[Feat]: Custom secrets plumbing for BE (#7891)

Co-authored-by: openhands <openhands@all-hands.dev>

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

@@ -1,53 +0,0 @@
-# Workflow that uses the DummyAgent to run a simple task
-name: Run E2E test with dummy agent
-
-# Always run on "main"
-# Always run on PRs
-on:
-  push:
-    branches:
-    - main
-  pull_request:
-
-# If triggered by a PR, it will be in the same group. However, each commit on main will be in its own unique group
-concurrency:
-  group: ${{ github.workflow }}-${{ (github.head_ref && github.ref) || github.run_id }}
-  cancel-in-progress: true
-
-jobs:
-  test:
-    runs-on: blacksmith-4vcpu-ubuntu-2204
-    steps:
-      - uses: actions/checkout@v4
-      - 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: Setup Node.js
-        uses: useblacksmith/setup-node@v5
-        with:
-          node-version: '22.x'
-      - name: Install poetry via pipx
-        run: pipx install poetry
-      - name: Set up Python
-        uses: useblacksmith/setup-python@v6
-        with:
-          python-version: '3.12'
-          cache: 'poetry'
-      - name: Install Python dependencies using Poetry
-        run: poetry install --without evaluation
-      - name: Build Environment
-        run: make build
-      - name: Run tests
-        run: |
-          set -e
-          SANDBOX_FORCE_REBUILD_RUNTIME=True poetry run python3 openhands/core/main.py -t "do a flip" -d ./workspace/ -c DummyAgent
-      - name: Check exit code
-        run: |
-          if [ $? -ne 0 ]; then
-            echo "Test failed"
-            exit 1
-          else
-            echo "Test passed"
-          fi

.github/workflows/openhands-resolver.yml

@@ -179,7 +179,7 @@ jobs:
 
           echo "MAX_ITERATIONS=${{ inputs.max_iterations || 50 }}" >> $GITHUB_ENV
           echo "SANDBOX_ENV_GITHUB_TOKEN=${{ secrets.PAT_TOKEN || github.token }}" >> $GITHUB_ENV
-          echo "SANDBOX_ENV_BASE_CONTAINER_IMAGE=${{ inputs.base_container_image }}" >> $GITHUB_ENV
+          echo "SANDBOX_BASE_CONTAINER_IMAGE=${{ inputs.base_container_image }}" >> $GITHUB_ENV
 
           # Set branch variables
           echo "TARGET_BRANCH=${{ inputs.target_branch || 'main' }}" >> $GITHUB_ENV

.openhands/microagents/repo.md

@@ -54,3 +54,20 @@ Frontend:
 ## Template for Github Pull Request
 
 If you are starting a pull request (PR), please follow the template in `.github/pull_request_template.md`.
+
+## Implementation Details
+
+These details may or may not be useful for your current task.
+
+### Frontend
+
+#### Action Handling:
+- Actions are defined in `frontend/src/types/action-type.ts`
+- The `HANDLED_ACTIONS` array in `frontend/src/state/chat-slice.ts` determines which actions are displayed as collapsible UI elements
+- To add a new action type to the UI:
+  1. Add the action type to the `HANDLED_ACTIONS` array
+  2. Implement the action handling in `addAssistantAction` function in chat-slice.ts
+  3. Add a translation key in the format `ACTION_MESSAGE$ACTION_NAME` to the i18n files
+- Actions with `thought` property are displayed in the UI based on their action type:
+  - Regular actions (like "run", "edit") display the thought as a separate message
+  - Special actions (like "think") are displayed as collapsible elements only

Development.md

@@ -118,7 +118,7 @@ poetry run pytest ./tests/unit/test_*.py
 To reduce build time (e.g., if no changes were made to the client-runtime component), you can use an existing Docker container image by
 setting the SANDBOX_RUNTIME_CONTAINER_IMAGE environment variable to the desired Docker image.
 
-Example: `export SANDBOX_RUNTIME_CONTAINER_IMAGE=ghcr.io/all-hands-ai/runtime:0.32-nikolaik`
+Example: `export SANDBOX_RUNTIME_CONTAINER_IMAGE=ghcr.io/all-hands-ai/runtime:0.34-nikolaik`
 
 ## Develop inside Docker container
 

Makefile

@@ -175,7 +175,7 @@ install-pre-commit-hooks:
 
 lint-backend:
 	@echo "$(YELLOW)Running linters...$(RESET)"
-	@poetry run pre-commit run --files openhands/**/* agenthub/**/* evaluation/**/* --show-diff-on-failure --config $(PRE_COMMIT_CONFIG_PATH)
+	@poetry run pre-commit run --files openhands/**/* evaluation/**/* tests/**/* --show-diff-on-failure --config $(PRE_COMMIT_CONFIG_PATH)
 
 lint-frontend:
 	@echo "$(YELLOW)Running linters for frontend...$(RESET)"

README.md

@@ -18,7 +18,7 @@
   <br/>
   <a href="https://docs.all-hands.dev/modules/usage/getting-started"><img src="https://img.shields.io/badge/Documentation-000?logo=googledocs&logoColor=FFE165&style=for-the-badge" alt="Check out the documentation"></a>
   <a href="https://arxiv.org/abs/2407.16741"><img src="https://img.shields.io/badge/Paper%20on%20Arxiv-000?logoColor=FFE165&logo=arxiv&style=for-the-badge" alt="Paper on Arxiv"></a>
-  <a href="https://huggingface.co/spaces/OpenHands/evaluation"><img src="https://img.shields.io/badge/Benchmark%20score-000?logoColor=FFE165&logo=huggingface&style=for-the-badge" alt="Evaluation Benchmark Score"></a>
+  <a href="https://docs.google.com/spreadsheets/d/1wOUdFCMyY6Nt0AIqF705KN4JKOWgeI4wUGUP60krXXs/edit?gid=0#gid=0"><img src="https://img.shields.io/badge/Benchmark%20score-000?logoColor=FFE165&logo=huggingface&style=for-the-badge" alt="Evaluation Benchmark Score"></a>
   <hr>
 </div>
 
@@ -52,17 +52,17 @@ system requirements and more information.
 
 
 ```bash
-docker pull docker.all-hands.dev/all-hands-ai/runtime:0.32-nikolaik
+docker pull docker.all-hands.dev/all-hands-ai/runtime:0.34-nikolaik
 
 docker run -it --rm --pull=always \
-    -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.all-hands.dev/all-hands-ai/runtime:0.32-nikolaik \
+    -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.all-hands.dev/all-hands-ai/runtime:0.34-nikolaik \
     -e LOG_ALL_EVENTS=true \
     -v /var/run/docker.sock:/var/run/docker.sock \
     -v ~/.openhands-state:/.openhands-state \
     -p 3000:3000 \
     --add-host host.docker.internal:host-gateway \
     --name openhands-app \
-    docker.all-hands.dev/all-hands-ai/openhands:0.32
+    docker.all-hands.dev/all-hands-ai/openhands:0.34
 ```
 
 You'll find OpenHands running at [http://localhost:3000](http://localhost:3000)!

config.template.toml

@@ -221,9 +221,22 @@ enable_browsing = true
 # Whether the LLM draft editor is enabled
 enable_llm_editor = false
 
+# Whether the standard editor tool (str_replace_editor) is enabled
+# Only has an effect if enable_llm_editor is False
+enable_editor = true
+
 # Whether the IPython tool is enabled
 enable_jupyter = true
 
+# Whether the command tool is enabled
+enable_cmd = true
+
+# Whether the think tool is enabled
+enable_think = true
+
+# Whether the finish tool is enabled
+enable_finish = true
+
 # LLM config group to use
 #llm_config = 'your-llm-config-group'
 

containers/dev/compose.yml

@@ -11,7 +11,7 @@ services:
       - BACKEND_HOST=${BACKEND_HOST:-"0.0.0.0"}
       - SANDBOX_API_HOSTNAME=host.docker.internal
       #
-      - SANDBOX_RUNTIME_CONTAINER_IMAGE=${SANDBOX_RUNTIME_CONTAINER_IMAGE:-ghcr.io/all-hands-ai/runtime:0.32-nikolaik}
+      - SANDBOX_RUNTIME_CONTAINER_IMAGE=${SANDBOX_RUNTIME_CONTAINER_IMAGE:-ghcr.io/all-hands-ai/runtime:0.34-nikolaik}
       - SANDBOX_USER_ID=${SANDBOX_USER_ID:-1234}
       - WORKSPACE_MOUNT_PATH=${WORKSPACE_BASE:-$PWD/workspace}
     ports:

docker-compose.yml

@@ -7,7 +7,7 @@ services:
     image: openhands:latest
     container_name: openhands-app-${DATE:-}
     environment:
-      - SANDBOX_RUNTIME_CONTAINER_IMAGE=${SANDBOX_RUNTIME_CONTAINER_IMAGE:-docker.all-hands.dev/all-hands-ai/runtime:0.32-nikolaik}
+      - SANDBOX_RUNTIME_CONTAINER_IMAGE=${SANDBOX_RUNTIME_CONTAINER_IMAGE:-docker.all-hands.dev/all-hands-ai/runtime:0.34-nikolaik}
       #- SANDBOX_USER_ID=${SANDBOX_USER_ID:-1234} # enable this only if you want a specific non-root sandbox user but you will have to manually adjust permissions of openhands-state for this user
       - WORKSPACE_MOUNT_PATH=${WORKSPACE_BASE:-$PWD/workspace}
     ports:

docs/i18n/fr/docusaurus-plugin-content-docs/current/usage/how-to/cli-mode.md

@@ -52,7 +52,7 @@ LLM_API_KEY="sk_test_12345"
 ```bash
 docker run -it \
     --pull=always \
-    -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.all-hands.dev/all-hands-ai/runtime:0.32-nikolaik \
+    -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.all-hands.dev/all-hands-ai/runtime:0.34-nikolaik \
     -e SANDBOX_USER_ID=$(id -u) \
     -e WORKSPACE_MOUNT_PATH=$WORKSPACE_BASE \
     -e LLM_API_KEY=$LLM_API_KEY \
@@ -61,7 +61,7 @@ docker run -it \
     -v /var/run/docker.sock:/var/run/docker.sock \
     --add-host host.docker.internal:host-gateway \
     --name openhands-app-$(date +%Y%m%d%H%M%S) \
-    docker.all-hands.dev/all-hands-ai/openhands:0.32 \
+    docker.all-hands.dev/all-hands-ai/openhands:0.34 \
     python -m openhands.core.cli
 ```
 

docs/i18n/fr/docusaurus-plugin-content-docs/current/usage/how-to/gui-mode.md

@@ -21,14 +21,18 @@ OpenHands fournit un mode Interface Graphique (GUI) convivial pour interagir ave
 3. Entrez la `Clé API` correspondante pour le fournisseur choisi.
 4. Cliquez sur "Enregistrer" pour appliquer les paramètres.
 
-### Configuration du Jeton GitHub
+### Jetons de Contrôle de Version
+
+OpenHands prend en charge plusieurs fournisseurs de contrôle de version. Vous pouvez configurer des jetons pour plusieurs fournisseurs simultanément.
+
+#### Configuration du Jeton GitHub
 
 OpenHands exporte automatiquement un `GITHUB_TOKEN` vers l'environnement shell s'il est disponible. Cela peut se produire de deux manières :
 
 1. **Localement (OSS)** : L'utilisateur saisit directement son jeton GitHub
 2. **En ligne (SaaS)** : Le jeton est obtenu via l'authentification OAuth GitHub
 
-#### Configuration d'un Jeton GitHub Local
+##### Configuration d'un Jeton GitHub Local
 
 1. **Générer un Personal Access Token (PAT)** :
    - Allez dans Paramètres GitHub > Paramètres développeur > Personal Access Tokens > Tokens (classique)
@@ -40,11 +44,11 @@ OpenHands exporte automatiquement un `GITHUB_TOKEN` vers l'environnement shell s
 
 2. **Entrer le Jeton dans OpenHands** :
    - Cliquez sur le bouton Paramètres (icône d'engrenage) en haut à droite
-   - Accédez à la section "GitHub"
+   - Accédez à la section "Git Provider Settings"
    - Collez votre jeton dans le champ "Jeton GitHub"
    - Cliquez sur "Enregistrer" pour appliquer les modifications
 
-#### Politiques de Jetons Organisationnels
+##### Politiques de Jetons Organisationnels
 
 Si vous travaillez avec des dépôts organisationnels, une configuration supplémentaire peut être nécessaire :
 
@@ -59,7 +63,7 @@ Si vous travaillez avec des dépôts organisationnels, une configuration supplé
    - Si nécessaire, cliquez sur "Activer SSO" à côté de votre organisation
    - Terminez le processus d'autorisation SSO
 
-#### Authentification OAuth (Mode En Ligne)
+##### Authentification OAuth (Mode En Ligne)
 
 Lorsque vous utilisez OpenHands en mode en ligne, le flux OAuth GitHub :
 
@@ -74,7 +78,7 @@ Lorsque vous utilisez OpenHands en mode en ligne, le flux OAuth GitHub :
    - Autorisez OpenHands à accéder à votre compte GitHub
    - Si vous utilisez une organisation, autorisez l'accès à l'organisation si vous y êtes invité
 
-#### Dépannage
+##### Dépannage
 
 Problèmes courants et solutions :
 
@@ -95,6 +99,43 @@ Problèmes courants et solutions :
    - Vérifiez la console du navigateur pour tout message d'erreur
    - Utilisez le bouton "Tester la connexion" dans les paramètres s'il est disponible
 
+#### Configuration du Jeton GitLab
+
+OpenHands exporte automatiquement un `GITLAB_TOKEN` vers l'environnement shell, uniquement pour les installations locales, s'il est disponible.
+
+##### Configuration d'un Jeton GitLab
+
+1. **Générer un Personal Access Token (PAT)** :
+   - Sur GitLab, allez dans Paramètres utilisateur > Jetons d'accès
+   - Créez un nouveau jeton avec les portées suivantes :
+     - `api` (Accès API)
+     - `read_user` (Lecture des informations utilisateur)
+     - `read_repository` (Lecture du dépôt)
+     - `write_repository` (Écriture du dépôt)
+   - Définissez une date d'expiration ou laissez vide pour un jeton sans expiration
+
+2. **Entrer le Jeton dans OpenHands** :
+   - Cliquez sur le bouton Paramètres (icône d'engrenage)
+   - Accédez à la section `Git Provider Settings`
+   - Collez votre jeton dans le champ `Jeton GitLab`
+   - Si vous utilisez GitLab auto-hébergé, entrez l'URL de votre instance GitLab
+   - Cliquez sur `Enregistrer les modifications` pour appliquer les changements
+
+##### Dépannage
+
+Problèmes courants et solutions :
+
+1. **Jeton Non Reconnu** :
+   - Assurez-vous que le jeton est correctement enregistré dans les paramètres
+   - Vérifiez que le jeton n'a pas expiré
+   - Vérifiez que le jeton a les portées requises
+   - Pour les instances auto-hébergées, vérifiez l'URL correcte de l'instance
+
+2. **Accès Refusé** :
+   - Vérifiez les permissions d'accès au projet
+   - Vérifiez si le jeton possède les portées nécessaires
+   - Pour les dépôts de groupe/organisation, assurez-vous d'avoir les accès appropriés
+
 ### Paramètres Avancés
 
 1. Basculez sur `Options Avancées` pour accéder aux paramètres supplémentaires.

docs/i18n/fr/docusaurus-plugin-content-docs/current/usage/how-to/headless-mode.md

@@ -46,7 +46,7 @@ LLM_API_KEY="sk_test_12345"
 ```bash
 docker run -it \
     --pull=always \
-    -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.all-hands.dev/all-hands-ai/runtime:0.32-nikolaik \
+    -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.all-hands.dev/all-hands-ai/runtime:0.34-nikolaik \
     -e SANDBOX_USER_ID=$(id -u) \
     -e WORKSPACE_MOUNT_PATH=$WORKSPACE_BASE \
     -e LLM_API_KEY=$LLM_API_KEY \
@@ -56,6 +56,6 @@ docker run -it \
     -v /var/run/docker.sock:/var/run/docker.sock \
     --add-host host.docker.internal:host-gateway \
     --name openhands-app-$(date +%Y%m%d%H%M%S) \
-    docker.all-hands.dev/all-hands-ai/openhands:0.32 \
+    docker.all-hands.dev/all-hands-ai/openhands:0.34 \
     python -m openhands.core.main -t "write a bash script that prints hi" --no-auto-continue
 ```

docs/i18n/fr/docusaurus-plugin-content-docs/current/usage/installation.mdx

@@ -13,16 +13,16 @@
 La façon la plus simple d'exécuter OpenHands est avec Docker.
 
 ```bash
-docker pull docker.all-hands.dev/all-hands-ai/runtime:0.32-nikolaik
+docker pull docker.all-hands.dev/all-hands-ai/runtime:0.34-nikolaik
 
 docker run -it --rm --pull=always \
-    -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.all-hands.dev/all-hands-ai/runtime:0.32-nikolaik \
+    -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.all-hands.dev/all-hands-ai/runtime:0.34-nikolaik \
     -e LOG_ALL_EVENTS=true \
     -v /var/run/docker.sock:/var/run/docker.sock \
     -p 3000:3000 \
     --add-host host.docker.internal:host-gateway \
     --name openhands-app \
-    docker.all-hands.dev/all-hands-ai/openhands:0.32
+    docker.all-hands.dev/all-hands-ai/openhands:0.34
 ```
 
 Vous pouvez également exécuter OpenHands en mode [headless scriptable](https://docs.all-hands.dev/modules/usage/how-to/headless-mode), en tant que [CLI interactive](https://docs.all-hands.dev/modules/usage/how-to/cli-mode), ou en utilisant l'[Action GitHub OpenHands](https://docs.all-hands.dev/modules/usage/how-to/github-action).

docs/i18n/fr/docusaurus-plugin-content-docs/current/usage/runtimes.md

@@ -13,7 +13,7 @@ C'est le Runtime par défaut qui est utilisé lorsque vous démarrez OpenHands.
 
 ```
 docker run # ...
-    -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.all-hands.dev/all-hands-ai/runtime:0.32-nikolaik \
+    -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.all-hands.dev/all-hands-ai/runtime:0.34-nikolaik \
     -v /var/run/docker.sock:/var/run/docker.sock \
     # ...
 ```

docs/i18n/ja/docusaurus-plugin-content-docs/current/usage/how-to/cli-mode.md

@@ -34,7 +34,7 @@ Docker で OpenHands を CLI モードで実行するには:
 ```bash
 docker run -it \
     --pull=always \
-    -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.all-hands.dev/all-hands-ai/runtime:0.32-nikolaik \
+    -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.all-hands.dev/all-hands-ai/runtime:0.34-nikolaik \
     -e SANDBOX_USER_ID=$(id -u) \
     -e WORKSPACE_MOUNT_PATH=$WORKSPACE_BASE \
     -e LLM_API_KEY=$LLM_API_KEY \
@@ -44,7 +44,7 @@ docker run -it \
     -v ~/.openhands-state:/.openhands-state \
     --add-host host.docker.internal:host-gateway \
     --name openhands-app-$(date +%Y%m%d%H%M%S) \
-    docker.all-hands.dev/all-hands-ai/openhands:0.32 \
+    docker.all-hands.dev/all-hands-ai/openhands:0.34 \
     python -m openhands.core.cli
 ```
 

docs/i18n/ja/docusaurus-plugin-content-docs/current/usage/how-to/gui-mode.md

@@ -16,7 +16,11 @@ OpenHandsは、AI アシスタントとやり取りするためのグラフィ
 3. 選択したプロバイダーに対応する`API Key`を入力します。
 4. `Save Changes`をクリックして設定を適用します。
 
-### GitHubトークンの設定
+### バージョン管理トークン
+
+OpenHandsは複数のバージョン管理プロバイダーをサポートしています。複数のプロバイダーのトークンを同時に設定できます。
+
+#### GitHubトークンの設定
 
 OpenHandsは、利用可能な場合、自動的に`GITHUB_TOKEN`をシェル環境にエクスポートします。これは2つの方法で行われます。
 
@@ -34,7 +38,7 @@ OpenHandsは、利用可能な場合、自動的に`GITHUB_TOKEN`をシェル環
      - Minimal Permissions（検索用に**Meta Data = Read-only**を選択し、ブランチ作成用に**Pull Requests = Read and Write**、**Content = Read and Write**を選択します）
   2. **OpenHandsにトークンを入力**:
    - 設定ボタン（歯車アイコン）をクリックします。
-   - `GitHub Settings`セクションに移動します。
+   - `Git Provider Settings`セクションに移動します。
    - `GitHub Token`フィールドにトークンを貼り付けます。
    - `Save Changes`をクリックして変更を適用します。
 </details>
@@ -94,6 +98,46 @@ OpenHandsは、利用可能な場合、自動的に`GITHUB_TOKEN`をシェル環
    - 組織を使用している場合は、プロンプトが表示されたら組織へのアクセスを承認します。
 </details>
 
+#### GitLabトークンの設定
+
+OpenHandsは、利用可能な場合、ローカルインストールのみ、自動的に`GITLAB_TOKEN`をシェル環境にエクスポートします。
+
+<details>
+  <summary>GitLabトークンの設定</summary>
+
+  1. **Personal Access Token（PAT）の生成**:
+   - GitLabで、User Settings > Access Tokensに移動します。
+   - 以下のスコープを持つ新しいトークンを作成します:
+     - `api`（APIアクセス）
+     - `read_user`（ユーザー情報の読み取り）
+     - `read_repository`（リポジトリ読み取り）
+     - `write_repository`（リポジトリ書き込み）
+   - 有効期限を設定するか、無期限トークンの場合は空白のままにします。
+  2. **OpenHandsにトークンを入力**:
+   - 設定ボタン（歯車アイコン）をクリックします。
+   - `Git Provider Settings`セクションに移動します。
+   - `GitLab Token`フィールドにトークンを貼り付けます。
+   - セルフホスト型GitLabを使用している場合は、GitLabインスタンスのURLを入力します。
+   - `Save Changes`をクリックして変更を適用します。
+</details>
+
+<details>
+  <summary>トラブルシューティング</summary>
+
+  一般的な問題と解決策:
+
+  - **トークンが認識されない**:
+     - トークンが設定に正しく保存されていることを確認します。
+     - トークンの有効期限が切れていないことを確認します。
+     - トークンに必要なスコープがあることを確認します。
+     - セルフホスト型インスタンスの場合は、正しいインスタンスURLを確認します。
+
+  - **アクセスが拒否された**:
+     - プロジェクトのアクセス権限を確認します。
+     - トークンに必要なスコープがあるかどうかを確認します。
+     - グループ/組織のリポジトリの場合は、適切なアクセス権があることを確認します。
+</details>
+
 ### 高度な設定
 
 1. 設定ページ内で、`Advanced`オプションを切り替えて追加の設定にアクセスします。

docs/i18n/ja/docusaurus-plugin-content-docs/current/usage/how-to/headless-mode.md

@@ -31,7 +31,7 @@ DockerでOpenHandsをヘッドレスモードで実行するには:
 ```bash
 docker run -it \
     --pull=always \
-    -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.all-hands.dev/all-hands-ai/runtime:0.32-nikolaik \
+    -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.all-hands.dev/all-hands-ai/runtime:0.34-nikolaik \
     -e SANDBOX_USER_ID=$(id -u) \
     -e WORKSPACE_MOUNT_PATH=$WORKSPACE_BASE \
     -e LLM_API_KEY=$LLM_API_KEY \
@@ -42,7 +42,7 @@ docker run -it \
     -v ~/.openhands-state:/.openhands-state \
     --add-host host.docker.internal:host-gateway \
     --name openhands-app-$(date +%Y%m%d%H%M%S) \
-    docker.all-hands.dev/all-hands-ai/openhands:0.32 \
+    docker.all-hands.dev/all-hands-ai/openhands:0.34 \
     python -m openhands.core.main -t "write a bash script that prints hi"
 ```
 

docs/i18n/ja/docusaurus-plugin-content-docs/current/usage/runtimes/docker.md

@@ -25,7 +25,7 @@ nikolaik の `SANDBOX_RUNTIME_CONTAINER_IMAGE` は、ランタイムサーバー
 
     ```bash
     docker run # ...
-        -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.all-hands.dev/all-hands-ai/runtime:0.32-nikolaik \
+        -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.all-hands.dev/all-hands-ai/runtime:0.34-nikolaik \
         -e SANDBOX_USER_ID=$(id -u) \
         -e WORKSPACE_MOUNT_PATH=$WORKSPACE_BASE \
         -v $WORKSPACE_BASE:/opt/workspace_base \
@@ -82,5 +82,5 @@ docker network create openhands-network
 # 分離されたネットワークで OpenHands を実行
 docker run # ... \
     --network openhands-network \
-    docker.all-hands.dev/all-hands-ai/openhands:0.32
+    docker.all-hands.dev/all-hands-ai/openhands:0.34
 ```

docs/i18n/pt-BR/docusaurus-plugin-content-docs/current/usage/how-to/cli-mode.md

@@ -35,7 +35,7 @@ Para executar o OpenHands no modo CLI com Docker:
 ```bash
 docker run -it \
     --pull=always \
-    -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.all-hands.dev/all-hands-ai/runtime:0.32-nikolaik \
+    -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.all-hands.dev/all-hands-ai/runtime:0.34-nikolaik \
     -e SANDBOX_USER_ID=$(id -u) \
     -e WORKSPACE_MOUNT_PATH=$WORKSPACE_BASE \
     -e LLM_API_KEY=$LLM_API_KEY \
@@ -45,7 +45,7 @@ docker run -it \
     -v ~/.openhands-state:/.openhands-state \
     --add-host host.docker.internal:host-gateway \
     --name openhands-app-$(date +%Y%m%d%H%M%S) \
-    docker.all-hands.dev/all-hands-ai/openhands:0.32 \
+    docker.all-hands.dev/all-hands-ai/openhands:0.34 \
     python -m openhands.core.cli
 ```
 

docs/i18n/pt-BR/docusaurus-plugin-content-docs/current/usage/how-to/gui-mode.md

@@ -17,7 +17,11 @@ O OpenHands fornece um modo de Interface Gráfica do Usuário (GUI) para interag
 3. Insira a `Chave de API` correspondente para o provedor escolhido.
 4. Clique em `Salvar Alterações` para aplicar as configurações.
 
-### Configuração do Token do GitHub
+### Tokens de Controle de Versão
+
+O OpenHands suporta múltiplos provedores de controle de versão. Você pode configurar tokens para vários provedores simultaneamente.
+
+#### Configuração do Token do GitHub
 
 O OpenHands exporta automaticamente um `GITHUB_TOKEN` para o ambiente shell se ele estiver disponível. Isso pode acontecer de duas maneiras:
 
@@ -35,7 +39,7 @@ O OpenHands exporta automaticamente um `GITHUB_TOKEN` para o ambiente shell se e
      - Minimal Permissions (Selecione **Meta Data = Read-only** para pesquisa, **Pull Requests = Read and Write**, **Content = Read and Write** para criação de branches)
   2. **Insira o Token no OpenHands**:
    - Clique no botão Settings (ícone de engrenagem).
-   - Navegue até a seção `GitHub Settings`.
+   - Navegue até a seção `Git Provider Settings`.
    - Cole seu token no campo `GitHub Token`.
    - Clique em `Save Changes` para aplicar as alterações.
 </details>
@@ -95,6 +99,46 @@ O OpenHands exporta automaticamente um `GITHUB_TOKEN` para o ambiente shell se e
    - Se estiver usando uma organização, autorize o acesso à organização se solicitado.
 </details>
 
+#### Configuração do Token do GitLab
+
+O OpenHands exporta automaticamente um `GITLAB_TOKEN` para o ambiente shell, apenas para instalações locais, se ele estiver disponível.
+
+<details>
+  <summary>Configurando um Token do GitLab</summary>
+
+  1. **Gere um Personal Access Token (PAT)**:
+   - No GitLab, vá para User Settings > Access Tokens.
+   - Crie um novo token com os seguintes escopos:
+     - `api` (Acesso à API)
+     - `read_user` (Leitura de informações do usuário)
+     - `read_repository` (Leitura do repositório)
+     - `write_repository` (Escrita no repositório)
+   - Defina uma data de expiração ou deixe em branco para um token sem expiração.
+  2. **Insira o Token no OpenHands**:
+   - Clique no botão Settings (ícone de engrenagem).
+   - Navegue até a seção `Git Provider Settings`.
+   - Cole seu token no campo `GitLab Token`.
+   - Se estiver usando GitLab auto-hospedado, insira a URL da sua instância GitLab.
+   - Clique em `Save Changes` para aplicar as alterações.
+</details>
+
+<details>
+  <summary>Solução de Problemas</summary>
+
+  Problemas comuns e soluções:
+
+  - **Token Não Reconhecido**:
+     - Certifique-se de que o token esteja salvo corretamente nas configurações.
+     - Verifique se o token não expirou.
+     - Verifique se o token possui os escopos necessários.
+     - Para instâncias auto-hospedadas, verifique a URL correta da instância.
+
+  - **Acesso Negado**:
+     - Verifique as permissões de acesso ao projeto.
+     - Verifique se o token possui os escopos necessários.
+     - Para repositórios de grupo/organização, certifique-se de ter o acesso adequado.
+</details>
+
 ### Configurações Avançadas
 
 1. Dentro da página Settings, ative as opções `Advanced` para acessar configurações adicionais.

docs/i18n/pt-BR/docusaurus-plugin-content-docs/current/usage/how-to/headless-mode.md

@@ -32,7 +32,7 @@ Para executar o OpenHands no modo Headless com Docker:
 ```bash
 docker run -it \
     --pull=always \
-    -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.all-hands.dev/all-hands-ai/runtime:0.32-nikolaik \
+    -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.all-hands.dev/all-hands-ai/runtime:0.34-nikolaik \
     -e SANDBOX_USER_ID=$(id -u) \
     -e WORKSPACE_MOUNT_PATH=$WORKSPACE_BASE \
     -e LLM_API_KEY=$LLM_API_KEY \
@@ -43,7 +43,7 @@ docker run -it \
     -v ~/.openhands-state:/.openhands-state \
     --add-host host.docker.internal:host-gateway \
     --name openhands-app-$(date +%Y%m%d%H%M%S) \
-    docker.all-hands.dev/all-hands-ai/openhands:0.32 \
+    docker.all-hands.dev/all-hands-ai/openhands:0.34 \
     python -m openhands.core.main -t "escreva um script bash que imprima oi"
 ```
 

docs/i18n/pt-BR/docusaurus-plugin-content-docs/current/usage/installation.mdx

@@ -58,17 +58,17 @@
 A maneira mais fácil de executar o OpenHands é no Docker.
 
 ```bash
-docker pull docker.all-hands.dev/all-hands-ai/runtime:0.32-nikolaik
+docker pull docker.all-hands.dev/all-hands-ai/runtime:0.34-nikolaik
 
 docker run -it --rm --pull=always \
-    -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.all-hands.dev/all-hands-ai/runtime:0.32-nikolaik \
+    -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.all-hands.dev/all-hands-ai/runtime:0.34-nikolaik \
     -e LOG_ALL_EVENTS=true \
     -v /var/run/docker.sock:/var/run/docker.sock \
     -v ~/.openhands-state:/.openhands-state \
     -p 3000:3000 \
     --add-host host.docker.internal:host-gateway \
     --name openhands-app \
-    docker.all-hands.dev/all-hands-ai/openhands:0.32
+    docker.all-hands.dev/all-hands-ai/openhands:0.34
 ```
 
 Você encontrará o OpenHands em execução em http://localhost:3000!

docs/i18n/pt-BR/docusaurus-plugin-content-docs/current/usage/runtimes.md

@@ -13,7 +13,7 @@ Este é o Runtime padrão que é usado quando você inicia o OpenHands. Você po
 
 ```
 docker run # ...
-    -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.all-hands.dev/all-hands-ai/runtime:0.32-nikolaik \
+    -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.all-hands.dev/all-hands-ai/runtime:0.34-nikolaik \
     -v /var/run/docker.sock:/var/run/docker.sock \
     # ...
 ```

docs/i18n/zh-Hans/docusaurus-plugin-content-docs/current/usage/how-to/cli-mode.md

@@ -50,7 +50,7 @@ LLM_API_KEY="sk_test_12345"
 ```bash
 docker run -it \
     --pull=always \
-    -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.all-hands.dev/all-hands-ai/runtime:0.32-nikolaik \
+    -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.all-hands.dev/all-hands-ai/runtime:0.34-nikolaik \
     -e SANDBOX_USER_ID=$(id -u) \
     -e WORKSPACE_MOUNT_PATH=$WORKSPACE_BASE \
     -e LLM_API_KEY=$LLM_API_KEY \
@@ -59,7 +59,7 @@ docker run -it \
     -v /var/run/docker.sock:/var/run/docker.sock \
     --add-host host.docker.internal:host-gateway \
     --name openhands-app-$(date +%Y%m%d%H%M%S) \
-    docker.all-hands.dev/all-hands-ai/openhands:0.32 \
+    docker.all-hands.dev/all-hands-ai/openhands:0.34 \
     python -m openhands.core.cli
 ```
 

docs/i18n/zh-Hans/docusaurus-plugin-content-docs/current/usage/how-to/gui-mode.md

@@ -19,14 +19,18 @@ OpenHands 提供了一个用户友好的图形用户界面（GUI）模式，用
 3. 输入所选提供商对应的 `API Key`。
 4. 点击"保存"应用设置。
 
-### GitHub Token 设置
+### 版本控制令牌
+
+OpenHands 支持多个版本控制提供商。您可以同时配置多个提供商的令牌。
+
+#### GitHub Token 设置
 
 如果可用，OpenHands 会自动将 `GITHUB_TOKEN` 导出到 shell 环境中。这可以通过两种方式实现：
 
 1. **本地（OSS）**：用户直接输入他们的 GitHub token
 2. **在线（SaaS）**：通过 GitHub OAuth 身份验证获取 token
 
-#### 设置本地 GitHub Token
+##### 设置本地 GitHub Token
 
 1. **生成个人访问令牌（PAT）**：
    - 转到 GitHub 设置 > 开发者设置 > 个人访问令牌 > 令牌（经典）
@@ -38,11 +42,11 @@ OpenHands 提供了一个用户友好的图形用户界面（GUI）模式，用
 
 2. **在 OpenHands 中输入令牌**：
    - 点击右上角的设置按钮（齿轮图标）
-   - 导航到"GitHub"部分
+   - 导航到"Git Provider Settings"部分
    - 将令牌粘贴到"GitHub Token"字段中
    - 点击"保存"应用更改
 
-#### 组织令牌策略
+##### 组织令牌策略
 
 如果您使用组织仓库，可能需要额外的设置：
 
@@ -57,7 +61,7 @@ OpenHands 提供了一个用户友好的图形用户界面（GUI）模式，用
    - 如果需要，点击组织旁边的"启用 SSO"
    - 完成 SSO 授权过程
 
-#### OAuth 身份验证（在线模式）
+##### OAuth 身份验证（在线模式）
 
 在在线模式下使用 OpenHands 时，GitHub OAuth 流程：
 
@@ -72,7 +76,7 @@ OpenHands 提供了一个用户友好的图形用户界面（GUI）模式，用
    - 授权 OpenHands 访问您的 GitHub 帐户
    - 如果使用组织，在出现提示时授权组织访问
 
-#### 故障排除
+##### 故障排除
 
 常见问题和解决方案：
 
@@ -93,6 +97,43 @@ OpenHands 提供了一个用户友好的图形用户界面（GUI）模式，用
    - 检查浏览器控制台中是否有任何错误消息
    - 如果可用，使用设置中的"测试连接"按钮
 
+#### GitLab Token 设置
+
+OpenHands 会自动将 `GITLAB_TOKEN` 导出到 shell 环境中，仅适用于本地安装，如果它可用的话。
+
+##### 设置 GitLab Token
+
+1. **生成个人访问令牌（PAT）**：
+   - 在 GitLab 中，转到用户设置 > 访问令牌
+   - 创建具有以下范围的新令牌：
+     - `api`（API 访问）
+     - `read_user`（读取用户信息）
+     - `read_repository`（读取仓库）
+     - `write_repository`（写入仓库）
+   - 设置过期日期或留空以获取永不过期的令牌
+
+2. **在 OpenHands 中输入令牌**：
+   - 点击设置按钮（齿轮图标）
+   - 导航到 `Git Provider Settings` 部分
+   - 将令牌粘贴到 `GitLab Token` 字段中
+   - 如果使用自托管 GitLab，请输入您的 GitLab 实例 URL
+   - 点击 `Save Changes` 应用更改
+
+##### 故障排除
+
+常见问题和解决方案：
+
+1. **令牌无法识别**：
+   - 确保令牌已正确保存在设置中
+   - 检查令牌是否已过期
+   - 验证令牌是否具有所需的范围
+   - 对于自托管实例，验证正确的实例 URL
+
+2. **访问被拒绝**：
+   - 验证项目访问权限
+   - 检查令牌是否具有必要的范围
+   - 对于组/组织仓库，确保您拥有适当的访问权限
+
 ### 高级设置
 
 1. 切换`高级选项`以访问其他设置。

docs/i18n/zh-Hans/docusaurus-plugin-content-docs/current/usage/how-to/headless-mode.md

@@ -47,7 +47,7 @@ LLM_API_KEY="sk_test_12345"
 ```bash
 docker run -it \
     --pull=always \
-    -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.all-hands.dev/all-hands-ai/runtime:0.32-nikolaik \
+    -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.all-hands.dev/all-hands-ai/runtime:0.34-nikolaik \
     -e SANDBOX_USER_ID=$(id -u) \
     -e WORKSPACE_MOUNT_PATH=$WORKSPACE_BASE \
     -e LLM_API_KEY=$LLM_API_KEY \
@@ -57,6 +57,6 @@ docker run -it \
     -v /var/run/docker.sock:/var/run/docker.sock \
     --add-host host.docker.internal:host-gateway \
     --name openhands-app-$(date +%Y%m%d%H%M%S) \
-    docker.all-hands.dev/all-hands-ai/openhands:0.32 \
+    docker.all-hands.dev/all-hands-ai/openhands:0.34 \
     python -m openhands.core.main -t "write a bash script that prints hi" --no-auto-continue
 ```

docs/i18n/zh-Hans/docusaurus-plugin-content-docs/current/usage/installation.mdx

@@ -11,16 +11,16 @@
 在 Docker 中运行 OpenHands 是最简单的方式。
 
 ```bash
-docker pull docker.all-hands.dev/all-hands-ai/runtime:0.32-nikolaik
+docker pull docker.all-hands.dev/all-hands-ai/runtime:0.34-nikolaik
 
 docker run -it --rm --pull=always \
-    -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.all-hands.dev/all-hands-ai/runtime:0.32-nikolaik \
+    -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.all-hands.dev/all-hands-ai/runtime:0.34-nikolaik \
     -e LOG_ALL_EVENTS=true \
     -v /var/run/docker.sock:/var/run/docker.sock \
     -p 3000:3000 \
     --add-host host.docker.internal:host-gateway \
     --name openhands-app \
-    docker.all-hands.dev/all-hands-ai/openhands:0.32
+    docker.all-hands.dev/all-hands-ai/openhands:0.34
 ```
 
 你也可以在可脚本化的[无头模式](https://docs.all-hands.dev/modules/usage/how-to/headless-mode)下运行 OpenHands，作为[交互式 CLI](https://docs.all-hands.dev/modules/usage/how-to/cli-mode)，或使用 [OpenHands GitHub Action](https://docs.all-hands.dev/modules/usage/how-to/github-action)。

docs/i18n/zh-Hans/docusaurus-plugin-content-docs/current/usage/runtimes.md

@@ -11,7 +11,7 @@
 
 ```
 docker run # ...
-    -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.all-hands.dev/all-hands-ai/runtime:0.32-nikolaik \
+    -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.all-hands.dev/all-hands-ai/runtime:0.34-nikolaik \
     -v /var/run/docker.sock:/var/run/docker.sock \
     # ...
 ```

docs/modules/usage/cloud/openhands-cloud.mdx

@@ -19,7 +19,7 @@ After visiting OpenHands Cloud, you will be asked to connect with your GitHub ac
 ### Adding Repository Access
 
 You can grant OpenHands specific repository access:
-1. Click the `Select a GitHub project` dropdown, select `Add more repositories...`.
+1. Click the `Select a Git project` dropdown, select `Add more repositories...`.
 2. Select the organization, then choose the specific repositories to grant OpenHands access to.
    <details>
      <summary>Permission Details for Repository Access</summary>
@@ -45,7 +45,7 @@ You can grant OpenHands specific repository access:
 ### Modifying Repository Access
 
 You can modify repository access at any time by:
-* Using the same `Select a GitHub project > Add more repositories` workflow, or
+* Using the same `Select a Git project > Add more repositories` workflow, or
 * Visiting the Settings page and selecting `Configure GitHub Repositories` under the `GitHub Settings` section.
 
 ## Conversation Persistence

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

@@ -35,7 +35,7 @@ To run OpenHands in CLI mode with Docker:
 ```bash
 docker run -it \
     --pull=always \
-    -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.all-hands.dev/all-hands-ai/runtime:0.32-nikolaik \
+    -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.all-hands.dev/all-hands-ai/runtime:0.34-nikolaik \
     -e SANDBOX_USER_ID=$(id -u) \
     -e WORKSPACE_MOUNT_PATH=$WORKSPACE_BASE \
     -e LLM_API_KEY=$LLM_API_KEY \
@@ -45,7 +45,7 @@ docker run -it \
     -v ~/.openhands-state:/.openhands-state \
     --add-host host.docker.internal:host-gateway \
     --name openhands-app-$(date +%Y%m%d%H%M%S) \
-    docker.all-hands.dev/all-hands-ai/openhands:0.32 \
+    docker.all-hands.dev/all-hands-ai/openhands:0.34 \
     python -m openhands.core.cli
 ```
 

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

@@ -1,7 +1,8 @@
 # Custom Sandbox
 
 :::note
-This guide is for users that would like to use their own custom Docker image for the runtime, e.g. with certain tools or programming languages pre-installed
+This guide is for users that would like to use their own custom Docker image for the runtime. For example
+with certain tools or programming languages pre-installed.
 :::
 
 The sandbox is where the agent performs its tasks. Instead of running commands directly on your computer

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

@@ -18,11 +18,14 @@ OpenHands provides a Graphical User Interface (GUI) mode for interacting with th
 3. Enter the corresponding `API Key` for your chosen provider.
 4. Click `Save Changes` to apply the settings.
 
-### GitHub Token Setup
+### Version Control Tokens
 
-OpenHands automatically exports a `GITHUB_TOKEN` to the shell environment if it is available. This can happen in two ways:
+OpenHands supports multiple version control providers. You can configure tokens for multiple providers simultaneously.
+
+#### GitHub Token Setup
+
+OpenHands automatically exports a `GITHUB_TOKEN` to the shell environment if provided:
 
-**Local Installation**: The user directly inputs their GitHub token.
 <details>
   <summary>Setting Up a GitHub Token</summary>
 
@@ -36,9 +39,8 @@ OpenHands automatically exports a `GITHUB_TOKEN` to the shell environment if it
      - Minimal Permissions ( Select `Meta Data = Read-only` read for search, `Pull Requests = Read and Write` and `Content = Read and Write` for branch creation)
   2. **Enter Token in OpenHands**:
    - Click the Settings button (gear icon).
-   - Navigate to the `GitHub Settings` section.
    - Paste your token in the `GitHub Token` field.
-   - Click `Save Changes` to apply the changes.
+   - Click `Save` to apply the changes.
 </details>
 
 <details>
@@ -79,21 +81,43 @@ OpenHands automatically exports a `GITHUB_TOKEN` to the shell environment if it
      - Check the browser console for any error messages.
 </details>
 
-**OpenHands Cloud**: The token is obtained through GitHub OAuth authentication.
+#### GitLab Token Setup
+
+OpenHands automatically exports a `GITLAB_TOKEN` to the shell environment if provided:
 
 <details>
-  <summary>OAuth Authentication</summary>
+  <summary>Setting Up a GitLab Token</summary>
 
-  When using OpenHands Cloud, the GitHub OAuth flow requests the following permissions:
-   - Repository access (read/write)
-   - Workflow management
-   - Organization read access
+  1. **Generate a Personal Access Token (PAT)**:
+   - On GitLab, go to User Settings > Access Tokens.
+   - Create a new token with the following scopes:
+     - `api` (API access)
+     - `read_user` (Read user information)
+     - `read_repository` (Read repository)
+     - `write_repository` (Write repository)
+   - Set an expiration date or leave it blank for a non-expiring token.
+  2. **Enter Token in OpenHands**:
+   - Click the Settings button (gear icon).
+   - Paste your token in the `GitLab Token` field.
+   - Enter your GitLab instance URL if using self-hosted GitLab.
+   - Click `Save` to apply the changes.
+</details>
 
-  To authenticate OpenHands:
-   - 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.
+<details>
+  <summary>Troubleshooting</summary>
+
+  Common issues and solutions:
+
+  - **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.
+     - For self-hosted instances, verify the correct instance URL.
+
+  - **Access Denied**:
+     - Verify project access permissions.
+     - Check if the token has the necessary scopes.
+     - For group/organization repositories, ensure you have proper access.
 </details>
 
 ### Advanced Settings

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

@@ -32,7 +32,7 @@ To run OpenHands in Headless mode with Docker:
 ```bash
 docker run -it \
     --pull=always \
-    -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.all-hands.dev/all-hands-ai/runtime:0.32-nikolaik \
+    -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.all-hands.dev/all-hands-ai/runtime:0.34-nikolaik \
     -e SANDBOX_USER_ID=$(id -u) \
     -e WORKSPACE_MOUNT_PATH=$WORKSPACE_BASE \
     -e LLM_API_KEY=$LLM_API_KEY \
@@ -43,7 +43,7 @@ docker run -it \
     -v ~/.openhands-state:/.openhands-state \
     --add-host host.docker.internal:host-gateway \
     --name openhands-app-$(date +%Y%m%d%H%M%S) \
-    docker.all-hands.dev/all-hands-ai/openhands:0.32 \
+    docker.all-hands.dev/all-hands-ai/openhands:0.34 \
     python -m openhands.core.main -t "write a bash script that prints hi"
 ```
 

docs/modules/usage/installation.mdx

@@ -58,17 +58,17 @@ A system with a modern processor and a minimum of **4GB RAM** is recommended to
 The easiest way to run OpenHands is in Docker.
 
 ```bash
-docker pull docker.all-hands.dev/all-hands-ai/runtime:0.32-nikolaik
+docker pull docker.all-hands.dev/all-hands-ai/runtime:0.34-nikolaik
 
 docker run -it --rm --pull=always \
-    -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.all-hands.dev/all-hands-ai/runtime:0.32-nikolaik \
+    -e SANDBOX_RUNTIME_CONTAINER_IMAGE=docker.all-hands.dev/all-hands-ai/runtime:0.34-nikolaik \
     -e LOG_ALL_EVENTS=true \
     -v /var/run/docker.sock:/var/run/docker.sock \
     -v ~/.openhands-state:/.openhands-state \
     -p 3000:3000 \
     --add-host host.docker.internal:host-gateway \
     --name openhands-app \
-    docker.all-hands.dev/all-hands-ai/openhands:0.32
+    docker.all-hands.dev/all-hands-ai/openhands:0.34
 ```
 
 You'll find OpenHands running at http://localhost:3000!

docs/modules/usage/key-features.md

@@ -12,6 +12,9 @@
   * Modify files
   * Upload and download files
 
+### Terminal
+- A space for OpenHands and users to run terminal commands.
+
 ### Jupyter
 - Shows all Python commands that were executed by OpenHands.
 - Particularly handy when using OpenHands to perform data visualization tasks.
@@ -23,6 +26,3 @@
 ### Browser
 - Used by OpenHands to browse websites.
 - The browser is non-interactive.
-
-### Terminal
-- A space for OpenHands and users to run terminal commands.

docs/modules/usage/llms/llms.md

@@ -17,6 +17,8 @@ Based on these findings and community feedback, the following models have been v
 - [gemini/gemini-2.5-pro](https://blog.google/technology/google-deepmind/gemini-model-thinking-updates-march-2025/)
 - [deepseek/deepseek-chat](https://api-docs.deepseek.com/)
 - [openai/o3-mini](https://openai.com/index/openai-o3-mini/)
+- [openai/o3](https://openai.com/index/introducing-o3-and-o4-mini/)
+- [openai/o4-mini](https://openai.com/index/introducing-o3-and-o4-mini/)
 - [all-hands/openhands-lm-32b-v0.1](https://www.all-hands.dev/blog/introducing-openhands-lm-32b----a-strong-open-coding-agent-model) -- available through [OpenRouter](https://openrouter.ai/all-hands/openhands-lm-32b-v0.1)
 
 

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

@@ -15,7 +15,7 @@ It is highly recommended that you use GPUs to serve local models for optimal exp
 For example, to download [OpenHands LM 32B v0.1](https://huggingface.co/all-hands/openhands-lm-32b-v0.1):
 
 ```bash
-huggingface-cli download all-hands/openhands-lm-32b-v0.1 --local-dir my_folder/openhands-lm-32b-v0.1
+huggingface-cli download all-hands/openhands-lm-32b-v0.1 --local-dir all-hands/openhands-lm-32b-v0.1
 ```
 
 ## Create an OpenAI-Compatible Endpoint With a Model Serving Framework
@@ -27,7 +27,7 @@ huggingface-cli download all-hands/openhands-lm-32b-v0.1 --local-dir my_folder/o
 
 ```bash
 SGLANG_ALLOW_OVERWRITE_LONGER_CONTEXT_LEN=1 python3 -m sglang.launch_server \
-    --model my_folder/openhands-lm-32b-v0.1 \
+    --model all-hands/openhands-lm-32b-v0.1 \
     --served-model-name openhands-lm-32b-v0.1 \
     --port 8000 \
     --tp 2 --dp 1 \
@@ -41,7 +41,7 @@ SGLANG_ALLOW_OVERWRITE_LONGER_CONTEXT_LEN=1 python3 -m sglang.launch_server \
 - Example launch command for OpenHands LM 32B (with at least 2 GPUs):
 
 ```bash
-vllm serve my_folder/openhands-lm-32b-v0.1 \
+vllm serve all-hands/openhands-lm-32b-v0.1 \
     --host 0.0.0.0 --port 8000 \
     --api-key mykey \
     --tensor-parallel-size 2 \
@@ -67,7 +67,7 @@ Ensure `config.toml` exists by running `make setup-config` which will create one
 workspace_base="/path/to/your/workspace"
 
 [llm]
-embedding_model="local"
+model="openhands-lm-32b-v0.1"
 ollama_base_url="http://localhost:8000"
 ```
 

evaluation/benchmarks/EDA/run_infer.py

@@ -14,7 +14,6 @@ from evaluation.utils.shared import (
     prepare_dataset,
     reset_logger_for_multiprocessing,
     run_evaluation,
-    update_agent_config_for_eval,
 )
 from openhands.controller.state.state import State
 from openhands.core.config import (
@@ -75,7 +74,6 @@ def get_config(
     )
     config.set_llm_config(metadata.llm_config)
     agent_config = config.get_agent_config(metadata.agent_class)
-    agent_config = update_agent_config_for_eval(agent_config)
     agent_config.enable_prompt_extensions = False
     return config
 

evaluation/benchmarks/agent_bench/run_infer.py

@@ -22,7 +22,6 @@ from evaluation.utils.shared import (
     prepare_dataset,
     reset_logger_for_multiprocessing,
     run_evaluation,
-    update_agent_config_for_eval,
 )
 from openhands.controller.state.state import State
 from openhands.core.config import (
@@ -56,8 +55,6 @@ def get_config(
     )
     config.set_llm_config(metadata.llm_config)
     agent_config = config.get_agent_config(metadata.agent_class)
-
-    agent_config = update_agent_config_for_eval(agent_config)
     agent_config.enable_prompt_extensions = False
     return config
 

evaluation/benchmarks/aider_bench/run_infer.py

@@ -21,7 +21,6 @@ from evaluation.utils.shared import (
     prepare_dataset,
     reset_logger_for_multiprocessing,
     run_evaluation,
-    update_agent_config_for_eval,
 )
 from openhands.controller.state.state import State
 from openhands.core.config import (
@@ -62,8 +61,6 @@ def get_config(
     )
     config.set_llm_config(metadata.llm_config)
     agent_config = config.get_agent_config(metadata.agent_class)
-
-    agent_config = update_agent_config_for_eval(agent_config)
     agent_config.enable_prompt_extensions = False
 
     # copy 'draft_editor' config if exists

evaluation/benchmarks/biocoder/run_infer.py

@@ -19,7 +19,6 @@ from evaluation.utils.shared import (
     prepare_dataset,
     reset_logger_for_multiprocessing,
     run_evaluation,
-    update_agent_config_for_eval,
 )
 from openhands.controller.state.state import State
 from openhands.core.config import (
@@ -73,8 +72,6 @@ def get_config(
     )
     config.set_llm_config(metadata.llm_config)
     agent_config = config.get_agent_config(metadata.agent_class)
-
-    agent_config = update_agent_config_for_eval(agent_config)
     agent_config.enable_prompt_extensions = False
     return config
 

evaluation/benchmarks/bird/run_infer.py

@@ -22,7 +22,6 @@ from evaluation.utils.shared import (
     prepare_dataset,
     reset_logger_for_multiprocessing,
     run_evaluation,
-    update_agent_config_for_eval,
 )
 from openhands.controller.state.state import State
 from openhands.core.config import (
@@ -87,8 +86,6 @@ def get_config(
     )
     config.set_llm_config(metadata.llm_config)
     agent_config = config.get_agent_config(metadata.agent_class)
-
-    agent_config = update_agent_config_for_eval(agent_config)
     agent_config.enable_prompt_extensions = False
     return config
 

evaluation/benchmarks/browsing_delegation/run_infer.py

@@ -15,7 +15,6 @@ from evaluation.utils.shared import (
     prepare_dataset,
     reset_logger_for_multiprocessing,
     run_evaluation,
-    update_agent_config_for_eval,
 )
 from openhands.controller.state.state import State
 from openhands.core.config import (
@@ -51,8 +50,6 @@ def get_config(
     )
     config.set_llm_config(metadata.llm_config)
     agent_config = config.get_agent_config(metadata.agent_class)
-
-    agent_config = update_agent_config_for_eval(agent_config)
     agent_config.enable_prompt_extensions = False
     return config
 

evaluation/benchmarks/commit0/run_infer.py

@@ -20,7 +20,6 @@ from evaluation.utils.shared import (
     prepare_dataset,
     reset_logger_for_multiprocessing,
     run_evaluation,
-    update_agent_config_for_eval,
     update_llm_config_for_completions_logging,
 )
 from openhands.controller.state.state import State
@@ -136,7 +135,6 @@ def get_config(
         enable_browsing=RUN_WITH_BROWSING,
         enable_llm_editor=False,
     )
-    agent_config = update_agent_config_for_eval(agent_config)
     config.set_agent_config(agent_config)
     return config
 

evaluation/benchmarks/discoverybench/run_infer.py

@@ -21,7 +21,6 @@ from evaluation.utils.shared import (
     prepare_dataset,
     reset_logger_for_multiprocessing,
     run_evaluation,
-    update_agent_config_for_eval,
 )
 from openhands.controller.state.state import State
 from openhands.core.config import (
@@ -77,8 +76,6 @@ def get_config(
     )
     config.set_llm_config(metadata.llm_config)
     agent_config = config.get_agent_config(metadata.agent_class)
-
-    agent_config = update_agent_config_for_eval(agent_config)
     agent_config.enable_prompt_extensions = False
     agent_config = AgentConfig(
         function_calling=False,

evaluation/benchmarks/gaia/run_infer.py

@@ -18,7 +18,6 @@ from evaluation.utils.shared import (
     prepare_dataset,
     reset_logger_for_multiprocessing,
     run_evaluation,
-    update_agent_config_for_eval,
 )
 from openhands.controller.state.state import State
 from openhands.core.config import (
@@ -67,8 +66,6 @@ def get_config(
     else:
         logger.info('Agent config not provided, using default settings')
         agent_config = config.get_agent_config(metadata.agent_class)
-
-        agent_config = update_agent_config_for_eval(agent_config)
         agent_config.enable_prompt_extensions = False
     return config
 

evaluation/benchmarks/gorilla/run_infer.py

@@ -16,7 +16,6 @@ from evaluation.utils.shared import (
     prepare_dataset,
     reset_logger_for_multiprocessing,
     run_evaluation,
-    update_agent_config_for_eval,
 )
 from openhands.controller.state.state import State
 from openhands.core.config import (
@@ -55,8 +54,6 @@ def get_config(
     )
     config.set_llm_config(metadata.llm_config)
     agent_config = config.get_agent_config(metadata.agent_class)
-
-    agent_config = update_agent_config_for_eval(agent_config)
     agent_config.enable_prompt_extensions = False
     return config
 

evaluation/benchmarks/gpqa/run_infer.py

@@ -34,7 +34,6 @@ from evaluation.utils.shared import (
     prepare_dataset,
     reset_logger_for_multiprocessing,
     run_evaluation,
-    update_agent_config_for_eval,
 )
 from openhands.controller.state.state import State
 from openhands.core.config import (
@@ -76,8 +75,6 @@ def get_config(
     )
     config.set_llm_config(metadata.llm_config)
     agent_config = config.get_agent_config(metadata.agent_class)
-
-    agent_config = update_agent_config_for_eval(agent_config)
     agent_config.enable_prompt_extensions = False
     return config
 

evaluation/benchmarks/humanevalfix/run_infer.py

@@ -27,7 +27,6 @@ from evaluation.utils.shared import (
     prepare_dataset,
     reset_logger_for_multiprocessing,
     run_evaluation,
-    update_agent_config_for_eval,
 )
 from openhands.controller.state.state import State
 from openhands.core.config import (
@@ -97,8 +96,6 @@ def get_config(
     )
     config.set_llm_config(metadata.llm_config)
     agent_config = config.get_agent_config(metadata.agent_class)
-
-    agent_config = update_agent_config_for_eval(agent_config)
     agent_config.enable_prompt_extensions = False
     return config
 

evaluation/benchmarks/lca_ci_build_repair/.gitignore

@@ -0,0 +1 @@
+config.yaml

evaluation/benchmarks/lca_ci_build_repair/README.MD

@@ -0,0 +1,35 @@
+# CI Builds Repair Benchmark Integration
+
+This module integrates the CI Builds Repair benchmark developed by [JetBrains-Research](https://github.com/JetBrains-Research/lca-baselines/tree/main/ci-builds-repair/ci-builds-repair-benchmark).
+
+For more information, refer to the [GitHub repository](https://github.com/JetBrains-Research/lca-baselines/tree/main/ci-builds-repair/ci-builds-repair-benchmark) and the associated [research paper](https://arxiv.org/abs/2406.11612).
+See notice below for details
+
+## Setup
+
+Before running any scripts, make sure to configure the benchmark by setting up `config.yaml`.
+This benchmark pushes to JetBrains' private GitHub repository. You will to request a `token_gh` provided by their team, to run this benchmark.
+
+## Inference
+
+To run inference with your model:
+
+```bash
+./evaluation/benchmarks/lca_ci_build_repair/scripts/run_infer.sh llm.yourmodel
+```
+
+## Evaluation
+
+To evaluate the predictions:
+
+```bash
+./evaluation/benchmarks/lca_ci_build_repair/scripts/eval_infer.sh predictions_path_containing_output
+```
+
+## Results
+The benchmark contains 68 instances, we skip instances #126 and #145, and only run 66 instances due to dockerization errors.
+
+Due to running in live GitHub machines, the benchmark is sensitive to the date it is run. Even the golden patches in the dataset might present failures due to updates.
+For example, on 2025-04-09, running the benchmark against the golden patches gave 57/67 successes, with 1 job left in the waiting list.
+
+On 2025-04-10, running the benchmark full with OH and no oracle, 37 succeeded. That is 54% of the complete set of 68 instances and 64% of the 57 that succeed with golden patches.

evaluation/benchmarks/lca_ci_build_repair/config_template.yaml

@@ -0,0 +1,11 @@
+LCA_PATH: path #where to clone lca-ci rep
+model_name: OpenHands
+benchmark_owner: ICML-25-BenchName-builds-repair
+token_gh: your_token
+#for lca-ci-repo
+repos_folder: /path/to/repos # here the cloned repos would be stored
+out_folder: /out/folder # here the result files would be stored
+data_cache_dir: /data/cache/dir/ # here the cached dataset would be stored
+username_gh: username-gh # your GitHub username
+# test_username: test_user # username that would be displayed in the benchmark. Optional. If ommitted, username_gh would be used
+language: Python # dataset language (now only Python is available)

evaluation/benchmarks/lca_ci_build_repair/eval_infer.py

@@ -0,0 +1,242 @@
+"""Implements evaluation on JetBrains CI builds repair baselines
+
+Please see https://github.com/JetBrains-Research/lca-baselines/tree/main/ci-builds-repair
+and https://huggingface.co/datasets/JetBrains-Research/lca-ci-builds-repair
+
+TODOs:
+- Add more flags
+"""
+
+import json
+import os
+from pathlib import Path
+
+import ruamel.yaml
+
+from evaluation.utils.shared import (
+    EvalMetadata,
+    get_default_sandbox_config_for_eval,
+    make_metadata,
+)
+from openhands.core.config import (
+    AppConfig,
+    LLMConfig,
+    get_parser,
+    load_app_config,
+)
+from openhands.core.logger import openhands_logger as logger
+from openhands.core.main import create_runtime
+from openhands.events.action import CmdRunAction
+from openhands.events.observation import CmdOutputObservation
+from openhands.runtime.base import Runtime
+from openhands.utils.async_utils import call_async_from_sync
+
+
+def get_config(
+    metadata: EvalMetadata,
+) -> AppConfig:
+    sandbox_config = get_default_sandbox_config_for_eval()
+    sandbox_config.base_container_image = 'python:3.12-bookworm'
+    config = AppConfig(
+        default_agent=metadata.agent_class,
+        run_as_openhands=False,
+        runtime='docker',
+        max_iterations=metadata.max_iterations,
+        sandbox=sandbox_config,
+        # do not mount workspace
+        workspace_base=None,
+        workspace_mount_path=None,
+    )
+    config.set_llm_config(metadata.llm_config)
+    agent_config = config.get_agent_config(metadata.agent_class)
+    agent_config.enable_prompt_extensions = False
+    return config
+
+
+config = load_app_config()
+
+
+def load_bench_config():
+    script_dir = os.path.dirname(
+        os.path.abspath(__file__)
+    )  # Get the absolute path of the script
+    config_path = os.path.join(script_dir, 'config.yaml')
+    yaml = ruamel.yaml.YAML(typ='rt')
+    with open(config_path, 'r') as file:
+        return yaml.load(file)
+
+
+bench_config = load_bench_config()
+
+
+def run_eval(
+    runtime: Runtime,
+):
+    """Run the evaluation and create report"""
+    logger.info(f"{'-' * 50} BEGIN Runtime Initialization Fn {'-' * 50}")
+    obs: CmdOutputObservation
+
+    lca_path = bench_config['LCA_PATH']
+    lca_ci_path = os.path.join(
+        lca_path, 'lca-baselines', 'ci-builds-repair', 'ci-builds-repair-benchmark'
+    )
+
+    model_name = bench_config['model_name']
+
+    action = CmdRunAction(command=f'mkdir {lca_path}')
+    logger.info(action, extra={'msg_type': 'ACTION'})
+    obs = runtime.run_action(action)
+    assert obs.exit_code == 0
+
+    action = CmdRunAction(command=f'cd {lca_path}')
+    logger.info(action, extra={'msg_type': 'ACTION'})
+    obs = runtime.run_action(action)
+    assert obs.exit_code == 0
+
+    lca_repo_url = 'https://github.com/juanmichelini/lca-baselines'
+    action = CmdRunAction(command=f'git clone {lca_repo_url}')
+    logger.info(action, extra={'msg_type': 'ACTION'})
+    obs = runtime.run_action(action)
+    assert obs.exit_code == 0
+
+    action = CmdRunAction(command=f'cd {lca_ci_path}')
+    logger.info(action, extra={'msg_type': 'ACTION'})
+    obs = runtime.run_action(action)
+    assert obs.exit_code == 0
+
+    action = CmdRunAction(command='git switch open-hands-integration')
+    logger.info(action, extra={'msg_type': 'ACTION'})
+    obs = runtime.run_action(action)
+    assert obs.exit_code == 0
+
+    script_dir = os.path.dirname(
+        os.path.abspath(__file__)
+    )  # Get the absolute path of the script
+    config_path = os.path.join(script_dir, 'config.yaml')
+    runtime.copy_to(config_path, lca_ci_path)
+
+    token_gh = bench_config['token_gh']
+    commandf = f'export TOKEN_GH={token_gh}'
+    action = CmdRunAction(command=commandf)
+    logger.info(action, extra={'msg_type': 'ACTION'})
+    obs = runtime.run_action(action)
+
+    action = CmdRunAction(command='poetry install')
+    logger.info(action, extra={'msg_type': 'ACTION'})
+    obs = runtime.run_action(action)
+
+    # Set up the task environment
+    commandf = f'poetry run python run_eval_jobs.py --model-name "{model_name}" --config-path "{lca_ci_path}/config.yaml" --job-ids-file "/tmp/output_lca.jsonl" --result-filename "testfile.jsonl"  > /tmp/single_output.txt'
+    action = CmdRunAction(command=commandf)
+    logger.info(action, extra={'msg_type': 'ACTION'})
+    obs = runtime.run_action(action)
+    logger.info(f'run_eval_jobs.py gave {obs.content} !')
+    # assert obs.exit_code == 0
+
+    commandf = 'cat /tmp/single_output.txt'
+    action = CmdRunAction(command=commandf)
+    logger.info(action, extra={'msg_type': 'ACTION'})
+    obs = runtime.run_action(action)
+    logger.info(f' {commandf} gave {obs.content}!')
+
+    testfile_path = os.path.join(bench_config['out_folder'], 'testfile.jsonl')
+    commandf = f'cat {testfile_path}'
+    action = CmdRunAction(command=commandf)
+    logger.info(action, extra={'msg_type': 'ACTION'})
+    obs = runtime.run_action(action)
+    report_str = obs.content
+
+    logger.info(f"{'-' * 50} END Runtime Initialization Fn {'-' * 50}")
+    return report_str
+
+
+def process_predictions(predictions_path: str):
+    output_path = Path(predictions_path)
+    if output_path.suffix != '.jsonl':
+        raise ValueError('output_path must end in .jsonl')
+
+    output_lca_path = output_path.with_name(output_path.stem + '_lca.jsonl')
+
+    with output_path.open() as infile, output_lca_path.open('w') as outfile:
+        for line in infile:
+            data = json.loads(line)
+            json.dump(data.get('test_result'), outfile)
+            outfile.write('\n')
+
+    return str(output_lca_path)
+
+
+if __name__ == '__main__':
+    parser = get_parser()
+    parser.add_argument(
+        '-s',
+        '--eval-split',
+        type=str,
+        default='test',
+        choices=['test'],
+        help='data split to evaluate on, must be test',
+    )
+    parser.add_argument(
+        '--predictions-path',
+        type=str,
+        help='Path to the directory containing the output.jsonl with the predictions.',
+    )
+    args, _ = parser.parse_known_args()
+
+    data_split = args.eval_split
+
+    llm_config = LLMConfig(model='dummy_model')
+
+    metadata = make_metadata(
+        llm_config,
+        f'jetbrains-lca-ci--{data_split}',
+        args.agent_cls,
+        args.max_iterations,
+        args.eval_note,
+        args.predictions_path,
+    )
+
+    # prepare image
+    config = get_config(metadata)
+    runtime = create_runtime(config)
+    call_async_from_sync(runtime.connect)
+    logger.info('Converting output.jsonl into output_lca.jsonl')
+    predictions_lca_path = process_predictions(
+        os.path.join(args.predictions_path, 'output.jsonl')
+    )
+    runtime.copy_to(predictions_lca_path, '/tmp')
+
+    # get results
+    results_str = run_eval(runtime)
+    results_path = os.path.join(args.predictions_path, 'results.jsonl')
+    with open(results_path, 'w') as file:
+        file.write(results_str)
+    logger.info(f'Saved results to {results_path}')
+
+    # make a summary
+    resolved_instances = []
+    unresolved_instances = []
+    for line in results_str.strip().splitlines():
+        data = json.loads(line)
+        conclusion = data.get('conclusion')
+        if conclusion == 'success':
+            resolved_instances.append(data)
+        elif conclusion == 'failure':
+            unresolved_instances.append(data)
+
+    completed_instances = resolved_instances + unresolved_instances
+
+    report = {
+        'success': len(resolved_instances),
+        'failure': len(unresolved_instances),
+        'resolved_instances': resolved_instances,
+        'unresolved_instances': unresolved_instances,
+        'completed_instances': completed_instances,
+    }
+
+    print(f'Results: {report}')
+    report_path = os.path.join(args.predictions_path, 'report.jsonl')
+    with open(report_path, 'w') as out_f:
+        out_f.write(json.dumps(report) + '\n')
+
+    logger.info(f'Saved report of results in swebench format to {report_path}')

evaluation/benchmarks/lca_ci_build_repair/run_infer.py

@@ -0,0 +1,406 @@
+"""Implements inference on JetBrains CI builds repair baselines
+
+Please see https://github.com/JetBrains-Research/lca-baselines/tree/main/ci-builds-repair
+and https://huggingface.co/datasets/JetBrains-Research/lca-ci-builds-repair
+
+TODOs:
+- Add EXP_NAME
+"""
+
+import asyncio
+import json
+import os
+from typing import Any
+
+import pandas as pd
+import ruamel.yaml
+from datasets import load_dataset
+
+from evaluation.utils.shared import (
+    EvalMetadata,
+    EvalOutput,
+    codeact_user_response,
+    compatibility_for_eval_history_pairs,
+    get_default_sandbox_config_for_eval,
+    make_metadata,
+    prepare_dataset,
+    reset_logger_for_multiprocessing,
+    run_evaluation,
+)
+from openhands.controller.state.state import State
+from openhands.core.config import (
+    AppConfig,
+    get_llm_config_arg,
+    get_parser,
+    load_app_config,
+)
+from openhands.core.logger import openhands_logger as logger
+from openhands.core.main import create_runtime, run_controller
+from openhands.events.action import CmdRunAction, MessageAction
+from openhands.events.observation import CmdOutputObservation
+from openhands.runtime.base import Runtime
+from openhands.utils.async_utils import call_async_from_sync
+
+
+def get_config(
+    metadata: EvalMetadata,
+) -> AppConfig:
+    sandbox_config = get_default_sandbox_config_for_eval()
+    sandbox_config.base_container_image = 'python:3.12-bookworm'
+    config = AppConfig(
+        default_agent=metadata.agent_class,
+        run_as_openhands=False,
+        runtime='docker',
+        max_iterations=metadata.max_iterations,
+        sandbox=sandbox_config,
+        # do not mount workspace
+        workspace_base=None,
+        workspace_mount_path=None,
+    )
+    config.set_llm_config(metadata.llm_config)
+    agent_config = config.get_agent_config(metadata.agent_class)
+    agent_config.enable_prompt_extensions = False
+    return config
+
+
+config = load_app_config()
+
+
+def load_bench_config():
+    script_dir = os.path.dirname(
+        os.path.abspath(__file__)
+    )  # Get the absolute path of the script
+    config_path = os.path.join(script_dir, 'config.yaml')
+    yaml = ruamel.yaml.YAML(typ='rt')
+    with open(config_path, 'r') as file:
+        return yaml.load(file)
+
+
+bench_config = load_bench_config()
+
+AGENT_CLS_TO_FAKE_USER_RESPONSE_FN = {
+    'CodeActAgent': codeact_user_response,
+}
+
+AGENT_CLS_TO_INST_SUFFIX = {
+    'CodeActAgent': 'When you think you have completed the task, please finish the interaction using the "finish" tool.\n'
+}
+
+
+def initialize_runtime(
+    runtime: Runtime,
+    instance: pd.Series,
+):
+    """Initialize the runtime for the agent.
+
+    This function is called before the runtime is used to run the agent.
+    """
+    logger.info(f"{'-' * 50} BEGIN Runtime Initialization Fn {'-' * 50}")
+    obs: CmdOutputObservation
+
+    lca_path = bench_config['LCA_PATH']
+    lca_ci_path = os.path.join(
+        lca_path, 'lca-baselines', 'ci-builds-repair', 'ci-builds-repair-benchmark'
+    )
+
+    repo_name = instance['repo_name']
+    repos_path = bench_config['repos_folder']
+    repo_owner = instance['repo_owner']
+    repo_path = os.path.join(repos_path, f'{repo_owner}__{repo_name}')
+    model_name = bench_config['model_name']
+
+    action = CmdRunAction(command=f'mkdir {lca_path}')
+    logger.info(action, extra={'msg_type': 'ACTION'})
+    obs = runtime.run_action(action)
+    assert obs.exit_code == 0
+
+    action = CmdRunAction(command=f'cd {lca_path}')
+    logger.info(action, extra={'msg_type': 'ACTION'})
+    obs = runtime.run_action(action)
+    assert obs.exit_code == 0
+
+    lca_repo_url = 'https://github.com/juanmichelini/lca-baselines'
+    action = CmdRunAction(command=f'git clone {lca_repo_url}')
+    logger.info(action, extra={'msg_type': 'ACTION'})
+    obs = runtime.run_action(action)
+    assert obs.exit_code == 0
+
+    action = CmdRunAction(command=f'cd {lca_ci_path}')
+    logger.info(action, extra={'msg_type': 'ACTION'})
+    obs = runtime.run_action(action)
+    assert obs.exit_code == 0
+
+    action = CmdRunAction(command='git switch open-hands-integration')
+    logger.info(action, extra={'msg_type': 'ACTION'})
+    obs = runtime.run_action(action)
+    assert obs.exit_code == 0
+
+    script_dir = os.path.dirname(
+        os.path.abspath(__file__)
+    )  # Get the absolute path of the script
+    config_path = os.path.join(script_dir, 'config.yaml')
+    with open(config_path, 'r') as file:
+        config_as_text = file.read()
+
+    commandf = f"echo '{config_as_text}' > config.yaml"
+    action = CmdRunAction(command=commandf)
+    logger.info(action, extra={'msg_type': 'ACTION'})
+    obs = runtime.run_action(action)
+
+    token_gh = bench_config['token_gh']
+    commandf = f'export TOKEN_GH={token_gh}'
+    action = CmdRunAction(command=commandf)
+    logger.info(action, extra={'msg_type': 'ACTION'})
+    obs = runtime.run_action(action)
+
+    action = CmdRunAction(command='poetry install')
+    logger.info(action, extra={'msg_type': 'ACTION'})
+    obs = runtime.run_action(action)
+
+    # Set up the task environment
+    commandf = f'poetry run python run_get_datapoint.py --model-name {model_name} --id {instance["id"]} > branch_name.txt'
+    action = CmdRunAction(command=commandf)
+    logger.info(action, extra={'msg_type': 'ACTION'})
+    obs = runtime.run_action(action)
+    if obs.exit_code != 0:
+        print(f'run_get_datapoint.py failed at {instance["id"]} with {obs.content}')
+    assert obs.exit_code == 0
+
+    commandf = 'cat branch_name.txt'
+    action = CmdRunAction(command=commandf)
+    logger.info(action, extra={'msg_type': 'ACTION'})
+    obs = runtime.run_action(action)
+    bench_config['user_branch_name'] = obs.content
+
+    # Navigate to the task's code path
+    action = CmdRunAction(command=f'cd {repo_path}')
+    logger.info(action, extra={'msg_type': 'ACTION'})
+    obs = runtime.run_action(action)
+
+    logger.info(f"{'-' * 50} END Runtime Initialization Fn {'-' * 50}")
+
+
+def complete_runtime(
+    runtime: Runtime,
+    instance: pd.Series,
+) -> dict[str, Any]:
+    """Complete the runtime for the agent.
+
+    This function is called before the runtime is used to run the agent.
+    If you need to do something in the sandbox to get the correctness metric after
+    the agent has run, modify this function.
+    """
+    logger.info(f"{'-' * 50} BEGIN Runtime Completion Fn {'-' * 50}")
+    obs: CmdOutputObservation
+
+    model_name = bench_config['model_name']
+
+    lca_path = bench_config['LCA_PATH']
+    lca_ci_path = os.path.join(
+        lca_path, 'lca-baselines', 'ci-builds-repair', 'ci-builds-repair-benchmark'
+    )
+
+    user_branch_name = bench_config['user_branch_name']
+
+    token_gh = bench_config['token_gh']
+    commandf = f'export TOKEN_GH={token_gh}'
+    action = CmdRunAction(command=commandf)
+    logger.info(action, extra={'msg_type': 'ACTION'})
+    obs = runtime.run_action(action)
+
+    # Navigate to the lca-baseslines scripts path
+    action = CmdRunAction(command=f'cd {lca_ci_path}')
+    logger.info(action, extra={'msg_type': 'ACTION'})
+    obs = runtime.run_action(action)
+    assert obs.exit_code == 0
+
+    commandf = f'poetry run python run_push_datapoint.py --id {instance["id"]} --model-name {model_name} --user-branch-name {user_branch_name} > single_output.json'
+    logger.info(f'Running push script: {commandf}')
+    action = CmdRunAction(command=commandf)
+    logger.info(action, extra={'msg_type': 'ACTION'})
+    obs = runtime.run_action(action)
+    # assert obs.exit_code == 0
+
+    commandf = 'cat single_output.json'
+    action = CmdRunAction(command=commandf)
+    logger.info(action, extra={'msg_type': 'ACTION'})
+    obs = runtime.run_action(action)
+    result = json.loads(obs.content)
+
+    logger.info(f"{'-' * 50} END Runtime Completion Fn {'-' * 50}")
+
+    return result
+
+
+def process_instance(instance: Any, metadata: EvalMetadata, reset_logger: bool = True):
+    config = get_config(metadata)
+
+    # Setup the logger properly, so you can run multi-processing to parallelize the evaluation
+    if reset_logger:
+        log_dir = os.path.join(metadata.eval_output_dir, 'infer_logs')
+        reset_logger_for_multiprocessing(logger, instance['instance_id'], log_dir)
+    else:
+        logger.info(f'Starting evaluation for instance {instance["instance_id"]}.')
+
+    repo_name = instance['repo_name']
+    repo_workflow = instance['workflow_path']
+    repo_logs = instance['logs']
+    repos_path = bench_config['repos_folder']
+    repo_owner = instance['repo_owner']
+    repo_path = os.path.join(repos_path, f'{repo_owner}__{repo_name}')
+
+    # Prepare the task instruction
+    instruction_no_oracle = f"""
+<uploaded_files>
+{repo_path}
+</uploaded_files>
+
+I've uploaded a python code repository in the directory {repo_path}, Consider the following issue:
+
+<issue_description>
+The repository must pass the CI workflow {repo_workflow}.
+but it gave the following error
+{repo_logs}
+</issue_description>
+
+Can you help me implement the necessary changes to the repository so that the requirements specified in the <issue_description> are met?
+I've already taken care of all changes to any of the test files described in the <issue_description>. This means you DON'T have to modify the testing logic or any of the tests in any way!
+Also the development Python environment is already set up for you (i.e., all dependencies already installed), so you don't need to install other packages.
+Your task is to make the minimal changes to non-test files in the {repo_path} directory to ensure the <issue_description> is satisfied.
+
+Follow these phases to resolve the issue:
+
+Phase 1. READING: read the problem and reword it in clearer terms
+   1.1 If there are code or config snippets. Express in words any best practices or conventions in them.
+   1.2 Hightlight message errors, method names, variables, file names, stack traces, and technical details.
+   1.3 Explain the problem in clear terms.
+   1.4 Enumerate the steps to reproduce the problem.
+   1.5 Hightlight any best practices to take into account when testing and fixing the issue
+
+Phase 2. RUNNING: install and run the tests on the repository
+   2.1 Follow the readme
+   2.2 Install the environment and anything needed
+   2.2 Iterate and figure out how to run the tests
+
+Phase 3. EXPLORATION: find the files that are related to the problem and possible solutions
+   3.1 Use `grep` to search for relevant methods, classes, keywords and error messages.
+   3.2 Identify all files related to the problem statement.
+   3.3 Propose the methods and files to fix the issue and explain why.
+   3.4 From the possible file locations, select the most likely location to fix the issue.
+
+Phase 4. TEST CREATION: before implementing any fix, create a script to reproduce and verify the issue.
+   4.1 Look at existing test files in the repository to understand the test format/structure.
+   4.2 Create a minimal reproduction script that reproduces the located issue.
+   4.3 Run the reproduction script to confirm you are reproducing the issue.
+   4.4 Adjust the reproduction script as necessary.
+
+Phase 5. FIX ANALYSIS: state clearly the problem and how to fix it
+   5.1 State clearly what the problem is.
+   5.2 State clearly where the problem is located.
+   5.3 State clearly how the test reproduces the issue.
+   5.4 State clearly the best practices to take into account in the fix.
+   5.5 State clearly how to fix the problem.
+
+Phase 6. FIX IMPLEMENTATION: Edit the source code to implement your chosen solution.
+   6.1 Make minimal, focused changes to fix the issue.
+
+Phase 7. VERIFICATION: Test your implementation thoroughly.
+   7.1 Run your reproduction script to verify the fix works.
+   7.2 Add edge cases to your test script to ensure comprehensive coverage.
+   7.3 Run existing tests related to the modified code to ensure you haven't broken anything. Run any tests in the repository related to:
+     7.2.1 The issue you are fixing
+     7.2.2 The files you modified
+     7.2.3 The functions you changed
+   7.4 If any tests fail, revise your implementation until all tests pass
+
+Phase 8. REVIEW: Carefully re-read the problem description and compare your changes with the base commit {instance["sha_fail"]}.
+   8.1 Ensure you've fully addressed all requirements.
+
+Once all phases are done, announce: 'Agent Task Complete'.
+Be thorough in your exploration, testing, and reasoning. It's fine if your thinking process is lengthy - quality and completeness are more important than brevity.
+"""
+    runtime = create_runtime(config)
+    call_async_from_sync(runtime.connect)
+    initialize_runtime(runtime, instance)
+
+    # Run the agent
+    state: State | None = asyncio.run(
+        run_controller(
+            config=config,
+            initial_user_action=MessageAction(content=instruction_no_oracle),
+            runtime=runtime,
+            fake_user_response_fn=AGENT_CLS_TO_FAKE_USER_RESPONSE_FN.get(
+                metadata.agent_class
+            ),
+        )
+    )
+    assert state is not None
+    metrics = state.metrics.get() if state.metrics else {}
+
+    test_result = complete_runtime(runtime, instance)
+
+    # history is now available as a stream of events, rather than list of pairs of (Action, Observation)
+    # for compatibility with the existing output format, we can remake the pairs here
+    # remove when it becomes unnecessary
+    histories = compatibility_for_eval_history_pairs(state.history)
+
+    # Save the output
+    output = EvalOutput(
+        instance_id=instance['instance_id'],
+        # instance=instance.to_dict(orient='recorods'),
+        instruction=instruction_no_oracle,
+        metadata=metadata,
+        history=histories,
+        test_result=test_result,
+        metrics=metrics,
+    )
+    return output
+
+
+if __name__ == '__main__':
+    parser = get_parser()
+    parser.add_argument(
+        '-s',
+        '--eval-split',
+        type=str,
+        default='test',
+        choices=['test'],
+        help='data split to evaluate on, must be test',
+    )
+    args, _ = parser.parse_known_args()
+
+    data_split = args.eval_split
+
+    bench = load_dataset(
+        'JetBrains-Research/lca-ci-builds-repair', split=data_split
+    ).to_pandas()
+    # todo: see why 126 is giving problems on inference
+    # todo: see why 145 is giving problems on eval
+    bench = bench[bench['id'] != 126]
+    bench = bench[bench['id'] != 145]
+    # bench = bench.iloc[0:56]
+    # add column instnace_id for compatibility with oh repo, old id column must be kept for lca repo
+    bench['instance_id'] = bench['id'].astype(str)
+
+    llm_config = None
+    if args.llm_config:
+        llm_config = get_llm_config_arg(args.llm_config)
+        # modify_params must be False for evaluation purpose, for reproducibility and accurancy of results
+        llm_config.modify_params = False
+    if llm_config is None:
+        raise ValueError(f'Could not find LLM config: --llm_config {args.llm_config}')
+
+    metadata = make_metadata(
+        llm_config,
+        f'jetbrains-lca-ci--{data_split}',
+        args.agent_cls,
+        args.max_iterations,
+        args.eval_note,
+        args.eval_output_dir,
+    )
+    output_file = os.path.join(metadata.eval_output_dir, 'output.jsonl')
+    instances = prepare_dataset(bench, output_file, args.eval_n_limit)
+
+    run_evaluation(
+        instances, metadata, output_file, args.eval_num_workers, process_instance
+    )

evaluation/benchmarks/lca_ci_build_repair/scripts/eval_infer.sh

@@ -0,0 +1,33 @@
+#!/usr/bin/env bash
+set -eo pipefail
+
+source "evaluation/utils/version_control.sh"
+
+PROCESS_FILEPATH=$1
+if [ -z "$PROCESS_FILEPATH" ]; then
+    echo "Error: PROCESS_FILEPATH is empty. Usage: ./eval_infer.sh <output_file> [instance_id] [dataset_name] [split]"
+    exit 1
+fi
+
+get_openhands_version
+
+PROCESS_FILEPATH=$(realpath $PROCESS_FILEPATH)
+echo "OPENHANDS_VERSION: $OPENHANDS_VERSION"
+echo "PROCESS_FILEPATH: $PROCESS_FILEPATH"
+
+EVAL_NOTE="$OPENHANDS_VERSION"
+if [ -n "$EXP_NAME" ]; then
+  EVAL_NOTE="$EVAL_NOTE-$EXP_NAME"
+fi
+
+function run_eval() {
+  COMMAND="poetry run python ./evaluation/benchmarks/lca_ci_build_repair/eval_infer.py \
+    --predictions-path $PROCESS_FILEPATH "
+
+  echo "RUNNING: $COMMAND"
+  # Run the command
+  eval $COMMAND
+}
+
+unset SANDBOX_ENV_GITHUB_TOKEN # prevent the agent from using the github token to push
+run_eval

evaluation/benchmarks/lca_ci_build_repair/scripts/run_infer.sh

@@ -0,0 +1,27 @@
+#!/usr/bin/env bash
+set -eo pipefail
+
+source "evaluation/utils/version_control.sh"
+
+MODEL_CONFIG=$1
+
+get_openhands_version
+
+echo "OPENHANDS_VERSION: $OPENHANDS_VERSION"
+echo "MODEL_CONFIG: $MODEL_CONFIG"
+
+EVAL_NOTE="$OPENHANDS_VERSION"
+if [ -n "$EXP_NAME" ]; then
+  EVAL_NOTE="$EVAL_NOTE-$EXP_NAME"
+fi
+
+function run_eval() {
+  COMMAND="poetry run python ./evaluation/benchmarks/lca_ci_build_repair/run_infer.py \
+    --llm-config $MODEL_CONFIG "
+
+  # Run the command
+  eval $COMMAND
+}
+
+#unset SANDBOX_ENV_GITHUB_TOKEN # prevent the agent from using the github token to push
+run_eval

evaluation/benchmarks/lca_ci_build_repair/setup.py

@@ -0,0 +1,60 @@
+"""Installs LCA CI Build Repair benchmark with scripts for OH integration."""
+
+import os
+import shutil
+import subprocess
+
+import yaml
+
+
+def setup():
+    # Read config.yaml
+    print('Reading config.yaml')
+    script_dir = os.path.dirname(
+        os.path.abspath(__file__)
+    )  # Get the absolute path of the script
+    config_path = os.path.join(script_dir, 'config.yaml')
+    with open(config_path, 'r') as f:
+        config = yaml.safe_load(f)
+
+    lca_path = config['LCA_PATH']
+    lca_ci_path = os.path.join(
+        lca_path, 'lca-baselines', 'ci-builds-repair', 'ci-builds-repair-benchmark'
+    )
+    repo_url = 'https://github.com/juanmichelini/lca-baselines'
+
+    # Clone the repository to LCA_CI_PATH
+    print(f'Cloning lca-baselines repository from {repo_url} into {lca_path}')
+    result = subprocess.run(
+        ['git', 'clone', repo_url], cwd=lca_path, capture_output=True, text=True
+    )
+    if result.returncode != 0:
+        print(f'Warning cloning repository: {result.stderr}')
+
+    # Clone the repository to LCA_CI_PATH
+    print('Switching branches')
+    result = subprocess.run(
+        ['git', 'switch', 'open-hands-integration'],
+        cwd=lca_ci_path,
+        capture_output=True,
+        text=True,
+    )
+    if result.returncode != 0:
+        print(f'Warning switching repository: {result.stderr}')
+
+    # Move and rename config_lca.yaml (overwrite if exists)
+    lca_ci_config_path = os.path.join(lca_ci_path, 'config.yaml')
+    print(f'Copying config.yaml to {lca_ci_config_path}')
+    shutil.copy(config_path, lca_ci_config_path)
+
+    # Run poetry install in LCA_CI_PATH
+    print(f"Running 'poetry install' in {lca_ci_path}")
+    result = subprocess.run(
+        ['poetry', 'install'], cwd=lca_ci_path, capture_output=True, text=True
+    )
+    if result.returncode != 0:
+        print(f'Warning during poetry install: {result.stderr}')
+
+
+if __name__ == '__main__':
+    setup()

evaluation/benchmarks/logic_reasoning/run_infer.py

@@ -14,7 +14,6 @@ from evaluation.utils.shared import (
     prepare_dataset,
     reset_logger_for_multiprocessing,
     run_evaluation,
-    update_agent_config_for_eval,
 )
 from openhands.controller.state.state import State
 from openhands.core.config import (
@@ -64,8 +63,6 @@ def get_config(
     )
     config.set_llm_config(metadata.llm_config)
     agent_config = config.get_agent_config(metadata.agent_class)
-
-    agent_config = update_agent_config_for_eval(agent_config)
     agent_config.enable_prompt_extensions = False
     return config
 

evaluation/benchmarks/mint/run_infer.py

@@ -19,7 +19,6 @@ from evaluation.utils.shared import (
     prepare_dataset,
     reset_logger_for_multiprocessing,
     run_evaluation,
-    update_agent_config_for_eval,
 )
 from openhands.controller.state.state import State
 from openhands.core.config import (
@@ -122,8 +121,6 @@ def get_config(
     )
     config.set_llm_config(metadata.llm_config)
     agent_config = config.get_agent_config(metadata.agent_class)
-
-    agent_config = update_agent_config_for_eval(agent_config)
     agent_config.enable_prompt_extensions = False
     return config
 

evaluation/benchmarks/ml_bench/run_infer.py

@@ -30,7 +30,6 @@ from evaluation.utils.shared import (
     prepare_dataset,
     reset_logger_for_multiprocessing,
     run_evaluation,
-    update_agent_config_for_eval,
 )
 from openhands.controller.state.state import State
 from openhands.core.config import (
@@ -92,8 +91,6 @@ def get_config(
     )
     config.set_llm_config(metadata.llm_config)
     agent_config = config.get_agent_config(metadata.agent_class)
-
-    agent_config = update_agent_config_for_eval(agent_config)
     agent_config.enable_prompt_extensions = False
     return config
 

evaluation/benchmarks/swe_bench/README.md

@@ -2,6 +2,8 @@
 
 This folder contains the evaluation harness that we built on top of the original [SWE-Bench benchmark](https://www.swebench.com/) ([paper](https://arxiv.org/abs/2310.06770)).
 
+**UPDATE (4/8/2025): We now support running SWT-Bench evaluation! For more details, checkout [the corresponding section](#SWT-Bench-Evaluation).**
+
 **UPDATE (03/27/2025): We now support SWE-Bench multimodal evaluation! Simply use "princeton-nlp/SWE-bench_Multimodal" as the dataset name in the `run_infer.sh` script to evaluate on multimodal instances.**
 
 **UPDATE (2/18/2025): We now support running SWE-Gym using the same evaluation harness here. For more details, checkout [this README](./SWE-Gym.md).**
@@ -141,7 +143,7 @@ With `output.jsonl` file, you can run `eval_infer.sh` to evaluate generated patc
 ./evaluation/benchmarks/swe_bench/scripts/eval_infer.sh $YOUR_OUTPUT_JSONL [instance_id] [dataset_name] [split]
 
 # Example
-./evaluation/benchmarks/swe_bench/scripts/eval_infer.sh evaluation/evaluation_outputs/outputs/swe_bench/CodeActAgent/gpt-4-1106-preview_maxiter_50_N_v1.0/output.jsonl
+./evaluation/benchmarks/swe_bench/scripts/eval_infer.sh evaluation/evaluation_outputs/outputs/princeton-nlp__SWE-bench_Lite/CodeActAgent/gpt-4-1106-preview_maxiter_50_N_v1.0/output.jsonl
 ```
 
 The script now accepts optional arguments:
@@ -182,3 +184,58 @@ To clean-up all existing runtimes that you've already started, run:
 ```bash
 ALLHANDS_API_KEY="YOUR-API-KEY" ./evaluation/utils/scripts/cleanup_remote_runtime.sh
 ```
+
+## SWT-Bench Evaluation
+
+[SWT-Bench](https://swtbench.com/) ([paper](https://arxiv.org/abs/2406.12952)) is a benchmark for evaluating the capability of LLMs at creating unit tests. It is performed on the same instances as SWE-Bench, but requires a separate evaluation harness to capture coverage and issue reproduction. We therefore detail below how to leverage the inference script in this folder to run inference on SWT-Bench and how to use the SWT-Bench evaluation harness to evaluate them.
+
+### Run inference on SWT-Bench
+
+To run inference on SWT-Bench, you can use the same `run_infer.sh` script as described for evaluation on plain SWE-Bench. The only differences is that you need to specify the `mode` parameter to `swt` or `swt-ci` when running the script. For example, to run inference on SWT-Bench Verified, run the following command:
+
+```bash
+./evaluation/benchmarks/swe_bench/scripts/run_infer.sh [model_config] [git-version] [agent] [eval_limit] [max_iter] [num_workers] [swe-dataset] test 1 swt
+
+# Example - This runs evaluation on CodeActAgent for 500 instances on "SWT-bench_Verified"'s test set (corresponding to SWE-bench_Verified), with max 100 iteration per instances, with 1 number of workers running in parallel
+./evaluation/benchmarks/swe_bench/scripts/run_infer.sh llm.eval_gpt4o-2024-11-20 HEAD CodeActAgent 500 100 1 princeton-nlp/SWE-bench_Verified test 1 swt
+```
+
+The two modes `swt` and `swt-ci` have the following effect:
+- `swt`: This mode will change the prompt to instruct the agent to generate reproducing test cases instead of resolving the issue.
+- `swt-ci`: In addition to the changes by `swt`, this mode sets up the CI environment by i) pre-installing the environment in the docker image, such that the test framework can be executed without errors and ii) telling the model the exact command to run the test framework.
+
+### Run evaluation for SWT-bench
+
+The evaluation of these results is done leveraging [the SWT-Bench evaluation harness](https://github.com/logic-star-ai/swt-bench/tree/master).
+
+#### Extracting results into SWT-Bench harness format
+In order to run evaluation of the obtained inference results in the SWT-Bench harness, we transform the results to a format that the SWT-Bench evaluation harness expects.
+
+```bash
+python3 evaluation/benchmarks/swe_bench/scripts/swtbench/convert.py --prediction_file [output.jsonl] > [output_swt.jsonl]
+
+# Example
+python3 evaluation/benchmarks/swe_bench/scripts/swtbench/convert.py --prediction_file "evaluation/evaluation_outputs/outputs/princeton-nlp__SWE-bench_Verified-test/CodeActAgent/gpt-4o-2024-11-20_maxiter_100_N_v0.31.0-no-hint-swt-run_1/output.jsonl" > OpenHands-gpt-4o-2024-11-20.jsonl
+```
+
+#### Running the results in SWT-Bench
+
+Next, we run the [SWT-Bench evaluation harness](https://github.com/logic-star-ai/swt-bench/tree/master) with these results.
+First set-up and validate the setup as described in the harness [here](https://github.com/logic-star-ai/swt-bench/tree/master?tab=readme-ov-file#-set-up).
+Then, run the evaluation with the following command:
+
+```bash
+# Example
+python3 -m src.main \
+    --dataset_name princeton-nlp/SWE-bench_Verified \
+    --predictions_path <pathTo>/OpenHands-gpt-4o-2024-11-20.jsonl \
+    --max_workers 12 \
+    --run_id OpenHands-CodeAct-gpt-4o-2024-11-20  --patch_types vanilla  --build_mode api
+```
+
+The results of the evaluation can be obtained by running the reporting script of the harness.
+
+```bash
+# Example
+python -m src.report run_instance_swt_logs/OpenHands-CodeAct-gpt-4o-2024-11-20/OpenHands__CodeActAgent__gpt-4o-2024-11-20 --dataset verified
+```

evaluation/benchmarks/swe_bench/resource/swt_bench_constants.py

@@ -0,0 +1,842 @@
+# Based on https://github.com/logic-star-ai/swt-bench/blob/master/src/constants.py
+
+# Constants - Installation Specifications
+MAP_VERSION_TO_INSTALL_SKLEARN = {
+    k: {
+        'python': '3.6',
+        'packages': 'numpy scipy cython pytest pandas matplotlib',
+        'install': 'python -m pip install -v --no-use-pep517 --no-build-isolation -e .',
+        'pip_packages': [
+            'cython',
+            'numpy==1.19.2',
+            'setuptools',
+            'scipy==1.5.2',
+        ],
+    }
+    for k in ['0.20', '0.21', '0.22']
+}
+MAP_VERSION_TO_INSTALL_SKLEARN.update(
+    {
+        k: {
+            'python': '3.9',
+            'packages': "'numpy==1.19.2' 'scipy==1.5.2' 'cython==3.0.10' pytest 'pandas<2.0.0' 'matplotlib<3.9.0' setuptools pytest joblib threadpoolctl",
+            'install': 'python -m pip install -v --no-use-pep517 --no-build-isolation -e .',
+            'pip_packages': ['cython', 'setuptools', 'numpy', 'scipy'],
+        }
+        for k in ['1.3', '1.4']
+    }
+)
+MAP_VERSION_TO_INSTALL_FLASK = {
+    '2.0': {
+        'python': '3.9',
+        'packages': 'requirements.txt',
+        'install': 'python -m pip install -e .',
+        'pip_packages': [
+            'setuptools==70.0.0',
+            'Werkzeug==2.3.7',
+            'Jinja2==3.0.1',
+            'itsdangerous==2.1.2',
+            'click==8.0.1',
+            'MarkupSafe==2.1.3',
+        ],
+    },
+    '2.1': {
+        'python': '3.10',
+        'packages': 'requirements.txt',
+        'install': 'python -m pip install -e .',
+        'pip_packages': [
+            'click==8.1.3',
+            'itsdangerous==2.1.2',
+            'Jinja2==3.1.2',
+            'MarkupSafe==2.1.1',
+            'Werkzeug==2.3.7',
+        ],
+    },
+}
+MAP_VERSION_TO_INSTALL_FLASK.update(
+    {
+        k: {
+            'python': '3.11',
+            'packages': 'requirements.txt',
+            'install': 'python -m pip install -e .',
+            'pip_packages': [
+                'click==8.1.3',
+                'itsdangerous==2.1.2',
+                'Jinja2==3.1.2',
+                'MarkupSafe==2.1.1',
+                'Werkzeug==2.3.7',
+            ],
+        }
+        for k in ['2.2', '2.3']
+    }
+)
+MAP_VERSION_TO_INSTALL_DJANGO = {
+    k: {
+        'python': '3.5',
+        'packages': 'requirements.txt',
+        'pre_install': [
+            'apt-get update && apt-get install -y locales',
+            "echo 'en_US UTF-8' > /etc/locale.gen",
+            'locale-gen en_US.UTF-8',
+        ],
+        'install': 'python setup.py install',
+        'pip_packages': ['setuptools'],
+        'eval_commands': [
+            'export LANG=en_US.UTF-8',
+            'export LC_ALL=en_US.UTF-8',
+            'export PYTHONIOENCODING=utf8',
+            'export LANGUAGE=en_US:en',
+        ],
+    }
+    for k in ['1.7', '1.8', '1.9', '1.10', '1.11', '2.0', '2.1', '2.2']
+}
+MAP_VERSION_TO_INSTALL_DJANGO.update(
+    {
+        k: {'python': '3.5', 'install': 'python setup.py install'}
+        for k in ['1.4', '1.5', '1.6']
+    }
+)
+MAP_VERSION_TO_INSTALL_DJANGO.update(
+    {
+        k: {
+            'python': '3.6',
+            'packages': 'requirements.txt',
+            'install': 'python -m pip install -e .',
+            'eval_commands': [
+                "sed -i '/en_US.UTF-8/s/^# //g' /etc/locale.gen && locale-gen",
+                'export LANG=en_US.UTF-8',
+                'export LANGUAGE=en_US:en',
+                'export LC_ALL=en_US.UTF-8',
+            ],
+        }
+        for k in ['3.0', '3.1', '3.2']
+    }
+)
+MAP_VERSION_TO_INSTALL_DJANGO.update(
+    {
+        k: {
+            'python': '3.8',
+            'packages': 'requirements.txt',
+            'install': 'python -m pip install -e .',
+        }
+        for k in ['4.0']
+    }
+)
+MAP_VERSION_TO_INSTALL_DJANGO.update(
+    {
+        k: {
+            'python': '3.9',
+            'packages': 'requirements.txt',
+            'install': 'python -m pip install -e .',
+        }
+        for k in ['4.1', '4.2']
+    }
+)
+MAP_VERSION_TO_INSTALL_DJANGO.update(
+    {
+        k: {
+            'python': '3.11',
+            'packages': 'requirements.txt',
+            'install': 'python -m pip install -e .',
+        }
+        for k in ['5.0']
+    }
+)
+MAP_VERSION_TO_INSTALL_REQUESTS = {
+    k: {'python': '3.9', 'packages': 'pytest', 'install': 'python -m pip install .'}
+    for k in ['0.7', '0.8', '0.9', '0.11', '0.13', '0.14', '1.1', '1.2', '2.0', '2.2']
+    + ['2.3', '2.4', '2.5', '2.7', '2.8', '2.9', '2.10', '2.11', '2.12', '2.17']
+    + ['2.18', '2.19', '2.22', '2.26', '2.25', '2.27', '3.0']
+}
+MAP_VERSION_TO_INSTALL_SEABORN = {
+    k: {
+        'python': '3.9',
+        'install': 'python -m pip install -e .',
+        'pip_packages': [
+            'contourpy==1.1.0',
+            'cycler==0.11.0',
+            'fonttools==4.42.1',
+            'importlib-resources==6.0.1',
+            'kiwisolver==1.4.5',
+            'matplotlib==3.7.2',
+            'numpy==1.25.2',
+            'packaging==23.1',
+            'pandas==1.3.5',  # 2.0.3
+            'pillow==10.0.0',
+            'pyparsing==3.0.9',
+            'pytest',
+            'python-dateutil==2.8.2',
+            'pytz==2023.3.post1',
+            'scipy==1.11.2',
+            'six==1.16.0',
+            'tzdata==2023.1',
+            'zipp==3.16.2',
+        ],
+    }
+    for k in ['0.11']
+}
+MAP_VERSION_TO_INSTALL_SEABORN.update(
+    {
+        k: {
+            'python': '3.9',
+            'install': 'python -m pip install -e .[dev]',
+            'pip_packages': [
+                'contourpy==1.1.0',
+                'cycler==0.11.0',
+                'fonttools==4.42.1',
+                'importlib-resources==6.0.1',
+                'kiwisolver==1.4.5',
+                'matplotlib==3.7.2',
+                'numpy==1.25.2',
+                'packaging==23.1',
+                'pandas==2.0.0',
+                'pillow==10.0.0',
+                'pyparsing==3.0.9',
+                'pytest',
+                'python-dateutil==2.8.2',
+                'pytz==2023.3.post1',
+                'scipy==1.11.2',
+                'six==1.16.0',
+                'tzdata==2023.1',
+                'zipp==3.16.2',
+            ],
+        }
+        for k in ['0.12', '0.13']
+    }
+)
+MAP_VERSION_TO_INSTALL_PYTEST = {
+    k: {'python': '3.9', 'install': 'python -m pip install -e .'}
+    for k in [
+        '4.4',
+        '4.5',
+        '4.6',
+        '5.0',
+        '5.1',
+        '5.2',
+        '5.3',
+        '5.4',
+        '6.0',
+        '6.2',
+        '6.3',
+        '7.0',
+        '7.1',
+        '7.2',
+        '7.4',
+        '8.0',
+    ]
+}
+MAP_VERSION_TO_INSTALL_PYTEST['4.4']['pip_packages'] = [
+    'atomicwrites==1.4.1',
+    'attrs==23.1.0',
+    'more-itertools==10.1.0',
+    'pluggy==0.13.1',
+    'py==1.11.0',
+    'setuptools==68.0.0',
+    'six==1.16.0',
+]
+MAP_VERSION_TO_INSTALL_PYTEST['4.5']['pip_packages'] = [
+    'atomicwrites==1.4.1',
+    'attrs==23.1.0',
+    'more-itertools==10.1.0',
+    'pluggy==0.11.0',
+    'py==1.11.0',
+    'setuptools==68.0.0',
+    'six==1.16.0',
+    'wcwidth==0.2.6',
+]
+MAP_VERSION_TO_INSTALL_PYTEST['4.6']['pip_packages'] = [
+    'atomicwrites==1.4.1',
+    'attrs==23.1.0',
+    'more-itertools==10.1.0',
+    'packaging==23.1',
+    'pluggy==0.13.1',
+    'py==1.11.0',
+    'six==1.16.0',
+    'wcwidth==0.2.6',
+]
+for k in ['5.0', '5.1', '5.2']:
+    MAP_VERSION_TO_INSTALL_PYTEST[k]['pip_packages'] = [
+        'atomicwrites==1.4.1',
+        'attrs==23.1.0',
+        'more-itertools==10.1.0',
+        'packaging==23.1',
+        'pluggy==0.13.1',
+        'py==1.11.0',
+        'wcwidth==0.2.6',
+    ]
+MAP_VERSION_TO_INSTALL_PYTEST['5.3']['pip_packages'] = [
+    'attrs==23.1.0',
+    'more-itertools==10.1.0',
+    'packaging==23.1',
+    'pluggy==0.13.1',
+    'py==1.11.0',
+    'wcwidth==0.2.6',
+]
+MAP_VERSION_TO_INSTALL_PYTEST['5.4']['pip_packages'] = [
+    'py==1.11.0',
+    'packaging==23.1',
+    'attrs==23.1.0',
+    'more-itertools==10.1.0',
+    'pluggy==0.13.1',
+]
+MAP_VERSION_TO_INSTALL_PYTEST['6.0']['pip_packages'] = [
+    'attrs==23.1.0',
+    'iniconfig==2.0.0',
+    'more-itertools==10.1.0',
+    'packaging==23.1',
+    'pluggy==0.13.1',
+    'py==1.11.0',
+    'toml==0.10.2',
+]
+for k in ['6.2', '6.3']:
+    MAP_VERSION_TO_INSTALL_PYTEST[k]['pip_packages'] = [
+        'attrs==23.1.0',
+        'iniconfig==2.0.0',
+        'packaging==23.1',
+        'pluggy==0.13.1',
+        'py==1.11.0',
+        'toml==0.10.2',
+    ]
+MAP_VERSION_TO_INSTALL_PYTEST['7.0']['pip_packages'] = [
+    'attrs==23.1.0',
+    'iniconfig==2.0.0',
+    'packaging==23.1',
+    'pluggy==0.13.1',
+    'py==1.11.0',
+]
+for k in ['7.1', '7.2']:
+    MAP_VERSION_TO_INSTALL_PYTEST[k]['pip_packages'] = [
+        'attrs==23.1.0',
+        'iniconfig==2.0.0',
+        'packaging==23.1',
+        'pluggy==0.13.1',
+        'py==1.11.0',
+        'tomli==2.0.1',
+    ]
+MAP_VERSION_TO_INSTALL_PYTEST['7.4']['pip_packages'] = [
+    'iniconfig==2.0.0',
+    'packaging==23.1',
+    'pluggy==1.3.0',
+    'exceptiongroup==1.1.3',
+    'tomli==2.0.1',
+]
+MAP_VERSION_TO_INSTALL_PYTEST['8.0']['pip_packages'] = [
+    'iniconfig==2.0.0',
+    'packaging==23.1',
+    'pluggy==1.3.0',
+    'exceptiongroup==1.1.3',
+    'tomli==2.0.1',
+]
+MAP_VERSION_TO_INSTALL_MATPLOTLIB = {
+    k: {
+        'python': '3.11',
+        'packages': 'environment.yml',
+        'install': 'python -m pip install -e .',
+        'pre_install': [
+            'apt-get -y update && apt-get -y upgrade && apt-get install -y imagemagick ffmpeg texlive texlive-latex-extra texlive-fonts-recommended texlive-xetex texlive-luatex cm-super dvipng'
+        ],
+        'pip_packages': [
+            'contourpy==1.1.0',
+            'cycler==0.11.0',
+            'fonttools==4.42.1',
+            'ghostscript',
+            'kiwisolver==1.4.5',
+            'numpy==1.25.2',
+            'packaging==23.1',
+            'pillow==10.0.0',
+            'pikepdf',
+            'pyparsing==3.0.9',
+            'python-dateutil==2.8.2',
+            'six==1.16.0',
+            'setuptools==68.1.2',
+            'setuptools-scm==7.1.0',
+            'typing-extensions==4.7.1',
+        ],
+    }
+    for k in ['3.5', '3.6', '3.7']
+}
+MAP_VERSION_TO_INSTALL_MATPLOTLIB.update(
+    {
+        k: {
+            'python': '3.8',
+            'packages': 'requirements.txt',
+            'install': 'python -m pip install -e .',
+            'pre_install': [
+                'apt-get -y update && apt-get -y upgrade && apt-get install -y imagemagick ffmpeg libfreetype6-dev pkg-config texlive texlive-latex-extra texlive-fonts-recommended texlive-xetex texlive-luatex cm-super'
+            ],
+            'pip_packages': ['pytest', 'ipython'],
+        }
+        for k in ['3.1', '3.2', '3.3', '3.4']
+    }
+)
+MAP_VERSION_TO_INSTALL_MATPLOTLIB.update(
+    {
+        k: {
+            'python': '3.7',
+            'packages': 'requirements.txt',
+            'install': 'python -m pip install -e .',
+            'pre_install': [
+                'apt-get -y update && apt-get -y upgrade && apt-get install -y imagemagick ffmpeg libfreetype6-dev pkg-config'
+            ],
+            'pip_packages': ['pytest'],
+        }
+        for k in ['3.0']
+    }
+)
+MAP_VERSION_TO_INSTALL_MATPLOTLIB.update(
+    {
+        k: {
+            'python': '3.5',
+            'install': 'python setup.py build; python setup.py install',
+            'pre_install': [
+                'apt-get -y update && apt-get -y upgrade && && apt-get install -y imagemagick ffmpeg'
+            ],
+            'pip_packages': ['pytest'],
+            'execute_test_as_nonroot': True,
+        }
+        for k in ['2.0', '2.1', '2.2', '1.0', '1.1', '1.2', '1.3', '1.4', '1.5']
+    }
+)
+MAP_VERSION_TO_INSTALL_SPHINX = {
+    k: {
+        'python': '3.9',
+        'pip_packages': ['tox==4.16.0', 'tox-current-env==0.0.11'],
+        'install': 'python -m pip install -e .[test]',
+        'pre_install': ["sed -i 's/pytest/pytest -rA/' tox.ini"],
+    }
+    for k in ['1.5', '1.6', '1.7', '1.8', '2.0', '2.1', '2.2', '2.3', '2.4', '3.0']
+    + ['3.1', '3.2', '3.3', '3.4', '3.5', '4.0', '4.1', '4.2', '4.3', '4.4']
+    + ['4.5', '5.0', '5.1', '5.2', '5.3', '6.0', '6.2', '7.0', '7.1', '7.2']
+}
+for k in ['3.0', '3.1', '3.2', '3.3', '3.4', '3.5', '4.0', '4.1', '4.2', '4.3', '4.4']:
+    MAP_VERSION_TO_INSTALL_SPHINX[k]['pre_install'].extend(
+        [
+            "sed -i 's/Jinja2>=2.3/Jinja2<3.0/' setup.py",
+            "sed -i 's/sphinxcontrib-applehelp/sphinxcontrib-applehelp<=1.0.7/' setup.py",
+            "sed -i 's/sphinxcontrib-devhelp/sphinxcontrib-devhelp<=1.0.5/' setup.py",
+            "sed -i 's/sphinxcontrib-qthelp/sphinxcontrib-qthelp<=1.0.6/' setup.py",
+            "sed -i 's/alabaster>=0.7,<0.8/alabaster>=0.7,<0.7.12/' setup.py",
+            "sed -i \"s/'packaging',/'packaging', 'markupsafe<=2.0.1',/\" setup.py",
+        ]
+    )
+    if k in ['4.2', '4.3', '4.4']:
+        MAP_VERSION_TO_INSTALL_SPHINX[k]['pre_install'].extend(
+            [
+                "sed -i 's/sphinxcontrib-htmlhelp>=2.0.0/sphinxcontrib-htmlhelp>=2.0.0,<=2.0.4/' setup.py",
+                "sed -i 's/sphinxcontrib-serializinghtml>=1.1.5/sphinxcontrib-serializinghtml>=1.1.5,<=1.1.9/' setup.py",
+            ]
+        )
+    elif k == '4.1':
+        MAP_VERSION_TO_INSTALL_SPHINX[k]['pre_install'].extend(
+            [
+                (
+                    "grep -q 'sphinxcontrib-htmlhelp>=2.0.0' setup.py && "
+                    "sed -i 's/sphinxcontrib-htmlhelp>=2.0.0/sphinxcontrib-htmlhelp>=2.0.0,<=2.0.4/' setup.py || "
+                    "sed -i 's/sphinxcontrib-htmlhelp/sphinxcontrib-htmlhelp<=2.0.4/' setup.py"
+                ),
+                (
+                    "grep -q 'sphinxcontrib-serializinghtml>=1.1.5' setup.py && "
+                    "sed -i 's/sphinxcontrib-serializinghtml>=1.1.5/sphinxcontrib-serializinghtml>=1.1.5,<=1.1.9/' setup.py || "
+                    "sed -i 's/sphinxcontrib-serializinghtml/sphinxcontrib-serializinghtml<=1.1.9/' setup.py"
+                ),
+            ]
+        )
+    else:
+        MAP_VERSION_TO_INSTALL_SPHINX[k]['pre_install'].extend(
+            [
+                "sed -i 's/sphinxcontrib-htmlhelp/sphinxcontrib-htmlhelp<=2.0.4/' setup.py",
+                "sed -i 's/sphinxcontrib-serializinghtml/sphinxcontrib-serializinghtml<=1.1.9/' setup.py",
+            ]
+        )
+MAP_VERSION_TO_INSTALL_SPHINX['7.2']['pre_install'] += [
+    'apt-get update && apt-get install -y graphviz'
+]
+MAP_VERSION_TO_INSTALL_ASTROPY = {
+    k: {
+        'python': '3.9',
+        'install': 'python -m pip install -e .[test] --verbose',
+        'pip_packages': [
+            'attrs==23.1.0',
+            'exceptiongroup==1.1.3',
+            'execnet==2.0.2',
+            'hypothesis==6.82.6',
+            'iniconfig==2.0.0',
+            'numpy==1.25.2',
+            'packaging==23.1',
+            'pluggy==1.3.0',
+            'psutil==5.9.5',
+            'pyerfa==2.0.0.3',
+            'pytest-arraydiff==0.5.0',
+            'pytest-astropy-header==0.2.2',
+            'pytest-astropy==0.10.0',
+            'pytest-cov==4.1.0',
+            'pytest-doctestplus==1.0.0',
+            'pytest-filter-subpackage==0.1.2',
+            'pytest-mock==3.11.1',
+            'pytest-openfiles==0.5.0',
+            'pytest-remotedata==0.4.0',
+            'pytest-xdist==3.3.1',
+            'pytest==7.4.0',
+            'PyYAML==6.0.1',
+            'setuptools==68.0.0',
+            'sortedcontainers==2.4.0',
+            'tomli==2.0.1',
+        ],
+    }
+    for k in ['0.1', '0.2', '0.3', '0.4', '1.1', '1.2', '1.3', '3.0', '3.1', '3.2']
+    + ['4.1', '4.2', '4.3', '5.0', '5.1', '5.2']
+}
+for k in ['4.1', '4.2', '4.3', '5.0', '5.1', '5.2']:
+    MAP_VERSION_TO_INSTALL_ASTROPY[k]['pre_install'] = [
+        'sed -i \'s/requires = \\["setuptools",/requires = \\["setuptools==68.0.0",/\' pyproject.toml'
+    ]
+MAP_VERSION_TO_INSTALL_SYMPY = {
+    k: {
+        'python': '3.9',
+        'packages': 'mpmath flake8',
+        'pip_packages': ['mpmath==1.3.0', 'flake8-comprehensions'],
+        'install': 'python -m pip install -e .',
+    }
+    for k in ['0.7', '1.0', '1.1', '1.10', '1.11', '1.12', '1.2', '1.4', '1.5', '1.6']
+    + ['1.7', '1.8', '1.9']
+}
+MAP_VERSION_TO_INSTALL_SYMPY.update(
+    {
+        k: {
+            'python': '3.9',
+            'packages': 'requirements.txt',
+            'install': 'python -m pip install -e .',
+            'pip_packages': ['mpmath==1.3.0'],
+        }
+        for k in ['1.13']
+    }
+)
+MAP_VERSION_TO_INSTALL_PYLINT = {
+    k: {
+        'python': '3.9',
+        'packages': 'requirements.txt',
+        'install': 'python -m pip install -e .',
+    }
+    for k in [
+        '2.10',
+        '2.11',
+        '2.13',
+        '2.14',
+        '2.15',
+        '2.16',
+        '2.17',
+        '2.8',
+        '2.9',
+        '3.0',
+    ]
+}
+MAP_VERSION_TO_INSTALL_PYLINT['2.8']['pip_packages'] = ['pyenchant==3.2']
+MAP_VERSION_TO_INSTALL_PYLINT['2.8']['pre_install'] = [
+    'apt-get update && apt-get install -y libenchant-2-dev hunspell-en-us'
+]
+MAP_VERSION_TO_INSTALL_PYLINT.update(
+    {
+        k: {
+            **MAP_VERSION_TO_INSTALL_PYLINT[k],
+            'pip_packages': ['astroid==3.0.0a6', 'setuptools'],
+        }
+        for k in ['3.0']
+    }
+)
+
+MAP_VERSION_TO_INSTALL_XARRAY = {
+    k: {
+        'python': '3.10',
+        'packages': 'environment.yml',
+        'install': 'python -m pip install -e .',
+        'pip_packages': [
+            'numpy==1.23.0',
+            'packaging==23.1',
+            'pandas==1.5.3',
+            'pytest==7.4.0',
+            'python-dateutil==2.8.2',
+            'pytz==2023.3',
+            'six==1.16.0',
+            'scipy==1.11.1',
+            'setuptools==68.0.0',
+        ],
+        'no_use_env': True,
+    }
+    for k in ['0.12', '0.18', '0.19', '0.20', '2022.03', '2022.06', '2022.09']
+}
+
+MAP_VERSION_TO_INSTALL_SQLFLUFF = {
+    k: {
+        'python': '3.9',
+        'packages': 'requirements.txt',
+        'install': 'python -m pip install -e .',
+    }
+    for k in [
+        '0.10',
+        '0.11',
+        '0.12',
+        '0.13',
+        '0.4',
+        '0.5',
+        '0.6',
+        '0.8',
+        '0.9',
+        '1.0',
+        '1.1',
+        '1.2',
+        '1.3',
+        '1.4',
+        '2.0',
+        '2.1',
+        '2.2',
+    ]
+}
+MAP_VERSION_TO_INSTALL_DBT_CORE = {
+    k: {
+        'python': '3.9',
+        'packages': 'requirements.txt',
+        'install': 'python -m pip install -e .',
+    }
+    for k in [
+        '0.13',
+        '0.14',
+        '0.15',
+        '0.16',
+        '0.17',
+        '0.18',
+        '0.19',
+        '0.20',
+        '0.21',
+        '1.0',
+        '1.1',
+        '1.2',
+        '1.3',
+        '1.4',
+        '1.5',
+        '1.6',
+        '1.7',
+    ]
+}
+MAP_VERSION_TO_INSTALL_PYVISTA = {
+    k: {
+        'python': '3.9',
+        'install': 'python -m pip install -e .',
+        'pip_packages': ['pytest'],
+    }
+    for k in ['0.20', '0.21', '0.22', '0.23']
+}
+MAP_VERSION_TO_INSTALL_PYVISTA.update(
+    {
+        k: {
+            'python': '3.9',
+            'packages': 'requirements.txt',
+            'install': 'python -m pip install -e .',
+            'pip_packages': ['pytest'],
+        }
+        for k in [
+            '0.24',
+            '0.25',
+            '0.26',
+            '0.27',
+            '0.28',
+            '0.29',
+            '0.30',
+            '0.31',
+            '0.32',
+            '0.33',
+            '0.34',
+            '0.35',
+            '0.36',
+            '0.37',
+            '0.38',
+            '0.39',
+            '0.40',
+            '0.41',
+            '0.42',
+            '0.43',
+        ]
+    }
+)
+MAP_VERSION_TO_INSTALL_ASTROID = {
+    k: {
+        'python': '3.9',
+        'install': 'python -m pip install -e .',
+        'pip_packages': ['pytest'],
+    }
+    for k in [
+        '2.10',
+        '2.12',
+        '2.13',
+        '2.14',
+        '2.15',
+        '2.16',
+        '2.5',
+        '2.6',
+        '2.7',
+        '2.8',
+        '2.9',
+        '3.0',
+    ]
+}
+MAP_VERSION_TO_INSTALL_MARSHMALLOW = {
+    k: {
+        'python': '3.9',
+        'install': "python -m pip install -e '.[dev]'",
+    }
+    for k in [
+        '2.18',
+        '2.19',
+        '2.20',
+        '3.0',
+        '3.1',
+        '3.10',
+        '3.11',
+        '3.12',
+        '3.13',
+        '3.15',
+        '3.16',
+        '3.19',
+        '3.2',
+        '3.4',
+        '3.8',
+        '3.9',
+    ]
+}
+MAP_VERSION_TO_INSTALL_PVLIB = {
+    k: {
+        'python': '3.9',
+        'install': 'python -m pip install -e .[all]',
+        'packages': 'pandas scipy',
+        'pip_packages': ['jupyter', 'ipython', 'matplotlib', 'pytest', 'flake8'],
+    }
+    for k in ['0.1', '0.2', '0.3', '0.4', '0.5', '0.6', '0.7', '0.8', '0.9']
+}
+MAP_VERSION_TO_INSTALL_PYDICOM = {
+    k: {'python': '3.6', 'install': 'python -m pip install -e .', 'packages': 'numpy'}
+    for k in [
+        '1.0',
+        '1.1',
+        '1.2',
+        '1.3',
+        '1.4',
+        '2.0',
+        '2.1',
+        '2.2',
+        '2.3',
+        '2.4',
+        '3.0',
+    ]
+}
+MAP_VERSION_TO_INSTALL_PYDICOM.update(
+    {k: {**MAP_VERSION_TO_INSTALL_PYDICOM[k], 'python': '3.8'} for k in ['1.4', '2.0']}
+)
+MAP_VERSION_TO_INSTALL_PYDICOM.update(
+    {k: {**MAP_VERSION_TO_INSTALL_PYDICOM[k], 'python': '3.9'} for k in ['2.1', '2.2']}
+)
+MAP_VERSION_TO_INSTALL_PYDICOM.update(
+    {k: {**MAP_VERSION_TO_INSTALL_PYDICOM[k], 'python': '3.10'} for k in ['2.3']}
+)
+MAP_VERSION_TO_INSTALL_PYDICOM.update(
+    {k: {**MAP_VERSION_TO_INSTALL_PYDICOM[k], 'python': '3.11'} for k in ['2.4', '3.0']}
+)
+MAP_VERSION_TO_INSTALL_HUMANEVAL = {k: {'python': '3.9'} for k in ['1.0']}
+MAP_VERSION_TO_INSTALL_HUMANEVAL_FIX = {
+    k: {'python': '3.10', 'packages': 'pytest'} for k in ['0.0.1']
+}
+
+# Constants - Task Instance Instllation Environment
+MAP_VERSION_TO_INSTALL = {
+    'astropy/astropy': MAP_VERSION_TO_INSTALL_ASTROPY,
+    'dbt-labs/dbt-core': MAP_VERSION_TO_INSTALL_DBT_CORE,
+    'django/django': MAP_VERSION_TO_INSTALL_DJANGO,
+    'matplotlib/matplotlib': MAP_VERSION_TO_INSTALL_MATPLOTLIB,
+    'marshmallow-code/marshmallow': MAP_VERSION_TO_INSTALL_MARSHMALLOW,
+    'mwaskom/seaborn': MAP_VERSION_TO_INSTALL_SEABORN,
+    'pallets/flask': MAP_VERSION_TO_INSTALL_FLASK,
+    'psf/requests': MAP_VERSION_TO_INSTALL_REQUESTS,
+    'pvlib/pvlib-python': MAP_VERSION_TO_INSTALL_PVLIB,
+    'pydata/xarray': MAP_VERSION_TO_INSTALL_XARRAY,
+    'pydicom/pydicom': MAP_VERSION_TO_INSTALL_PYDICOM,
+    'pylint-dev/astroid': MAP_VERSION_TO_INSTALL_ASTROID,
+    'pylint-dev/pylint': MAP_VERSION_TO_INSTALL_PYLINT,
+    'pytest-dev/pytest': MAP_VERSION_TO_INSTALL_PYTEST,
+    'pyvista/pyvista': MAP_VERSION_TO_INSTALL_PYVISTA,
+    'scikit-learn/scikit-learn': MAP_VERSION_TO_INSTALL_SKLEARN,
+    'sphinx-doc/sphinx': MAP_VERSION_TO_INSTALL_SPHINX,
+    'sqlfluff/sqlfluff': MAP_VERSION_TO_INSTALL_SQLFLUFF,
+    'swe-bench/humaneval': MAP_VERSION_TO_INSTALL_HUMANEVAL,
+    'nielstron/humaneval_fix': MAP_VERSION_TO_INSTALL_HUMANEVAL_FIX,
+    'sympy/sympy': MAP_VERSION_TO_INSTALL_SYMPY,
+}
+
+# Constants - Repository Specific Installation Instructions
+MAP_REPO_TO_INSTALL = {}
+
+# Constants - Task Instance Test Frameworks
+TEST_PYTEST_VERBOSE = 'pytest -rA --tb=long -p no:cacheprovider'
+MAP_REPO_TO_TEST_FRAMEWORK_VERBOSE = {
+    'astropy/astropy': {
+        k: TEST_PYTEST_VERBOSE for k in MAP_VERSION_TO_INSTALL_ASTROPY.keys()
+    },
+    'django/django': {
+        k: './tests/runtests.py --verbosity 2 --settings=test_sqlite --parallel 1'
+        for k in MAP_VERSION_TO_INSTALL_DJANGO.keys()
+    },
+    'marshmallow-code/marshmallow': {
+        k: TEST_PYTEST_VERBOSE for k in MAP_VERSION_TO_INSTALL_MARSHMALLOW.keys()
+    },
+    'matplotlib/matplotlib': {
+        k: TEST_PYTEST_VERBOSE for k in MAP_VERSION_TO_INSTALL_MATPLOTLIB.keys()
+    },
+    'mwaskom/seaborn': {
+        k: 'pytest -rA --tb=long' for k in MAP_VERSION_TO_INSTALL_SEABORN.keys()
+    },
+    'pallets/flask': {
+        k: TEST_PYTEST_VERBOSE for k in MAP_VERSION_TO_INSTALL_FLASK.keys()
+    },
+    'psf/requests': {
+        k: TEST_PYTEST_VERBOSE for k in MAP_VERSION_TO_INSTALL_REQUESTS.keys()
+    },
+    'pvlib/pvlib-python': {
+        k: TEST_PYTEST_VERBOSE for k in MAP_VERSION_TO_INSTALL_PVLIB.keys()
+    },
+    'pydata/xarray': {
+        k: TEST_PYTEST_VERBOSE for k in MAP_VERSION_TO_INSTALL_XARRAY.keys()
+    },
+    'pydicom/pydicom': {
+        k: TEST_PYTEST_VERBOSE for k in MAP_VERSION_TO_INSTALL_PYDICOM.keys()
+    },
+    'pylint-dev/astroid': {
+        k: TEST_PYTEST_VERBOSE for k in MAP_VERSION_TO_INSTALL_ASTROID.keys()
+    },
+    'pylint-dev/pylint': {
+        k: TEST_PYTEST_VERBOSE for k in MAP_VERSION_TO_INSTALL_PYLINT.keys()
+    },
+    'pytest-dev/pytest': {
+        k: 'pytest -rA --tb=long' for k in MAP_VERSION_TO_INSTALL_PYTEST.keys()
+    },
+    'pyvista/pyvista': {
+        k: TEST_PYTEST_VERBOSE for k in MAP_VERSION_TO_INSTALL_PYVISTA.keys()
+    },
+    'scikit-learn/scikit-learn': {
+        k: TEST_PYTEST_VERBOSE for k in MAP_VERSION_TO_INSTALL_SKLEARN.keys()
+    },
+    'sphinx-doc/sphinx': {
+        k: 'tox -epy39 -v --' for k in MAP_VERSION_TO_INSTALL_SPHINX.keys()
+    },
+    'sqlfluff/sqlfluff': {
+        k: TEST_PYTEST_VERBOSE for k in MAP_VERSION_TO_INSTALL_SQLFLUFF.keys()
+    },
+    'swe-bench/humaneval': {
+        k: 'python' for k in MAP_VERSION_TO_INSTALL_HUMANEVAL.keys()
+    },
+    'nielstron/humaneval_fix': {
+        k: TEST_PYTEST_VERBOSE for k in MAP_VERSION_TO_INSTALL_HUMANEVAL.keys()
+    },
+    'sympy/sympy': {
+        k: 'bin/test -C --verbose' for k in MAP_VERSION_TO_INSTALL_SYMPY.keys()
+    },
+}
+MAP_REPO_TO_TEST_FRAMEWORK_VERBOSE['django/django']['1.9'] = (
+    './tests/runtests.py --verbosity 2'
+)

evaluation/benchmarks/swe_bench/run_infer.py

@@ -3,7 +3,7 @@ import copy
 import json
 import os
 import tempfile
-from typing import Any
+from typing import Any, Literal
 
 import pandas as pd
 import toml
@@ -17,6 +17,11 @@ from evaluation.benchmarks.swe_bench.binary_patch_utils import (
 from evaluation.benchmarks.swe_bench.resource.mapping import (
     get_instance_resource_factor,
 )
+from evaluation.benchmarks.swe_bench.resource.swt_bench_constants import (
+    MAP_REPO_TO_INSTALL,
+    MAP_REPO_TO_TEST_FRAMEWORK_VERBOSE,
+    MAP_VERSION_TO_INSTALL,
+)
 from evaluation.utils.shared import (
     EvalException,
     EvalMetadata,
@@ -30,7 +35,6 @@ from evaluation.utils.shared import (
     prepare_dataset,
     reset_logger_for_multiprocessing,
     run_evaluation,
-    update_agent_config_for_eval,
     update_llm_config_for_completions_logging,
 )
 from openhands.controller.state.state import State
@@ -56,6 +60,7 @@ from openhands.utils.shutdown_listener import sleep_if_should_continue
 
 USE_HINT_TEXT = os.environ.get('USE_HINT_TEXT', 'false').lower() == 'true'
 RUN_WITH_BROWSING = os.environ.get('RUN_WITH_BROWSING', 'false').lower() == 'true'
+BenchMode = Literal['swe', 'swt', 'swt-ci']
 
 
 AGENT_CLS_TO_FAKE_USER_RESPONSE_FN = {
@@ -69,7 +74,36 @@ def _get_swebench_workspace_dir_name(instance: pd.Series) -> str:
 
 def get_instruction(instance: pd.Series, metadata: EvalMetadata) -> MessageAction:
     workspace_dir_name = _get_swebench_workspace_dir_name(instance)
-    instruction = f"""
+    mode = metadata.details['mode']
+    if mode.startswith('swt'):
+        test_instructions = (
+            f'The following command can be used to run the tests: `{list(MAP_REPO_TO_TEST_FRAMEWORK_VERBOSE[instance.repo].values())[0]}`. Make sure they fail in the expected way.\n'
+            if mode.endswith('ci')
+            else ''
+        )
+        instruction = f"""\
+<uploaded_files>
+/workspace/{workspace_dir_name}
+</uploaded_files>
+I've uploaded a python code repository in the directory {workspace_dir_name}. Consider the following issue description:
+
+<issue_description>
+{instance.problem_statement}
+</issue_description>
+
+
+Can you help me implement the necessary changes to the repository to test whether the issue in <issue_description> was resolved?
+I will take care of all changes to any of the non-test files. This means you DON'T have to modify the actual logic and ONLY have to update test logic and tests!
+Your task is to make the minimal changes to tests files in the /workspace directory to reproduce the issue in the <issue_description>, i.e., such that the generated tests fail in the current state (where the issue is unresolved) and pass when the issue will be resolved.
+Follow these steps to reproduce the issue:
+1. As a first step, it might be a good idea to explore the repo to familiarize yourself with its structure.
+2. Create a script `reproduction.py` to reproduce the error and execute it with `python reproduction.py` using the BashTool, to confirm the error
+3. Edit the sourcecode of the repo to integrate your reproduction script into the test framework
+4. Run the test framework and make sure your tests fail! Only submit FAILING tests! Never submit passing tests.
+{test_instructions}Your thinking should be thorough and so it's fine if it's very long.
+"""
+    else:
+        instruction = f"""
 <uploaded_files>
 /workspace/{workspace_dir_name}
 </uploaded_files>
@@ -232,7 +266,6 @@ def get_config(
         condenser=metadata.condenser_config,
         enable_prompt_extensions=False,
     )
-    agent_config = update_agent_config_for_eval(agent_config)
     config.set_agent_config(agent_config)
     return config
 
@@ -358,6 +391,30 @@ 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)}')
 
+    if metadata.details['mode'] == 'swt-ci':
+        # set up repo
+        setup_commands = []
+        if instance['repo'] in MAP_REPO_TO_INSTALL:
+            setup_commands.append(MAP_REPO_TO_INSTALL[instance['repo']])
+
+        # Run pre-install set up if provided
+        install = MAP_VERSION_TO_INSTALL.get(instance['repo'], {}).get(
+            instance['version'], []
+        )
+        if 'pre_install' in install:
+            for pre_install in install['pre_install']:
+                setup_commands.append(pre_install)
+
+        if 'install' in install:
+            setup_commands.append(install['install'])
+
+        for command in setup_commands:
+            action = CmdRunAction(command=command)
+            action.set_hard_timeout(600)
+            logger.info(action, extra={'msg_type': 'ACTION'})
+            obs = runtime.run_action(action)
+            logger.info(obs, extra={'msg_type': 'OBSERVATION'})
+
     if 'multimodal' not in metadata.dataset.lower():
         # Only for non-multimodal datasets, we need to activate the testbed environment for Python
         # SWE-Bench multimodal datasets are not using the testbed environment
@@ -680,6 +737,13 @@ if __name__ == '__main__':
         default='test',
         help='split to evaluate on',
     )
+    parser.add_argument(
+        '--mode',
+        type=str,
+        default='swe',
+        choices=['swe', 'swt', 'swt-ci'],
+        help="mode to run the evaluation, either 'swe', 'swt', or 'swt-ci'",
+    )
     args, _ = parser.parse_known_args()
 
     # NOTE: It is preferable to load datasets from huggingface datasets and perform post-processing
@@ -716,7 +780,7 @@ if __name__ == '__main__':
     if llm_config is None:
         raise ValueError(f'Could not find LLM config: --llm_config {args.llm_config}')
 
-    details = {}
+    details = {'mode': args.mode}
     _agent_cls = openhands.agenthub.Agent.get_cls(args.agent_cls)
 
     dataset_descrption = (
@@ -866,7 +930,7 @@ if __name__ == '__main__':
                     # Also make sure git_patch is not empty - otherwise we fall back to previous attempt (empty patch is worse than anything else)
                     if (
                         instance['instance_id'] not in added_instance_ids
-                        and instance['test_result']['git_patch'].strip()
+                        and instance['test_result'].get('git_patch', '').strip()
                     ):
                         fout.write(line)
                         added_instance_ids.add(instance['instance_id'])

evaluation/benchmarks/swe_bench/scripts/run_infer.sh

@@ -12,6 +12,7 @@ NUM_WORKERS=$6
 DATASET=$7
 SPLIT=$8
 N_RUNS=$9
+MODE=${10}
 
 if [ -z "$NUM_WORKERS" ]; then
   NUM_WORKERS=1
@@ -45,6 +46,11 @@ if [ -z "$SPLIT" ]; then
   SPLIT="test"
 fi
 
+if [ -z "$MODE" ]; then
+  MODE="swe"
+  echo "MODE not specified, use default $MODE"
+fi
+
 export RUN_WITH_BROWSING=$RUN_WITH_BROWSING
 echo "RUN_WITH_BROWSING: $RUN_WITH_BROWSING"
 
@@ -55,6 +61,10 @@ echo "OPENHANDS_VERSION: $OPENHANDS_VERSION"
 echo "MODEL_CONFIG: $MODEL_CONFIG"
 echo "DATASET: $DATASET"
 echo "SPLIT: $SPLIT"
+echo "MAX_ITER: $MAX_ITER"
+echo "NUM_WORKERS: $NUM_WORKERS"
+echo "COMMIT_HASH: $COMMIT_HASH"
+echo "MODE: $MODE"
 
 # Default to NOT use Hint
 if [ -z "$USE_HINT_TEXT" ]; then
@@ -74,9 +84,13 @@ fi
 if [ -n "$EXP_NAME" ]; then
   EVAL_NOTE="$EVAL_NOTE-$EXP_NAME"
 fi
+# if mode != swe, add mode to the eval note
+if [ "$MODE" != "swe" ]; then
+  EVAL_NOTE="${EVAL_NOTE}-${MODE}"
+fi
 
 function run_eval() {
-  local eval_note=$1
+  local eval_note="${1}"
   COMMAND="poetry run python evaluation/benchmarks/swe_bench/run_infer.py \
     --agent-cls $AGENT \
     --llm-config $MODEL_CONFIG \
@@ -84,7 +98,8 @@ function run_eval() {
     --eval-num-workers $NUM_WORKERS \
     --eval-note $eval_note \
     --dataset $DATASET \
-    --split $SPLIT"
+    --split $SPLIT \
+    --mode $MODE"
 
   if [ -n "$EVAL_LIMIT" ]; then
     echo "EVAL_LIMIT: $EVAL_LIMIT"

evaluation/benchmarks/swe_bench/scripts/swtbench/convert.py

@@ -0,0 +1,95 @@
+import argparse
+import json
+import logging
+
+import unidiff
+
+from evaluation.benchmarks.swe_bench.resource.swt_bench_constants import (
+    MAP_VERSION_TO_INSTALL,
+)
+
+_LOGGER = logging.getLogger(__name__)
+
+
+def remove_setup_files(model_patch: str, instance: dict, delete_setup_changes: bool):
+    """Discard all changes that a patch applies to files changes by the pre_install script and that are reproduction scripts (top-level script)"""
+    setup_files = ['setup.py', 'tox.ini', 'pyproject.toml']
+    pre_install = (
+        MAP_VERSION_TO_INSTALL.get(instance['repo'], {})
+        .get(instance['version'], {})
+        .get('pre_install', [])
+    )
+    relevant_files = (
+        [
+            file
+            for file in setup_files
+            if any(file in install and 'sed' in install for install in pre_install)
+        ]
+        if delete_setup_changes
+        else []
+    )
+    for i in range(10):
+        try:
+            # Appearently outputs.jsonl has .strip() applied, so we try to reconstruct the original patch by adding auxiliary whitespace
+            patch = unidiff.PatchSet(model_patch + i * '\n')
+            break
+        except unidiff.UnidiffParseError:
+            pass
+
+    to_delete = []
+    for i, file in enumerate(patch):
+        if (
+            any(f in file.source_file for f in relevant_files)
+            or file.target_file.count('/') == 1
+        ):
+            to_delete.append(i)
+    for i in reversed(to_delete):
+        del patch[i]
+    return str(patch)
+
+
+def main(
+    prediction_file: str,
+):
+    """Main function to extract the model patches from the OpenHands prediction file and turn them into the expected SWT-Bench format."""
+    with open(prediction_file) as f:
+        for line in f:
+            pred = json.loads(line)
+            try:
+                git_diff = pred['test_result']['git_patch']
+            except KeyError:
+                _LOGGER.warning(
+                    'Warning: No git diff found for instance %s', pred['instance_id']
+                )
+                continue
+            ci_mode = pred['metadata']['details'].get('mode', '') == 'swt-ci'
+            try:
+                git_diff = remove_setup_files(git_diff, pred['instance'], ci_mode)
+            except:  # noqa: E722
+                _LOGGER.warning(
+                    'Warning: Invalid git diff found for instance %s',
+                    pred['instance_id'],
+                )
+            print(
+                json.dumps(
+                    {
+                        'instance_id': pred['instance_id'],
+                        'model_name_or_path': f'{pred["metadata"]["llm_config"]["openrouter_app_name"]}__{pred["metadata"]["agent_class"]}__{pred["metadata"]["llm_config"]["model"]}',
+                        'model_patch': git_diff,
+                        'full_output': json.dumps(pred),
+                    }
+                )
+            )
+
+
+if __name__ == '__main__':
+    parser = argparse.ArgumentParser()
+    parser.add_argument(
+        '--prediction_file',
+        type=str,
+        required=True,
+        help='Path to the prediction file (.../outputs.jsonl)',
+    )
+    args = parser.parse_args()
+
+    main(args.prediction_file)

evaluation/benchmarks/testgeneval/run_infer.py

@@ -30,7 +30,6 @@ from evaluation.utils.shared import (
     prepare_dataset,
     reset_logger_for_multiprocessing,
     run_evaluation,
-    update_agent_config_for_eval,
     update_llm_config_for_completions_logging,
 )
 from openhands.controller.state.state import State
@@ -165,7 +164,6 @@ def get_config(
         condenser=metadata.condenser_config,
         enable_prompt_extensions=False,
     )
-    agent_config = update_agent_config_for_eval(agent_config)
     config.set_agent_config(agent_config)
     return config
 

evaluation/benchmarks/the_agent_company/run_infer.py

@@ -13,10 +13,7 @@ from typing import List
 import yaml
 from browsing import pre_login
 
-from evaluation.utils.shared import (
-    get_default_sandbox_config_for_eval,
-    update_agent_config_for_eval,
-)
+from evaluation.utils.shared import get_default_sandbox_config_for_eval
 from openhands.controller.state.state import State
 from openhands.core.config import (
     AppConfig,
@@ -61,14 +58,12 @@ def get_config(
     )
     config.set_llm_config(llm_config)
     if agent_config:
-        agent_config = update_agent_config_for_eval(agent_config)
         config.set_agent_config(agent_config)
     else:
         logger.info('Agent config not provided, using default settings')
         agent_config = AgentConfig(
             enable_prompt_extensions=False,
         )
-        agent_config = update_agent_config_for_eval(agent_config)
         config.set_agent_config(agent_config)
     return config
 

evaluation/benchmarks/toolqa/run_infer.py

@@ -15,7 +15,6 @@ from evaluation.utils.shared import (
     prepare_dataset,
     reset_logger_for_multiprocessing,
     run_evaluation,
-    update_agent_config_for_eval,
 )
 from openhands.controller.state.state import State
 from openhands.core.config import (
@@ -56,8 +55,6 @@ def get_config(
     )
     config.set_llm_config(metadata.llm_config)
     agent_config = config.get_agent_config(metadata.agent_class)
-
-    agent_config = update_agent_config_for_eval(agent_config)
     agent_config.enable_prompt_extensions = False
     return config
 

evaluation/benchmarks/webarena/run_infer.py

@@ -16,7 +16,6 @@ from evaluation.utils.shared import (
     prepare_dataset,
     reset_logger_for_multiprocessing,
     run_evaluation,
-    update_agent_config_for_eval,
 )
 from openhands.controller.state.state import State
 from openhands.core.config import (
@@ -77,8 +76,6 @@ def get_config(
     )
     config.set_llm_config(metadata.llm_config)
     agent_config = config.get_agent_config(metadata.agent_class)
-
-    agent_config = update_agent_config_for_eval(agent_config)
     agent_config.enable_prompt_extensions = False
     return config
 

evaluation/utils/shared.py

@@ -160,26 +160,6 @@ def cleanup():
         process.join()
 
 
-def update_agent_config_for_eval(
-    agent_config: AgentConfig | None = None,
-) -> AgentConfig:
-    """Update agent config with evaluation-specific settings.
-
-    Args:
-        agent_config: The agent config to update. If None, a new AgentConfig will be created.
-
-    Returns:
-        The updated agent config.
-    """
-    if agent_config is None:
-        agent_config = AgentConfig()
-
-    # Note: We're not disabling repository memory here as requested
-    # This function can be used for other evaluation-specific settings
-
-    return agent_config
-
-
 def make_metadata(
     llm_config: LLMConfig,
     dataset_name: str,
@@ -192,8 +172,6 @@ def make_metadata(
     agent_config: AgentConfig | None = None,
     condenser_config: CondenserConfig | None = None,
 ) -> EvalMetadata:
-    # Update agent config with evaluation-specific settings
-    agent_config = update_agent_config_for_eval(agent_config)
     model_name = llm_config.model.split('/')[-1]
     model_path = model_name.replace(':', '_').replace('@', '-')
     eval_note = f'_N_{eval_note}' if eval_note else ''

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

@@ -223,7 +223,7 @@ describe("ChatInput", () => {
     render(<ChatInput onSubmit={onSubmitMock} />);
     const textarea = screen.getByRole("textbox");
     expect(textarea).toBeInTheDocument();
-    
+
     // The actual verification of maxRows=16 is handled internally by the TextareaAutosize component
     // and affects how many rows the textarea can expand to
   });

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

@@ -1,8 +1,8 @@
 import { afterEach, beforeAll, describe, expect, it, vi } from "vitest";
-import type { Message } from "#/message";
 import { act, screen, waitFor, within } from "@testing-library/react";
 import userEvent from "@testing-library/user-event";
 import { renderWithProviders } from "test-utils";
+import type { Message } from "#/message";
 import { addUserMessage } from "#/state/chat-slice";
 import { SUGGESTIONS } from "#/utils/suggestions";
 import * as ChatSlice from "#/state/chat-slice";
@@ -45,7 +45,15 @@ describe("Empty state", () => {
   it("should render suggestions if empty", () => {
     const { store } = renderWithProviders(<ChatInterface />, {
       preloadedState: {
-        chat: { messages: [] },
+        chat: {
+          messages: [],
+          systemMessage: {
+            content: "",
+            tools: [],
+            openhands_version: null,
+            agent_class: null
+          }
+        },
       },
     });
 
@@ -68,7 +76,15 @@ describe("Empty state", () => {
   it("should render the default suggestions", () => {
     renderWithProviders(<ChatInterface />, {
       preloadedState: {
-        chat: { messages: [] },
+        chat: {
+          messages: [],
+          systemMessage: {
+            content: "",
+            tools: [],
+            openhands_version: null,
+            agent_class: null
+          }
+        },
       },
     });
 
@@ -98,7 +114,15 @@ describe("Empty state", () => {
       const user = userEvent.setup();
       const { store } = renderWithProviders(<ChatInterface />, {
         preloadedState: {
-          chat: { messages: [] },
+          chat: {
+            messages: [],
+            systemMessage: {
+              content: "",
+              tools: [],
+              openhands_version: null,
+              agent_class: null
+            }
+          },
         },
       });
 
@@ -127,7 +151,15 @@ describe("Empty state", () => {
       const user = userEvent.setup();
       const { rerender } = renderWithProviders(<ChatInterface />, {
         preloadedState: {
-          chat: { messages: [] },
+          chat: {
+            messages: [],
+            systemMessage: {
+              content: "",
+              tools: [],
+              openhands_version: null,
+              agent_class: null
+            }
+          },
         },
       });
 

frontend/__tests__/components/chat/expandable-message.test.tsx

@@ -95,6 +95,23 @@ describe("ExpandableMessage", () => {
     expect(screen.queryByTestId("status-icon")).not.toBeInTheDocument();
   });
 
+  it("should render with neutral border and no icon for action messages with undefined success (timeout case)", () => {
+    renderWithProviders(
+      <ExpandableMessage
+        id="OBSERVATION_MESSAGE$RUN"
+        message="Command timed out"
+        type="action"
+        success={undefined}
+      />,
+    );
+    const element = screen.getByText("OBSERVATION_MESSAGE$RUN");
+    const container = element.closest(
+      "div.flex.gap-2.items-center.justify-start",
+    );
+    expect(container).toHaveClass("border-neutral-300");
+    expect(screen.queryByTestId("status-icon")).not.toBeInTheDocument();
+  });
+
   it("should render the out of credits message when the user is out of credits", async () => {
     const getConfigSpy = vi.spyOn(OpenHands, "getConfig");
     // @ts-expect-error - We only care about the APP_MODE and FEATURE_FLAGS fields

frontend/__tests__/components/features/auth-modal.test.tsx

@@ -3,34 +3,46 @@ import { it, describe, expect, vi, beforeAll, afterAll } from "vitest";
 import userEvent from "@testing-library/user-event";
 import { AuthModal } from "#/components/features/waitlist/auth-modal";
 import * as CaptureConsent from "#/utils/handle-capture-consent";
+import * as AuthHook from "#/context/auth-context";
 
 describe("AuthModal", () => {
   beforeAll(() => {
     vi.stubGlobal("location", { href: "" });
+    vi.spyOn(AuthHook, "useAuth").mockReturnValue({
+      providersAreSet: false,
+      setProvidersAreSet: vi.fn(),
+      providerTokensSet: [],
+      setProviderTokensSet: vi.fn()
+    });
   });
 
   afterAll(() => {
     vi.unstubAllGlobals();
+    vi.restoreAllMocks();
   });
 
   it("should render a tos checkbox that is unchecked by default", () => {
-    render(<AuthModal githubAuthUrl={null} />);
+    render(<AuthModal githubAuthUrl={null} appMode="saas" />);
     const checkbox = screen.getByRole("checkbox");
 
     expect(checkbox).not.toBeChecked();
   });
 
-  it("should only enable the GitHub button if the tos checkbox is checked", async () => {
+  it("should only enable the identity provider buttons if the tos checkbox is checked", async () => {
     const user = userEvent.setup();
-    render(<AuthModal githubAuthUrl={null} />);
-    const checkbox = screen.getByRole("checkbox");
-    const button = screen.getByRole("button", { name: "GITHUB$CONNECT_TO_GITHUB" });
+    render(<AuthModal githubAuthUrl={null} appMode="saas" />);
 
-    expect(button).toBeDisabled();
+    const checkbox = screen.getByRole("checkbox");
+    const githubButton = screen.getByRole("button", { name: "GITHUB$CONNECT_TO_GITHUB" });
+    const gitlabButton = screen.getByRole("button", { name: "GITLAB$CONNECT_TO_GITLAB" });
+
+    expect(githubButton).toBeDisabled();
+    expect(gitlabButton).toBeDisabled();
 
     await user.click(checkbox);
 
-    expect(button).not.toBeDisabled();
+    expect(githubButton).not.toBeDisabled();
+    expect(gitlabButton).not.toBeDisabled();
   });
 
   it("should set user analytics consent to true when the user checks the tos checkbox", async () => {
@@ -40,7 +52,7 @@ describe("AuthModal", () => {
     );
 
     const user = userEvent.setup();
-    render(<AuthModal githubAuthUrl="mock-url" />);
+    render(<AuthModal githubAuthUrl="mock-url" appMode="saas" />);
 
     const checkbox = screen.getByRole("checkbox");
     await user.click(checkbox);

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

@@ -56,12 +56,16 @@ describe("GitRepositorySelector", () => {
         full_name: "test/repo1",
         git_provider: "github" as Provider,
         stargazers_count: 100,
+        is_public: true,
+        pushed_at: "2023-01-01T00:00:00Z",
       },
       {
         id: 2,
         full_name: "test/repo2",
         git_provider: "github" as Provider,
         stargazers_count: 200,
+        is_public: true,
+        pushed_at: "2023-01-02T00:00:00Z",
       },
     ];
 

frontend/__tests__/components/features/home/home-header.test.tsx

@@ -0,0 +1,70 @@
+import { QueryClientProvider, QueryClient } from "@tanstack/react-query";
+import { render, screen } from "@testing-library/react";
+import { Provider } from "react-redux";
+import { createRoutesStub } from "react-router";
+import { setupStore } from "test-utils";
+import { describe, expect, it, vi } from "vitest";
+import userEvent from "@testing-library/user-event";
+import { AuthProvider } from "#/context/auth-context";
+import { HomeHeader } from "#/components/features/home/home-header";
+import OpenHands from "#/api/open-hands";
+
+const renderHomeHeader = () => {
+  const RouterStub = createRoutesStub([
+    {
+      Component: HomeHeader,
+      path: "/",
+    },
+    {
+      Component: () => <div data-testid="conversation-screen" />,
+      path: "/conversations/:conversationId",
+    },
+  ]);
+
+  return render(<RouterStub />, {
+    wrapper: ({ children }) => (
+      <Provider store={setupStore()}>
+        <AuthProvider initialProvidersAreSet>
+          <QueryClientProvider client={new QueryClient()}>
+            {children}
+          </QueryClientProvider>
+        </AuthProvider>
+      </Provider>
+    ),
+  });
+};
+
+describe("HomeHeader", () => {
+  it("should create an empty conversation and redirect when pressing the launch from scratch button", async () => {
+    const createConversationSpy = vi.spyOn(OpenHands, "createConversation");
+
+    renderHomeHeader();
+
+    const launchButton = screen.getByRole("button", {
+      name: /launch from scratch/i,
+    });
+    await userEvent.click(launchButton);
+
+    expect(createConversationSpy).toHaveBeenCalledExactlyOnceWith(
+      undefined,
+      undefined,
+      [],
+      undefined,
+    );
+
+    // expect to be redirected to /conversations/:conversationId
+    await screen.findByTestId("conversation-screen");
+  });
+
+  it("should change the launch button text to 'Loading...' when creating a conversation", async () => {
+    renderHomeHeader();
+
+    const launchButton = screen.getByRole("button", {
+      name: /launch from scratch/i,
+    });
+    await userEvent.click(launchButton);
+
+    expect(launchButton).toHaveTextContent(/Loading/i);
+    expect(launchButton).toBeDisabled();
+  });
+});

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

@@ -0,0 +1,229 @@
+import { render, screen, waitFor, within } from "@testing-library/react";
+import { describe, expect, it, vi } from "vitest";
+import userEvent from "@testing-library/user-event";
+import { QueryClientProvider, QueryClient } from "@tanstack/react-query";
+import { setupStore } from "test-utils";
+import { Provider } from "react-redux";
+import { createRoutesStub } from "react-router";
+import OpenHands from "#/api/open-hands";
+import { AuthProvider } from "#/context/auth-context";
+import { GitRepository } from "#/types/git";
+import * as GitService from "#/api/git";
+import { RepoConnector } from "#/components/features/home/repo-connector";
+
+const renderRepoConnector = (initialProvidersAreSet = true) => {
+  const mockRepoSelection = vi.fn();
+  const RouterStub = createRoutesStub([
+    {
+      Component: () => <RepoConnector onRepoSelection={mockRepoSelection} />,
+      path: "/",
+    },
+    {
+      Component: () => <div data-testid="conversation-screen" />,
+      path: "/conversations/:conversationId",
+    },
+    {
+      Component: () => <div data-testid="settings-screen" />,
+      path: "/settings",
+    },
+  ]);
+
+  return render(<RouterStub />, {
+    wrapper: ({ children }) => (
+      <Provider store={setupStore()}>
+        <AuthProvider initialProvidersAreSet={initialProvidersAreSet}>
+          <QueryClientProvider client={new QueryClient()}>
+            {children}
+          </QueryClientProvider>
+        </AuthProvider>
+      </Provider>
+    ),
+  });
+};
+
+const MOCK_RESPOSITORIES: GitRepository[] = [
+  {
+    id: 1,
+    full_name: "rbren/polaris",
+    git_provider: "github",
+    is_public: true,
+  },
+  {
+    id: 2,
+    full_name: "All-Hands-AI/OpenHands",
+    git_provider: "github",
+    is_public: true,
+  },
+];
+
+describe("RepoConnector", () => {
+  it("should render the repository connector section", () => {
+    renderRepoConnector();
+    screen.getByTestId("repo-connector");
+  });
+
+  it("should render the available repositories in the dropdown", async () => {
+    const retrieveUserGitRepositoriesSpy = vi.spyOn(
+      GitService,
+      "retrieveUserGitRepositories",
+    );
+    retrieveUserGitRepositoriesSpy.mockResolvedValue({
+      data: MOCK_RESPOSITORIES,
+      nextPage: null,
+    });
+
+    renderRepoConnector();
+
+    // Wait for the loading state to be replaced with the dropdown
+    const dropdown = await waitFor(() => screen.getByTestId("repo-dropdown"));
+    await userEvent.click(dropdown);
+
+    await waitFor(() => {
+      screen.getByText("rbren/polaris");
+      screen.getByText("All-Hands-AI/OpenHands");
+    });
+  });
+
+  it("should only enable the launch button if a repo is selected", async () => {
+    const retrieveUserGitRepositoriesSpy = vi.spyOn(
+      GitService,
+      "retrieveUserGitRepositories",
+    );
+    retrieveUserGitRepositoriesSpy.mockResolvedValue({
+      data: MOCK_RESPOSITORIES,
+      nextPage: null,
+    });
+
+    renderRepoConnector();
+
+    const launchButton = screen.getByTestId("repo-launch-button");
+    expect(launchButton).toBeDisabled();
+
+    // Wait for the loading state to be replaced with the dropdown
+    const dropdown = await waitFor(() => screen.getByTestId("repo-dropdown"));
+    await userEvent.click(dropdown);
+    await userEvent.click(screen.getByText("rbren/polaris"));
+
+    expect(launchButton).toBeEnabled();
+  });
+
+  it("should render the 'add git(hub|lab) repos' links if saas mode", async () => {
+    const getConfiSpy = vi.spyOn(OpenHands, "getConfig");
+    // @ts-expect-error - only return the APP_MODE
+    getConfiSpy.mockResolvedValue({
+      APP_MODE: "saas",
+    });
+
+    renderRepoConnector();
+
+    await screen.findByText("Add GitHub repos");
+  });
+
+  it("should not render the 'add git(hub|lab) repos' links if oss mode", async () => {
+    const getConfiSpy = vi.spyOn(OpenHands, "getConfig");
+    // @ts-expect-error - only return the APP_MODE
+    getConfiSpy.mockResolvedValue({
+      APP_MODE: "oss",
+    });
+
+    renderRepoConnector();
+
+    expect(screen.queryByText("Add GitHub repos")).not.toBeInTheDocument();
+    expect(screen.queryByText("Add GitLab repos")).not.toBeInTheDocument();
+  });
+
+  it("should create a conversation and redirect with the selected repo when pressing the launch button", async () => {
+    const createConversationSpy = vi.spyOn(OpenHands, "createConversation");
+    const retrieveUserGitRepositoriesSpy = vi.spyOn(
+      GitService,
+      "retrieveUserGitRepositories",
+    );
+    retrieveUserGitRepositoriesSpy.mockResolvedValue({
+      data: MOCK_RESPOSITORIES,
+      nextPage: null,
+    });
+
+    renderRepoConnector();
+
+    const repoConnector = screen.getByTestId("repo-connector");
+    const launchButton =
+      within(repoConnector).getByTestId("repo-launch-button");
+    await userEvent.click(launchButton);
+
+    // repo not selected yet
+    expect(createConversationSpy).not.toHaveBeenCalled();
+
+    // select a repository from the dropdown
+    const dropdown = await waitFor(() => 
+      within(repoConnector).getByTestId("repo-dropdown")
+    );
+    await userEvent.click(dropdown);
+
+    const repoOption = screen.getByText("rbren/polaris");
+    await userEvent.click(repoOption);
+    await userEvent.click(launchButton);
+
+    expect(createConversationSpy).toHaveBeenCalledExactlyOnceWith(
+      {
+        full_name: "rbren/polaris",
+        git_provider: "github",
+        id: 1,
+        is_public: true,
+      },
+      undefined,
+      [],
+      undefined,
+    );
+  });
+
+  it("should change the launch button text to 'Loading...' when creating a conversation", async () => {
+    const retrieveUserGitRepositoriesSpy = vi.spyOn(
+      GitService,
+      "retrieveUserGitRepositories",
+    );
+    retrieveUserGitRepositoriesSpy.mockResolvedValue({
+      data: MOCK_RESPOSITORIES,
+      nextPage: null,
+    });
+
+    renderRepoConnector();
+
+    const launchButton = screen.getByTestId("repo-launch-button");
+
+    // Wait for the loading state to be replaced with the dropdown
+    const dropdown = await waitFor(() => screen.getByTestId("repo-dropdown"));
+    await userEvent.click(dropdown);
+    await userEvent.click(screen.getByText("rbren/polaris"));
+
+    await userEvent.click(launchButton);
+    expect(launchButton).toBeDisabled();
+    expect(launchButton).toHaveTextContent(/Loading/i);
+  });
+
+  it("should not display a button to settings if the user is signed in with their git provider", async () => {
+    renderRepoConnector(true);
+    expect(
+      screen.queryByTestId("navigate-to-settings-button"),
+    ).not.toBeInTheDocument();
+  });
+
+  it("should display a button to settings if the user needs to sign in with their git provider", async () => {
+    renderRepoConnector(false);
+
+    const goToSettingsButton = await screen.findByTestId(
+      "navigate-to-settings-button",
+    );
+    const dropdown = screen.queryByTestId("repo-dropdown");
+    const launchButton = screen.queryByTestId("repo-launch-button");
+    const providerLinks = screen.queryAllByText(/add git(hub|lab) repos/i);
+
+    expect(dropdown).not.toBeInTheDocument();
+    expect(launchButton).not.toBeInTheDocument();
+    expect(providerLinks.length).toBe(0);
+
+    expect(goToSettingsButton).toBeInTheDocument();
+
+    await userEvent.click(goToSettingsButton);
+    await screen.findByTestId("settings-screen");
+  });
+});

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

@@ -0,0 +1,186 @@
+import { render, screen } from "@testing-library/react";
+import { beforeEach, describe, expect, it, vi } from "vitest";
+import { QueryClient, QueryClientProvider } from "@tanstack/react-query";
+import userEvent from "@testing-library/user-event";
+import { Provider } from "react-redux";
+import { createRoutesStub } from "react-router";
+import { setupStore } from "test-utils";
+import { SuggestedTask } from "#/components/features/home/tasks/task.types";
+import OpenHands from "#/api/open-hands";
+import { AuthProvider } from "#/context/auth-context";
+import { TaskCard } from "#/components/features/home/tasks/task-card";
+import * as GitService from "#/api/git";
+import { GitRepository } from "#/types/git";
+import {
+  getFailingChecksPrompt,
+  getMergeConflictPrompt,
+  getOpenIssuePrompt,
+  getUnresolvedCommentsPrompt,
+} from "#/components/features/home/tasks/get-prompt-for-query";
+
+const MOCK_TASK_1: SuggestedTask = {
+  issue_number: 123,
+  repo: "repo1",
+  title: "Task 1",
+  task_type: "MERGE_CONFLICTS",
+};
+
+const MOCK_TASK_2: SuggestedTask = {
+  issue_number: 456,
+  repo: "repo2",
+  title: "Task 2",
+  task_type: "FAILING_CHECKS",
+};
+
+const MOCK_TASK_3: SuggestedTask = {
+  issue_number: 789,
+  repo: "repo3",
+  title: "Task 3",
+  task_type: "UNRESOLVED_COMMENTS",
+};
+
+const MOCK_TASK_4: SuggestedTask = {
+  issue_number: 101112,
+  repo: "repo4",
+  title: "Task 4",
+  task_type: "OPEN_ISSUE",
+};
+
+const MOCK_RESPOSITORIES: GitRepository[] = [
+  { id: 1, full_name: "repo1", git_provider: "github", is_public: true },
+  { id: 2, full_name: "repo2", git_provider: "github", is_public: true },
+  { id: 3, full_name: "repo3", git_provider: "gitlab", is_public: true },
+  { id: 4, full_name: "repo4", git_provider: "gitlab", is_public: true },
+];
+
+const renderTaskCard = (task = MOCK_TASK_1) => {
+  const RouterStub = createRoutesStub([
+    {
+      Component: () => <TaskCard task={task} />,
+      path: "/",
+    },
+    {
+      Component: () => <div data-testid="conversation-screen" />,
+      path: "/conversations/:conversationId",
+    },
+  ]);
+
+  return render(<RouterStub />, {
+    wrapper: ({ children }) => (
+      <Provider store={setupStore()}>
+        <AuthProvider initialProvidersAreSet>
+          <QueryClientProvider client={new QueryClient()}>
+            {children}
+          </QueryClientProvider>
+        </AuthProvider>
+      </Provider>
+    ),
+  });
+};
+
+describe("TaskCard", () => {
+  it("format the issue id", async () => {
+    renderTaskCard();
+
+    const taskId = screen.getByTestId("task-id");
+    expect(taskId).toHaveTextContent(/#123/i);
+  });
+
+  it("should call createConversation when clicking the launch button", async () => {
+    const createConversationSpy = vi.spyOn(OpenHands, "createConversation");
+
+    renderTaskCard();
+
+    const launchButton = screen.getByTestId("task-launch-button");
+    await userEvent.click(launchButton);
+
+    expect(createConversationSpy).toHaveBeenCalled();
+  });
+
+  describe("creating conversation prompts", () => {
+    beforeEach(() => {
+      const retrieveUserGitRepositoriesSpy = vi.spyOn(
+        GitService,
+        "retrieveUserGitRepositories",
+      );
+      retrieveUserGitRepositoriesSpy.mockResolvedValue({
+        data: MOCK_RESPOSITORIES,
+        nextPage: null,
+      });
+    });
+
+    it("should call create conversation with the merge conflict prompt", async () => {
+      const createConversationSpy = vi.spyOn(OpenHands, "createConversation");
+
+      renderTaskCard(MOCK_TASK_1);
+
+      const launchButton = screen.getByTestId("task-launch-button");
+      await userEvent.click(launchButton);
+
+      expect(createConversationSpy).toHaveBeenCalledWith(
+        MOCK_RESPOSITORIES[0],
+        getMergeConflictPrompt(MOCK_TASK_1.issue_number, MOCK_TASK_1.repo),
+        [],
+        undefined,
+      );
+    });
+
+    it("should call create conversation with the failing checks prompt", async () => {
+      const createConversationSpy = vi.spyOn(OpenHands, "createConversation");
+
+      renderTaskCard(MOCK_TASK_2);
+
+      const launchButton = screen.getByTestId("task-launch-button");
+      await userEvent.click(launchButton);
+
+      expect(createConversationSpy).toHaveBeenCalledWith(
+        MOCK_RESPOSITORIES[1],
+        getFailingChecksPrompt(MOCK_TASK_2.issue_number, MOCK_TASK_2.repo),
+        [],
+        undefined,
+      );
+    });
+
+    it("should call create conversation with the unresolved comments prompt", async () => {
+      const createConversationSpy = vi.spyOn(OpenHands, "createConversation");
+
+      renderTaskCard(MOCK_TASK_3);
+
+      const launchButton = screen.getByTestId("task-launch-button");
+      await userEvent.click(launchButton);
+
+      expect(createConversationSpy).toHaveBeenCalledWith(
+        MOCK_RESPOSITORIES[2],
+        getUnresolvedCommentsPrompt(MOCK_TASK_3.issue_number, MOCK_TASK_3.repo),
+        [],
+        undefined,
+      );
+    });
+
+    it("should call create conversation with the open issue prompt", async () => {
+      const createConversationSpy = vi.spyOn(OpenHands, "createConversation");
+
+      renderTaskCard(MOCK_TASK_4);
+
+      const launchButton = screen.getByTestId("task-launch-button");
+      await userEvent.click(launchButton);
+
+      expect(createConversationSpy).toHaveBeenCalledWith(
+        MOCK_RESPOSITORIES[3],
+        getOpenIssuePrompt(MOCK_TASK_4.issue_number, MOCK_TASK_4.repo),
+        [],
+        undefined,
+      );
+    });
+  });
+
+  it("should disable the launch button and update text content when creating a conversation", async () => {
+    renderTaskCard();
+
+    const launchButton = screen.getByTestId("task-launch-button");
+    await userEvent.click(launchButton);
+
+    expect(launchButton).toHaveTextContent(/Loading/i);
+    expect(launchButton).toBeDisabled();
+  });
+});

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

@@ -0,0 +1,113 @@
+import { render, screen, waitFor } from "@testing-library/react";
+import { afterEach, describe, expect, it, vi } from "vitest";
+import { QueryClient, QueryClientProvider } from "@tanstack/react-query";
+import { Provider } from "react-redux";
+import { createRoutesStub } from "react-router";
+import { setupStore } from "test-utils";
+import userEvent from "@testing-library/user-event";
+import { TaskSuggestions } from "#/components/features/home/tasks/task-suggestions";
+import { SuggestionsService } from "#/api/suggestions-service/suggestions-service.api";
+import { MOCK_TASKS } from "#/mocks/task-suggestions-handlers";
+import { AuthProvider } from "#/context/auth-context";
+
+const renderTaskSuggestions = (initialProvidersAreSet = true) => {
+  const RouterStub = createRoutesStub([
+    {
+      Component: TaskSuggestions,
+      path: "/",
+    },
+    {
+      Component: () => <div data-testid="conversation-screen" />,
+      path: "/conversations/:conversationId",
+    },
+    {
+      Component: () => <div data-testid="settings-screen" />,
+      path: "/settings",
+    },
+  ]);
+
+  return render(<RouterStub />, {
+    wrapper: ({ children }) => (
+      <Provider store={setupStore()}>
+        <AuthProvider initialProvidersAreSet={initialProvidersAreSet}>
+          <QueryClientProvider client={new QueryClient()}>
+            {children}
+          </QueryClientProvider>
+        </AuthProvider>
+      </Provider>
+    ),
+  });
+};
+
+describe("TaskSuggestions", () => {
+  const getSuggestedTasksSpy = vi.spyOn(
+    SuggestionsService,
+    "getSuggestedTasks",
+  );
+
+  afterEach(() => {
+    vi.clearAllMocks();
+  });
+
+  it("should render the task suggestions section", () => {
+    renderTaskSuggestions();
+    screen.getByTestId("task-suggestions");
+  });
+
+  it("should render an empty message if there are no tasks", async () => {
+    getSuggestedTasksSpy.mockResolvedValue([]);
+    renderTaskSuggestions();
+    await screen.findByText(/No tasks available/i);
+  });
+
+  it("should render the task groups with the correct titles", async () => {
+    getSuggestedTasksSpy.mockResolvedValue(MOCK_TASKS);
+    renderTaskSuggestions();
+
+    await waitFor(() => {
+      MOCK_TASKS.forEach((taskGroup) => {
+        screen.getByText(taskGroup.title);
+      });
+    });
+  });
+
+  it("should render the task cards with the correct task details", async () => {
+    getSuggestedTasksSpy.mockResolvedValue(MOCK_TASKS);
+    renderTaskSuggestions();
+
+    await waitFor(() => {
+      MOCK_TASKS.forEach((task) => {
+        screen.getByText(task.title);
+      });
+    });
+  });
+
+  it("should render skeletons when loading", async () => {
+    getSuggestedTasksSpy.mockResolvedValue(MOCK_TASKS);
+    renderTaskSuggestions();
+
+    const skeletons = screen.getAllByTestId("task-group-skeleton");
+    expect(skeletons.length).toBeGreaterThan(0);
+
+    await waitFor(() => {
+      MOCK_TASKS.forEach((taskGroup) => {
+        screen.getByText(taskGroup.title);
+      });
+    });
+
+    expect(screen.queryByTestId("task-group-skeleton")).not.toBeInTheDocument();
+  });
+
+  it("should display a button to settings if the user needs to sign in with their git provider", async () => {
+    renderTaskSuggestions(false);
+
+    expect(getSuggestedTasksSpy).not.toHaveBeenCalled();
+    const goToSettingsButton = await screen.findByTestId(
+      "navigate-to-settings-button",
+    );
+    expect(goToSettingsButton).toBeInTheDocument();
+
+    await userEvent.click(goToSettingsButton);
+    await screen.findByTestId("settings-screen");
+  });
+});

frontend/__tests__/components/features/payment/payment-form.test.tsx

@@ -61,25 +61,25 @@ describe("PaymentForm", () => {
     renderPaymentForm();
 
     const topUpInput = await screen.findByTestId("top-up-input");
-    await user.type(topUpInput, "50.12");
+    await user.type(topUpInput, "50");
 
     const topUpButton = screen.getByText("PAYMENT$ADD_CREDIT");
     await user.click(topUpButton);
 
-    expect(createCheckoutSessionSpy).toHaveBeenCalledWith(50.12);
+    expect(createCheckoutSessionSpy).toHaveBeenCalledWith(50);
   });
 
-  it("should round the top-up amount to two decimal places", async () => {
+  it("should only accept integer values", async () => {
     const user = userEvent.setup();
     renderPaymentForm();
 
     const topUpInput = await screen.findByTestId("top-up-input");
-    await user.type(topUpInput, "50.125456");
+    await user.type(topUpInput, "50");
 
     const topUpButton = screen.getByText("PAYMENT$ADD_CREDIT");
     await user.click(topUpButton);
 
-    expect(createCheckoutSessionSpy).toHaveBeenCalledWith(50.13);
+    expect(createCheckoutSessionSpy).toHaveBeenCalledWith(50);
   });
 
   it("should disable the top-up button if the user enters an invalid amount", async () => {
@@ -100,7 +100,7 @@ describe("PaymentForm", () => {
     renderPaymentForm();
 
     const topUpInput = await screen.findByTestId("top-up-input");
-    await user.type(topUpInput, "50.12");
+    await user.type(topUpInput, "50");
 
     const topUpButton = screen.getByText("PAYMENT$ADD_CREDIT");
     await user.click(topUpButton);
@@ -114,7 +114,7 @@ describe("PaymentForm", () => {
       renderPaymentForm();
 
       const topUpInput = await screen.findByTestId("top-up-input");
-      await user.type(topUpInput, "-50.12");
+      await user.type(topUpInput, "-50");
 
       const topUpButton = screen.getByText("PAYMENT$ADD_CREDIT");
       await user.click(topUpButton);
@@ -139,6 +139,8 @@ describe("PaymentForm", () => {
       const user = userEvent.setup();
       renderPaymentForm();
 
+      // With type="number", the browser would prevent non-numeric input,
+      // but we'll test the validation logic anyway
       const topUpInput = await screen.findByTestId("top-up-input");
       await user.type(topUpInput, "abc");
 
@@ -160,5 +162,19 @@ describe("PaymentForm", () => {
 
       expect(createCheckoutSessionSpy).not.toHaveBeenCalled();
     });
+
+    test("user enters a decimal value", async () => {
+      const user = userEvent.setup();
+      renderPaymentForm();
+
+      // With step="1", the browser would validate this, but we'll test our validation logic
+      const topUpInput = await screen.findByTestId("top-up-input");
+      await user.type(topUpInput, "50.5");
+
+      const topUpButton = screen.getByText("PAYMENT$ADD_CREDIT");
+      await user.click(topUpButton);
+
+      expect(createCheckoutSessionSpy).not.toHaveBeenCalled();
+    });
   });
 });

frontend/__tests__/components/file-operations.test.tsx

@@ -1,9 +1,18 @@
 import { render, screen } from "@testing-library/react";
-import { describe, it, expect } from "vitest";
+import { describe, it, expect, vi } from "vitest";
 import { Messages } from "#/components/features/chat/messages";
 import type { Message } from "#/message";
 import { renderWithProviders } from "test-utils";
 
+// Mock the useParams hook to provide a conversationId
+vi.mock("react-router", async () => {
+  const actual = await vi.importActual<typeof import("react-router")>("react-router");
+  return {
+    ...actual,
+    useParams: () => ({ conversationId: "test-conversation-id" }),
+  };
+});
+
 describe("File Operations Messages", () => {
   it("should show success indicator for successful file read operation", () => {
     const messages: Message[] = [

frontend/__tests__/components/terminal/terminal.test.tsx

@@ -5,7 +5,7 @@ import { Command, appendInput, appendOutput } from "#/state/command-slice";
 import Terminal from "#/components/features/terminal/terminal";
 
 const renderTerminal = (commands: Command[] = []) =>
-  renderWithProviders(<Terminal secrets={[]} />, {
+  renderWithProviders(<Terminal />, {
     preloadedState: {
       cmd: {
         commands,
@@ -121,7 +121,7 @@ describe.skip("Terminal", () => {
 
   // This test fails because it expects `disposeMock` to have been called before the component is unmounted.
   it.skip("should dispose the terminal on unmount", () => {
-    const { unmount } = renderWithProviders(<Terminal secrets={[]} />);
+    const { unmount } = renderWithProviders(<Terminal />);
 
     expect(mockTerminal.dispose).not.toHaveBeenCalled();
 

frontend/__tests__/hooks/use-terminal.test.tsx

@@ -1,31 +1,31 @@
 import { beforeAll, describe, expect, it, vi } from "vitest";
-import { render } from "@testing-library/react";
 import { afterEach } from "node:test";
-import { ReactNode } from "react";
 import { useTerminal } from "#/hooks/use-terminal";
 import { Command } from "#/state/command-slice";
+import { AgentState } from "#/types/agent-state";
+import { renderWithProviders } from "../../test-utils";
+
+// Mock the WsClient context
+vi.mock("#/context/ws-client-provider", () => ({
+  useWsClient: () => ({
+    send: vi.fn(),
+    status: "CONNECTED",
+    isLoadingMessages: false,
+    events: [],
+  }),
+}));
 
 interface TestTerminalComponentProps {
   commands: Command[];
-  secrets: string[];
 }
 
 function TestTerminalComponent({
   commands,
-  secrets,
 }: TestTerminalComponentProps) {
-  const ref = useTerminal({ commands, secrets, disabled: false });
+  const ref = useTerminal({ commands });
   return <div ref={ref} />;
 }
 
-interface WrapperProps {
-  children: ReactNode;
-}
-
-function Wrapper({ children }: WrapperProps) {
-  return <div>{children}</div>;
-}
-
 describe("useTerminal", () => {
   const mockTerminal = vi.hoisted(() => ({
     loadAddon: vi.fn(),
@@ -57,8 +57,11 @@ describe("useTerminal", () => {
   });
 
   it("should render", () => {
-    render(<TestTerminalComponent commands={[]} secrets={[]} />, {
-      wrapper: Wrapper,
+    renderWithProviders(<TestTerminalComponent commands={[]} />, {
+      preloadedState: {
+        agent: { curAgentState: AgentState.RUNNING },
+        cmd: { commands: [] },
+      },
     });
   });
 
@@ -68,15 +71,19 @@ describe("useTerminal", () => {
       { content: "hello", type: "output" },
     ];
 
-    render(<TestTerminalComponent commands={commands} secrets={[]} />, {
-      wrapper: Wrapper,
+    renderWithProviders(<TestTerminalComponent commands={commands} />, {
+      preloadedState: {
+        agent: { curAgentState: AgentState.RUNNING },
+        cmd: { commands },
+      },
     });
 
     expect(mockTerminal.writeln).toHaveBeenNthCalledWith(1, "echo hello");
     expect(mockTerminal.writeln).toHaveBeenNthCalledWith(2, "hello");
   });
 
-  it("should hide secrets in the terminal", () => {
+  // This test is no longer relevant as secrets filtering has been removed
+  it.skip("should hide secrets in the terminal", () => {
     const secret = "super_secret_github_token";
     const anotherSecret = "super_secret_another_token";
     const commands: Command[] = [
@@ -87,23 +94,18 @@ describe("useTerminal", () => {
       { content: secret, type: "output" },
     ];
 
-    render(
+    renderWithProviders(
       <TestTerminalComponent
         commands={commands}
-        secrets={[secret, anotherSecret]}
       />,
       {
-        wrapper: Wrapper,
+        preloadedState: {
+          agent: { curAgentState: AgentState.RUNNING },
+          cmd: { commands },
+        },
       },
     );
 
-    // BUG: `vi.clearAllMocks()` does not clear the number of calls
-    // therefore, we need to assume the order of the calls based
-    // on the test order
-    expect(mockTerminal.writeln).toHaveBeenNthCalledWith(
-      3,
-      `export GITHUB_TOKEN=${"*".repeat(10)},${"*".repeat(10)},${"*".repeat(10)}`,
-    );
-    expect(mockTerminal.writeln).toHaveBeenNthCalledWith(4, "*".repeat(10));
+    // This test is no longer relevant as secrets filtering has been removed
   });
 });

frontend/__tests__/routes/home-screen.test.tsx

@@ -0,0 +1,370 @@
+import { render, screen, waitFor, within } from "@testing-library/react";
+import { beforeEach, describe, expect, it, vi } from "vitest";
+import { QueryClientProvider, QueryClient } from "@tanstack/react-query";
+import userEvent from "@testing-library/user-event";
+import { createRoutesStub } from "react-router";
+import { Provider } from "react-redux";
+import { setupStore } from "test-utils";
+import { AxiosError } from "axios";
+import HomeScreen from "#/routes/home";
+import { AuthProvider } from "#/context/auth-context";
+import * as GitService from "#/api/git";
+import { GitRepository } from "#/types/git";
+import OpenHands from "#/api/open-hands";
+import MainApp from "#/routes/root-layout";
+
+const createAxiosNotFoundErrorObject = () =>
+  new AxiosError(
+    "Request failed with status code 404",
+    "ERR_BAD_REQUEST",
+    undefined,
+    undefined,
+    {
+      status: 404,
+      statusText: "Not Found",
+      data: { message: "Settings not found" },
+      headers: {},
+      // @ts-expect-error - we only need the response object for this test
+      config: {},
+    },
+  );
+
+const RouterStub = createRoutesStub([
+  {
+    Component: MainApp,
+    path: "/",
+    children: [
+      {
+        Component: HomeScreen,
+        path: "/",
+      },
+      {
+        Component: () => <div data-testid="conversation-screen" />,
+        path: "/conversations/:conversationId",
+      },
+      {
+        Component: () => <div data-testid="settings-screen" />,
+        path: "/settings",
+      },
+    ],
+  },
+]);
+
+const renderHomeScreen = (initialProvidersAreSet = true) =>
+  render(<RouterStub />, {
+    wrapper: ({ children }) => (
+      <Provider store={setupStore()}>
+        <AuthProvider initialProvidersAreSet={initialProvidersAreSet}>
+          <QueryClientProvider client={new QueryClient()}>
+            {children}
+          </QueryClientProvider>
+        </AuthProvider>
+      </Provider>
+    ),
+  });
+
+const MOCK_RESPOSITORIES: GitRepository[] = [
+  {
+    id: 1,
+    full_name: "octocat/hello-world",
+    git_provider: "github",
+    is_public: true,
+  },
+  {
+    id: 2,
+    full_name: "octocat/earth",
+    git_provider: "github",
+    is_public: true,
+  },
+];
+
+describe("HomeScreen", () => {
+  it("should render", () => {
+    renderHomeScreen();
+    screen.getByTestId("home-screen");
+  });
+
+  it("should render the repository connector and suggested tasks sections", async () => {
+    renderHomeScreen();
+
+    screen.getByTestId("repo-connector");
+    screen.getByTestId("task-suggestions");
+  });
+
+  it("should filter the suggested tasks based on the selected repository", async () => {
+    const retrieveUserGitRepositoriesSpy = vi.spyOn(
+      GitService,
+      "retrieveUserGitRepositories",
+    );
+    retrieveUserGitRepositoriesSpy.mockResolvedValue({
+      data: MOCK_RESPOSITORIES,
+      nextPage: null,
+    });
+
+    renderHomeScreen();
+
+    const taskSuggestions = screen.getByTestId("task-suggestions");
+
+    // Initially, all tasks should be visible
+    await waitFor(() => {
+      within(taskSuggestions).getByText("octocat/hello-world");
+      within(taskSuggestions).getByText("octocat/earth");
+    });
+
+    // Select a repository from the dropdown
+    const repoConnector = screen.getByTestId("repo-connector");
+
+    const dropdown = within(repoConnector).getByTestId("repo-dropdown");
+    await userEvent.click(dropdown);
+
+    const repoOption = screen.getAllByText("octocat/hello-world")[1];
+    await userEvent.click(repoOption);
+
+    // After selecting a repository, only tasks related to that repository should be visible
+    await waitFor(() => {
+      within(taskSuggestions).getByText("octocat/hello-world");
+      expect(
+        within(taskSuggestions).queryByText("octocat/earth"),
+      ).not.toBeInTheDocument();
+    });
+  });
+
+  it("should reset the filtered tasks when the selected repository is cleared", async () => {
+    const retrieveUserGitRepositoriesSpy = vi.spyOn(
+      GitService,
+      "retrieveUserGitRepositories",
+    );
+    retrieveUserGitRepositoriesSpy.mockResolvedValue({
+      data: MOCK_RESPOSITORIES,
+      nextPage: null,
+    });
+
+    renderHomeScreen();
+
+    const taskSuggestions = screen.getByTestId("task-suggestions");
+
+    // Initially, all tasks should be visible
+    await waitFor(() => {
+      within(taskSuggestions).getByText("octocat/hello-world");
+      within(taskSuggestions).getByText("octocat/earth");
+    });
+
+    // Select a repository from the dropdown
+    const repoConnector = screen.getByTestId("repo-connector");
+
+    const dropdown = within(repoConnector).getByTestId("repo-dropdown");
+    await userEvent.click(dropdown);
+
+    const repoOption = screen.getAllByText("octocat/hello-world")[1];
+    await userEvent.click(repoOption);
+
+    // After selecting a repository, only tasks related to that repository should be visible
+    await waitFor(() => {
+      within(taskSuggestions).getByText("octocat/hello-world");
+      expect(
+        within(taskSuggestions).queryByText("octocat/earth"),
+      ).not.toBeInTheDocument();
+    });
+
+    // Clear the selected repository
+    await userEvent.clear(dropdown);
+
+    // All tasks should be visible again
+    await waitFor(() => {
+      within(taskSuggestions).getByText("octocat/hello-world");
+      within(taskSuggestions).getByText("octocat/earth");
+    });
+  });
+
+  describe("launch buttons", () => {
+    const setupLaunchButtons = async () => {
+      let headerLaunchButton = screen.getByTestId("header-launch-button");
+      let repoLaunchButton = screen.getByTestId("repo-launch-button");
+      let tasksLaunchButtons =
+        await screen.findAllByTestId("task-launch-button");
+
+      // Select a repository from the dropdown to enable the repo launch button
+      const repoConnector = screen.getByTestId("repo-connector");
+      const dropdown = within(repoConnector).getByTestId("repo-dropdown");
+      await userEvent.click(dropdown);
+      const repoOption = screen.getAllByText("octocat/hello-world")[1];
+      await userEvent.click(repoOption);
+
+      expect(headerLaunchButton).not.toBeDisabled();
+      expect(repoLaunchButton).not.toBeDisabled();
+      tasksLaunchButtons.forEach((button) => {
+        expect(button).not.toBeDisabled();
+      });
+
+      headerLaunchButton = screen.getByTestId("header-launch-button");
+      repoLaunchButton = screen.getByTestId("repo-launch-button");
+      tasksLaunchButtons = await screen.findAllByTestId("task-launch-button");
+
+      return {
+        headerLaunchButton,
+        repoLaunchButton,
+        tasksLaunchButtons,
+      };
+    };
+
+    beforeEach(() => {
+      const retrieveUserGitRepositoriesSpy = vi.spyOn(
+        GitService,
+        "retrieveUserGitRepositories",
+      );
+      retrieveUserGitRepositoriesSpy.mockResolvedValue({
+        data: MOCK_RESPOSITORIES,
+        nextPage: null,
+      });
+    });
+
+    it("should disable the other launch buttons when the header launch button is clicked", async () => {
+      renderHomeScreen();
+      const { headerLaunchButton, repoLaunchButton } =
+        await setupLaunchButtons();
+
+      const tasksLaunchButtonsAfter =
+        await screen.findAllByTestId("task-launch-button");
+
+      // All other buttons should be disabled when the header button is clicked
+      await userEvent.click(headerLaunchButton);
+
+      expect(headerLaunchButton).toBeDisabled();
+      expect(repoLaunchButton).toBeDisabled();
+      tasksLaunchButtonsAfter.forEach((button) => {
+        expect(button).toBeDisabled();
+      });
+    });
+
+    it("should disable the other launch buttons when the repo launch button is clicked", async () => {
+      renderHomeScreen();
+      const { headerLaunchButton, repoLaunchButton } =
+        await setupLaunchButtons();
+
+      const tasksLaunchButtonsAfter =
+        await screen.findAllByTestId("task-launch-button");
+
+      // All other buttons should be disabled when the repo button is clicked
+      await userEvent.click(repoLaunchButton);
+
+      expect(headerLaunchButton).toBeDisabled();
+      expect(repoLaunchButton).toBeDisabled();
+      tasksLaunchButtonsAfter.forEach((button) => {
+        expect(button).toBeDisabled();
+      });
+    });
+
+    it("should disable the other launch buttons when any task launch button is clicked", async () => {
+      renderHomeScreen();
+      const { headerLaunchButton, repoLaunchButton, tasksLaunchButtons } =
+        await setupLaunchButtons();
+
+      const tasksLaunchButtonsAfter =
+        await screen.findAllByTestId("task-launch-button");
+
+      // All other buttons should be disabled when the task button is clicked
+      await userEvent.click(tasksLaunchButtons[0]);
+
+      expect(headerLaunchButton).toBeDisabled();
+      expect(repoLaunchButton).toBeDisabled();
+      tasksLaunchButtonsAfter.forEach((button) => {
+        expect(button).toBeDisabled();
+      });
+    });
+  });
+
+  it("should hide the suggested tasks section if not authed with git(hub|lab)", async () => {
+    renderHomeScreen(false);
+
+    const taskSuggestions = screen.queryByTestId("task-suggestions");
+    const repoConnector = screen.getByTestId("repo-connector");
+
+    expect(taskSuggestions).not.toBeInTheDocument();
+    expect(repoConnector).toBeInTheDocument();
+  });
+});
+
+describe("Settings 404", () => {
+  const getConfigSpy = vi.spyOn(OpenHands, "getConfig");
+  const getSettingsSpy = vi.spyOn(OpenHands, "getSettings");
+
+  it("should open the settings modal if GET /settings fails with a 404", async () => {
+    const error = createAxiosNotFoundErrorObject();
+    getSettingsSpy.mockRejectedValue(error);
+
+    renderHomeScreen();
+
+    const settingsModal = await screen.findByTestId("ai-config-modal");
+    expect(settingsModal).toBeInTheDocument();
+  });
+
+  it("should navigate to the settings screen when clicking the advanced settings button", async () => {
+    const error = createAxiosNotFoundErrorObject();
+    getSettingsSpy.mockRejectedValue(error);
+
+    const user = userEvent.setup();
+    renderHomeScreen();
+
+    const settingsScreen = screen.queryByTestId("settings-screen");
+    expect(settingsScreen).not.toBeInTheDocument();
+
+    const settingsModal = await screen.findByTestId("ai-config-modal");
+    expect(settingsModal).toBeInTheDocument();
+
+    const advancedSettingsButton = await screen.findByTestId(
+      "advanced-settings-link",
+    );
+    await user.click(advancedSettingsButton);
+
+    const settingsScreenAfter = await screen.findByTestId("settings-screen");
+    expect(settingsScreenAfter).toBeInTheDocument();
+
+    const settingsModalAfter = screen.queryByTestId("ai-config-modal");
+    expect(settingsModalAfter).not.toBeInTheDocument();
+  });
+
+  it("should not open the settings modal if GET /settings fails but is SaaS mode", async () => {
+    // @ts-expect-error - we only need APP_MODE for this test
+    getConfigSpy.mockResolvedValue({
+      APP_MODE: "saas",
+      FEATURE_FLAGS: {
+        ENABLE_BILLING: false,
+        HIDE_LLM_SETTINGS: false,
+      },
+    });
+    const error = createAxiosNotFoundErrorObject();
+    getSettingsSpy.mockRejectedValue(error);
+
+    renderHomeScreen();
+
+    // small hack to wait for the modal to not appear
+    await expect(
+      screen.findByTestId("ai-config-modal", {}, { timeout: 1000 }),
+    ).rejects.toThrow();
+  });
+});
+
+describe("Setup Payment modal", () => {
+  const getConfigSpy = vi.spyOn(OpenHands, "getConfig");
+  const getSettingsSpy = vi.spyOn(OpenHands, "getSettings");
+
+  it("should only render if SaaS mode and is new user", async () => {
+    // @ts-expect-error - we only need the APP_MODE for this test
+    getConfigSpy.mockResolvedValue({
+      APP_MODE: "saas",
+      FEATURE_FLAGS: {
+        ENABLE_BILLING: true,
+        HIDE_LLM_SETTINGS: false,
+      },
+    });
+    const error = createAxiosNotFoundErrorObject();
+    getSettingsSpy.mockRejectedValue(error);
+
+    renderHomeScreen();
+
+    const setupPaymentModal = await screen.findByTestId(
+      "proceed-to-stripe-button",
+    );
+    expect(setupPaymentModal).toBeInTheDocument();
+  });
+});

frontend/__tests__/routes/home.test.tsx

@@ -1,177 +0,0 @@
-import { createRoutesStub } from "react-router";
-import { afterEach, describe, expect, it, vi } from "vitest";
-import { renderWithProviders } from "test-utils";
-import userEvent from "@testing-library/user-event";
-import { screen } from "@testing-library/react";
-import { AxiosError } from "axios";
-import MainApp from "#/routes/root-layout";
-import SettingsScreen from "#/routes/settings";
-import Home from "#/routes/home";
-import OpenHands from "#/api/open-hands";
-
-const createAxiosNotFoundErrorObject = () =>
-  new AxiosError(
-    "Request failed with status code 404",
-    "ERR_BAD_REQUEST",
-    undefined,
-    undefined,
-    {
-      status: 404,
-      statusText: "Not Found",
-      data: { message: "Settings not found" },
-      headers: {},
-      // @ts-expect-error - we only need the response object for this test
-      config: {},
-    },
-  );
-
-const getSettingsSpy = vi.spyOn(OpenHands, "getSettings");
-
-const RouterStub = createRoutesStub([
-  {
-    // layout route
-    Component: MainApp,
-    path: "/",
-    children: [
-      {
-        // home route
-        Component: Home,
-        path: "/",
-      },
-      {
-        Component: SettingsScreen,
-        path: "/settings",
-      },
-    ],
-  },
-]);
-
-afterEach(() => {
-  vi.clearAllMocks();
-});
-
-describe("Home Screen", () => {
-  const getConfigSpy = vi.spyOn(OpenHands, "getConfig");
-
-  it("should render the home screen", () => {
-    renderWithProviders(<RouterStub initialEntries={["/"]} />);
-  });
-
-  it("should navigate to the settings screen when the settings button is clicked", async () => {
-    const user = userEvent.setup();
-    renderWithProviders(<RouterStub initialEntries={["/"]} />);
-
-    const settingsButton = await screen.findByTestId("settings-button");
-    await user.click(settingsButton);
-
-    const settingsScreen = await screen.findByTestId("settings-screen");
-    expect(settingsScreen).toBeInTheDocument();
-  });
-
-  it("should navigate to the settings when pressing 'Connect to GitHub' if the user isn't authenticated", async () => {
-    // @ts-expect-error - we only need APP_MODE for this test
-    getConfigSpy.mockResolvedValue({
-      APP_MODE: "oss",
-      FEATURE_FLAGS: {
-        ENABLE_BILLING: false,
-        HIDE_LLM_SETTINGS: false,
-      },
-    });
-    const user = userEvent.setup();
-    renderWithProviders(<RouterStub initialEntries={["/"]} />);
-
-    const connectToGitHubButton =
-      await screen.findByTestId("connect-to-github");
-    await user.click(connectToGitHubButton);
-
-    const settingsScreen = await screen.findByTestId("settings-screen");
-    expect(settingsScreen).toBeInTheDocument();
-  });
-});
-
-describe("Settings 404", () => {
-  const getConfigSpy = vi.spyOn(OpenHands, "getConfig");
-
-  it("should open the settings modal if GET /settings fails with a 404", async () => {
-    const error = createAxiosNotFoundErrorObject();
-    getSettingsSpy.mockRejectedValue(error);
-
-    renderWithProviders(<RouterStub initialEntries={["/"]} />);
-
-    const settingsModal = await screen.findByTestId("ai-config-modal");
-    expect(settingsModal).toBeInTheDocument();
-  });
-
-  it("should navigate to the settings screen when clicking the advanced settings button", async () => {
-    const error = createAxiosNotFoundErrorObject();
-    getSettingsSpy.mockRejectedValue(error);
-
-    const user = userEvent.setup();
-    renderWithProviders(<RouterStub initialEntries={["/"]} />);
-
-    const settingsScreen = screen.queryByTestId("settings-screen");
-    expect(settingsScreen).not.toBeInTheDocument();
-
-    const settingsModal = await screen.findByTestId("ai-config-modal");
-    expect(settingsModal).toBeInTheDocument();
-
-    const advancedSettingsButton = await screen.findByTestId(
-      "advanced-settings-link",
-    );
-    await user.click(advancedSettingsButton);
-
-    const settingsScreenAfter = await screen.findByTestId("settings-screen");
-    expect(settingsScreenAfter).toBeInTheDocument();
-
-    const settingsModalAfter = screen.queryByTestId("ai-config-modal");
-    expect(settingsModalAfter).not.toBeInTheDocument();
-  });
-
-  it("should not open the settings modal if GET /settings fails but is SaaS mode", async () => {
-    // @ts-expect-error - we only need APP_MODE for this test
-    getConfigSpy.mockResolvedValue({
-      APP_MODE: "saas",
-      FEATURE_FLAGS: {
-        ENABLE_BILLING: false,
-        HIDE_LLM_SETTINGS: false,
-      },
-    });
-    const error = createAxiosNotFoundErrorObject();
-    getSettingsSpy.mockRejectedValue(error);
-
-    renderWithProviders(<RouterStub initialEntries={["/"]} />);
-
-    // small hack to wait for the modal to not appear
-    await expect(
-      screen.findByTestId("ai-config-modal", {}, { timeout: 1000 }),
-    ).rejects.toThrow();
-  });
-});
-
-describe("Setup Payment modal", () => {
-  const getConfigSpy = vi.spyOn(OpenHands, "getConfig");
-
-  afterEach(() => {
-    vi.resetAllMocks();
-  });
-
-  it("should only render if SaaS mode and is new user", async () => {
-    // @ts-expect-error - we only need the APP_MODE for this test
-    getConfigSpy.mockResolvedValue({
-      APP_MODE: "saas",
-      FEATURE_FLAGS: {
-        ENABLE_BILLING: true,
-        HIDE_LLM_SETTINGS: false,
-      },
-    });
-    const error = createAxiosNotFoundErrorObject();
-    getSettingsSpy.mockRejectedValue(error);
-
-    renderWithProviders(<RouterStub initialEntries={["/"]} />);
-
-    const setupPaymentModal = await screen.findByTestId(
-      "proceed-to-stripe-button",
-    );
-    expect(setupPaymentModal).toBeInTheDocument();
-  });
-});

frontend/__tests__/routes/settings-with-payment.test.tsx

@@ -43,10 +43,12 @@ describe("Settings Billing", () => {
 
     renderSettingsScreen();
 
-    await waitFor(() => {
-      const navbar = screen.queryByTestId("settings-navbar");
-      expect(navbar).not.toBeInTheDocument();
-    });
+    // Wait for the settings screen to be rendered
+    await screen.findByTestId("settings-screen");
+
+    // Then check that the navbar is not present
+    const navbar = screen.queryByTestId("settings-navbar");
+    expect(navbar).not.toBeInTheDocument();
   });
 
   it("should render the navbar if SaaS mode", async () => {

frontend/__tests__/services/observations.test.ts

@@ -48,4 +48,4 @@ describe("Observations Service", () => {
       });
     });
   });
-});
+});

frontend/__tests__/utils/group-suggested-tasks.test.ts

@@ -0,0 +1,91 @@
+import { expect, test } from "vitest";
+import {
+  SuggestedTask,
+  SuggestedTaskGroup,
+} from "#/components/features/home/tasks/task.types";
+import { groupSuggestedTasks } from "#/utils/group-suggested-tasks";
+
+const rawTasks: SuggestedTask[] = [
+  {
+    issue_number: 1,
+    repo: "repo1",
+    title: "Task 1",
+    task_type: "MERGE_CONFLICTS",
+  },
+  {
+    issue_number: 2,
+    repo: "repo1",
+    title: "Task 2",
+    task_type: "FAILING_CHECKS",
+  },
+  {
+    issue_number: 3,
+    repo: "repo2",
+    title: "Task 3",
+    task_type: "UNRESOLVED_COMMENTS",
+  },
+  {
+    issue_number: 4,
+    repo: "repo2",
+    title: "Task 4",
+    task_type: "OPEN_ISSUE",
+  },
+  {
+    issue_number: 5,
+    repo: "repo3",
+    title: "Task 5",
+    task_type: "FAILING_CHECKS",
+  },
+];
+
+const groupedTasks: SuggestedTaskGroup[] = [
+  {
+    title: "repo1",
+    tasks: [
+      {
+        issue_number: 1,
+        repo: "repo1",
+        title: "Task 1",
+        task_type: "MERGE_CONFLICTS",
+      },
+      {
+        issue_number: 2,
+        repo: "repo1",
+        title: "Task 2",
+        task_type: "FAILING_CHECKS",
+      },
+    ],
+  },
+  {
+    title: "repo2",
+    tasks: [
+      {
+        issue_number: 3,
+        repo: "repo2",
+        title: "Task 3",
+        task_type: "UNRESOLVED_COMMENTS",
+      },
+      {
+        issue_number: 4,
+        repo: "repo2",
+        title: "Task 4",
+        task_type: "OPEN_ISSUE",
+      },
+    ],
+  },
+  {
+    title: "repo3",
+    tasks: [
+      {
+        issue_number: 5,
+        repo: "repo3",
+        title: "Task 5",
+        task_type: "FAILING_CHECKS",
+      },
+    ],
+  },
+];
+
+test("groupSuggestedTasks", () => {
+  expect(groupSuggestedTasks(rawTasks)).toEqual(groupedTasks);
+});

frontend/package.json

@@ -1,6 +1,6 @@
 {
   "name": "openhands-frontend",
-  "version": "0.32.0",
+  "version": "0.34.0",
   "private": true,
   "type": "module",
   "engines": {
@@ -8,28 +8,30 @@
   },
   "dependencies": {
     "@heroui/react": "2.7.6",
+    "@microlink/react-json-view": "^1.26.1",
     "@monaco-editor/react": "^4.7.0-rc.0",
-    "@react-router/node": "^7.5.0",
-    "@react-router/serve": "^7.5.0",
-    "@react-types/shared": "^3.28.0",
-    "@reduxjs/toolkit": "^2.6.1",
+    "@react-router/node": "^7.5.1",
+    "@react-router/serve": "^7.5.1",
+    "@react-types/shared": "^3.29.0",
+    "@reduxjs/toolkit": "^2.7.0",
     "@stripe/react-stripe-js": "^3.6.0",
-    "@stripe/stripe-js": "^7.0.0",
-    "@tanstack/react-query": "^5.72.1",
-    "@vitejs/plugin-react": "^4.3.2",
+    "@stripe/stripe-js": "^7.1.0",
+    "@tanstack/react-query": "^5.74.4",
+    "@vitejs/plugin-react": "^4.4.0",
     "@xterm/addon-fit": "^0.10.0",
     "@xterm/xterm": "^5.4.0",
     "axios": "^1.8.4",
     "clsx": "^2.1.1",
     "eslint-config-airbnb-typescript": "^18.0.0",
-    "framer-motion": "^12.6.3",
-    "i18next": "^24.2.3",
-    "i18next-browser-languagedetector": "^8.0.4",
+    "framer-motion": "^12.7.4",
+    "i18next": "^25.0.0",
+    "i18next-browser-languagedetector": "^8.0.5",
     "i18next-http-backend": "^3.0.2",
     "isbot": "^5.1.25",
     "jose": "^6.0.10",
+    "lucide-react": "^0.501.0",
     "monaco-editor": "^0.52.2",
-    "posthog-js": "^1.235.0",
+    "posthog-js": "^1.236.2",
     "react": "^19.1.0",
     "react-dom": "^19.1.0",
     "react-highlight": "^0.15.0",
@@ -38,14 +40,14 @@
     "react-icons": "^5.5.0",
     "react-markdown": "^10.1.0",
     "react-redux": "^9.2.0",
-    "react-router": "^7.5.0",
+    "react-router": "^7.5.1",
     "react-syntax-highlighter": "^15.6.1",
     "react-textarea-autosize": "^8.5.9",
     "remark-gfm": "^4.0.1",
     "sirv-cli": "^3.0.1",
     "socket.io-client": "^4.8.1",
     "tailwind-merge": "^3.2.0",
-    "vite": "^6.2.5",
+    "vite": "^6.3.2",
     "web-vitals": "^3.5.2",
     "ws": "^8.18.1"
   },
@@ -79,16 +81,16 @@
     "@babel/traverse": "^7.27.0",
     "@babel/types": "^7.27.0",
     "@mswjs/socket.io-binding": "^0.1.1",
-    "@playwright/test": "^1.51.1",
-    "@react-router/dev": "^7.5.0",
+    "@playwright/test": "^1.52.0",
+    "@react-router/dev": "^7.5.1",
     "@tailwindcss/typography": "^0.5.16",
-    "@tanstack/eslint-plugin-query": "^5.72.1",
+    "@tanstack/eslint-plugin-query": "^5.73.3",
     "@testing-library/dom": "^10.4.0",
     "@testing-library/jest-dom": "^6.6.1",
     "@testing-library/react": "^16.3.0",
     "@testing-library/user-event": "^14.6.1",
-    "@types/node": "^22.14.0",
-    "@types/react": "^19.1.0",
+    "@types/node": "^22.14.1",
+    "@types/react": "^19.1.2",
     "@types/react-dom": "^19.1.1",
     "@types/react-highlight": "^0.12.8",
     "@types/react-syntax-highlighter": "^15.5.13",
@@ -101,7 +103,7 @@
     "eslint": "^8.57.0",
     "eslint-config-airbnb": "^19.0.4",
     "eslint-config-airbnb-typescript": "^18.0.0",
-    "eslint-config-prettier": "^10.1.1",
+    "eslint-config-prettier": "^10.1.2",
     "eslint-plugin-import": "^2.29.1",
     "eslint-plugin-jsx-a11y": "^6.10.2",
     "eslint-plugin-prettier": "^5.2.6",
@@ -109,8 +111,8 @@
     "eslint-plugin-react-hooks": "^4.6.2",
     "eslint-plugin-unused-imports": "^4.1.4",
     "husky": "^9.1.7",
-    "jsdom": "^26.0.0",
-    "lint-staged": "^15.5.0",
+    "jsdom": "^26.1.0",
+    "lint-staged": "^15.5.1",
     "msw": "^2.6.6",
     "postcss": "^8.5.2",
     "prettier": "^3.5.3",

frontend/public/mockServiceWorker.js

@@ -8,7 +8,7 @@
  * - Please do NOT serve this file on production.
  */
 
-const PACKAGE_VERSION = '2.7.3'
+const PACKAGE_VERSION = '2.7.5'
 const INTEGRITY_CHECKSUM = '00729d72e3b82faf54ca8b9621dbb96f'
 const IS_MOCKED_RESPONSE = Symbol('isMockedResponse')
 const activeClientIds = new Set()

frontend/scripts/check-unlocalized-strings.cjs

@@ -105,6 +105,7 @@ function isRawTranslationKey(str) {
 // Specific technical strings that should be excluded from localization
 const EXCLUDED_TECHNICAL_STRINGS = [
   "openid email profile", // OAuth scope string - not user-facing
+  "OPEN_ISSUE", // Task type identifier, not a UI string
 ];
 
 function isExcludedTechnicalString(str) {
@@ -275,8 +276,8 @@ function isCommonDevelopmentString(str) {
 
   // HTML tags and attributes
   if (
-    /^<[a-z0-9]+>.*<\/[a-z0-9]+>$/.test(str) ||
-    /^<[a-z0-9]+ [^>]+\/>$/.test(str)
+    /^<[a-z0-9]+(?:\s[^>]*)?>.*<\/[a-z0-9]+>$/i.test(str) ||
+    /^<[a-z0-9]+ [^>]+\/>$/i.test(str)
   ) {
     return true;
   }
@@ -344,6 +345,7 @@ function isCommonDevelopmentString(str) {
 
   // Check for specific error messages and UI text
   if (
+    str === "Failed to accept TOS:" ||
     str === "All data associated with this project will be lost." ||
     str === "You will lose any unsaved information." ||
     str ===

frontend/src/api/api-keys.ts

@@ -0,0 +1,49 @@
+import { openHands } from "./open-hands-axios";
+
+export interface ApiKey {
+  id: string;
+  name: string;
+  prefix: string;
+  created_at: string;
+  last_used_at: string | null;
+}
+
+export interface CreateApiKeyResponse {
+  id: string;
+  name: string;
+  key: string; // Full key, only returned once upon creation
+  prefix: string;
+  created_at: string;
+}
+
+class ApiKeysClient {
+  /**
+   * Get all API keys for the current user
+   */
+  static async getApiKeys(): Promise<ApiKey[]> {
+    const { data } = await openHands.get<unknown>("/api/keys");
+    // Ensure we always return an array, even if the API returns something else
+    return Array.isArray(data) ? (data as ApiKey[]) : [];
+  }
+
+  /**
+   * Create a new API key
+   * @param name - A descriptive name for the API key
+   */
+  static async createApiKey(name: string): Promise<CreateApiKeyResponse> {
+    const { data } = await openHands.post<CreateApiKeyResponse>("/api/keys", {
+      name,
+    });
+    return data;
+  }
+
+  /**
+   * Delete an API key
+   * @param id - The ID of the API key to delete
+   */
+  static async deleteApiKey(id: string): Promise<void> {
+    await openHands.delete(`/api/keys/${id}`);
+  }
+}
+
+export default ApiKeysClient;

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

@@ -1,5 +1,31 @@
 import axios from "axios";
+import { saveLastPage } from "#/utils/last-page";
 
 export const openHands = axios.create({
   baseURL: `${window.location.protocol}//${import.meta.env.VITE_BACKEND_BASE_URL || window?.location.host}`,
 });
+
+// We don't need the request interceptor anymore since we're using a separate QueryClient
+// for the TOS page that disables all queries
+
+// Add response interceptor to handle 401 and 403 errors
+openHands.interceptors.response.use(
+  (response) => response,
+  (error) => {
+    if (error.response?.status === 401) {
+      // Save the last page before redirecting
+      saveLastPage();
+    } else if (
+      error.response?.status === 403 &&
+      error.response?.data?.tos_not_accepted
+    ) {
+      // Save the last page before redirecting to TOS
+      // Only redirect if we're not already on the TOS page
+      if (window.location.pathname !== "/tos") {
+        saveLastPage();
+        window.location.href = "/tos";
+      }
+    }
+    return Promise.reject(error);
+  },
+);

frontend/src/api/open-hands.ts

@@ -8,6 +8,8 @@ import {
   Conversation,
   ResultSet,
   GetTrajectoryResponse,
+  GitChangeDiff,
+  GitChange,
 } from "./open-hands.types";
 import { openHands } from "./open-hands-axios";
 import { ApiSettings, PostApiSettings } from "#/types/settings";
@@ -277,6 +279,35 @@ class OpenHands {
       appMode === "saas" ? "/api/logout" : "/api/unset-settings-tokens";
     await openHands.post(endpoint);
   }
+
+  /**
+   * Accept the Terms of Service
+   * @returns true if successful, false otherwise
+   */
+  static async acceptTOS(): Promise<boolean> {
+    const response = await openHands.post("/api/accept_tos");
+    return response.status === 200;
+  }
+  
+  static async getGitChanges(conversationId: string): Promise<GitChange[]> {
+    const { data } = await openHands.get<GitChange[]>(
+      `/api/conversations/${conversationId}/git/changes`,
+    );
+    return data;
+  }
+
+  static async getGitChangeDiff(
+    conversationId: string,
+    path: string,
+  ): Promise<GitChangeDiff> {
+    const { data } = await openHands.get<GitChangeDiff>(
+      `/api/conversations/${conversationId}/git/diff`,
+      {
+        params: { path },
+      },
+    );
+    return data;
+  }
 }
 
 export default OpenHands;

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

@@ -70,6 +70,8 @@ export interface AuthenticateResponse {
   error?: string;
 }
 
+export type ConversationTrigger = "resolver" | "gui";
+
 export interface Conversation {
   conversation_id: string;
   title: string;
@@ -77,9 +79,22 @@ export interface Conversation {
   last_updated_at: string;
   created_at: string;
   status: ProjectStatus;
+  trigger?: ConversationTrigger;
 }
 
 export interface ResultSet<T> {
   results: T[];
   next_page_id: string | null;
 }
+
+export type GitChangeStatus = "M" | "A" | "D" | "R" | "U";
+
+export interface GitChange {
+  status: GitChangeStatus;
+  path: string;
+}
+
+export interface GitChangeDiff {
+  modified: string;
+  original: string;
+}

frontend/src/api/suggestions-service/suggestions-service.api.ts

@@ -0,0 +1,9 @@
+import { SuggestedTask } from "#/components/features/home/tasks/task.types";
+import { openHands } from "../open-hands-axios";
+
+export class SuggestionsService {
+  static async getSuggestedTasks(): Promise<SuggestedTask[]> {
+    const { data } = await openHands.get("/api/user/suggested-tasks");
+    return data;
+  }
+}

frontend/src/assets/branding/all-hands-logo-spark.svg

@@ -15,13 +15,13 @@
         fill="black" />
       <path
         d="M38.7381 10.5084C38.5759 10.5084 38.4106 10.4788 38.2545 10.4076C37.6821 10.1526 37.4312 9.49736 37.6944 8.94289C38.5453 7.1431 39.791 5.48266 41.2938 4.14245C41.7559 3.73031 42.4782 3.75699 42.9037 4.20768C43.3291 4.65541 43.3016 5.35516 42.8363 5.76731C41.5539 6.91182 40.4919 8.32912 39.7634 9.86502C39.5737 10.2653 39.1666 10.5055 38.7381 10.5084Z"
-        fill="white" />
+        fill="black" />
       <path
         d="M34.898 9.87074C34.3073 9.87667 33.8023 9.43784 33.7533 8.85669C33.536 6.25633 33.5268 3.62039 33.7319 1.02003C33.7808 0.412188 34.3287 -0.0414663 34.9531 0.00300963C35.5805 0.0504507 36.0488 0.578232 36.0029 1.18607C35.807 3.67079 35.8162 6.1911 36.0243 8.67582C36.0763 9.28366 35.6081 9.81737 34.9806 9.86481C34.9531 9.86481 34.9255 9.86778 34.898 9.86778V9.87074Z"
-        fill="white" />
+        fill="black" />
       <path
         d="M30.976 10.5558C30.4649 10.5618 29.9935 10.2267 29.8619 9.7256C29.3783 7.88726 28.4632 6.14084 27.2175 4.67906C26.8165 4.20762 26.8869 3.51379 27.3705 3.12537C27.8572 2.73695 28.5734 2.80514 28.9743 3.27362C30.4312 4.98743 31.5024 7.03036 32.0656 9.18003C32.2217 9.77008 31.8514 10.372 31.2423 10.5232C31.1505 10.5469 31.0617 10.5558 30.9699 10.5588L30.976 10.5558Z"
-        fill="white" />
+        fill="black" />
     </g>
   </g>
   <defs>

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

@@ -0,0 +1,6 @@
+<svg width="22" height="22" viewBox="0 0 22 22" fill="none" xmlns="http://www.w3.org/2000/svg">
+  <path d="M11 21L16.5 8H5.5L11 21Z" stroke="currentColor" stroke-linecap="round" stroke-linejoin="round"/>
+  <path d="M1 8L3.5 15.5L11 21L18.5 15.5L21 8" stroke="currentColor" stroke-linecap="round" stroke-linejoin="round"/>
+  <path d="M1 8L5.5 8L8.25 1" stroke="currentColor" stroke-linecap="round" stroke-linejoin="round"/>
+  <path d="M21 8L16.5 8L13.75 1" stroke="currentColor" stroke-linecap="round" stroke-linejoin="round"/>
+</svg>