before pushing.
 * Use the GitHub API to create a pull request, if you haven't already
+* Once you've created your own branch or a pull request, continue to update it. Do NOT create a new one unless you are explicitly asked to. Update the PR title and description as necessary, but don't change the branch name.
 * Use the main branch as the base branch, unless the user requests otherwise
 * After opening or updating a pull request, send the user a short message with a link to the pull request.
+* Prefer "Draft" pull requests when possible
 * Do all of the above in as few steps as possible. E.g. you could open a PR with one step by running the following bash commands:
 ```bash
 git remote -v && git branch # to find the current org, repo and branch

microagents/knowledge/npm.md

@@ -1,5 +1,7 @@
 ---
 name: npm
+type: knowledge
+version: 1.0.0
 agent: CodeActAgent
 triggers:
 - npm

microagents/tasks/address_pr_comments.md

@@ -0,0 +1,20 @@
+---
+name: address_pr_comments
+type: task
+version: 1.0.0
+author: openhands
+agent: CodeActAgent
+inputs:
+  - name: PR_URL
+    description: "URL of the pull request"
+    required: true
+  - name: BRANCH_NAME
+    description: "Branch name corresponds to the pull request"
+    required: true
+---
+
+First, check the branch {{ BRANCH_NAME }} and read the diff against the main branch to understand the purpose.
+
+This branch corresponds to this PR {{ PR_URL }}
+
+Next, you should use the GitHub API to read the reviews and comments on this PR and address them.