fix: Use runtime to load microagents

- Update CodeActAgent to use runtime.get_microagents_from_selected_repo
- Add test to verify microagent loading from runtime
- Fix #6304

.github/workflows/ghcr-build.yml

@@ -56,7 +56,7 @@ jobs:
           docker-images: false
           swap-storage: true
       - name: Set up QEMU
-        uses: docker/setup-qemu-action@v3.2.0
+        uses: docker/setup-qemu-action@v3.3.0
         with:
           image: tonistiigi/binfmt:latest
       - name: Login to GHCR
@@ -119,7 +119,7 @@ jobs:
           docker-images: false
           swap-storage: true
       - name: Set up QEMU
-        uses: docker/setup-qemu-action@v3.2.0
+        uses: docker/setup-qemu-action@v3.3.0
         with:
           image: tonistiigi/binfmt:latest
       - name: Login to GHCR

.github/workflows/integration-runner.yml

@@ -56,6 +56,7 @@ jobs:
           LLM_MODEL: "litellm_proxy/claude-3-5-haiku-20241022"
           LLM_API_KEY: ${{ secrets.LLM_API_KEY }}
           LLM_BASE_URL: ${{ secrets.LLM_BASE_URL }}
+          MAX_ITERATIONS: 10
         run: |
           echo "[llm.eval]" > config.toml
           echo "model = \"$LLM_MODEL\"" >> config.toml
@@ -70,7 +71,7 @@ jobs:
         env:
           SANDBOX_FORCE_REBUILD_RUNTIME: True
         run: |
-          poetry run ./evaluation/integration_tests/scripts/run_infer.sh llm.eval HEAD CodeActAgent '' $N_PROCESSES '' 'haiku_run'
+          poetry run ./evaluation/integration_tests/scripts/run_infer.sh llm.eval HEAD CodeActAgent '' 10 $N_PROCESSES '' 'haiku_run'
 
           # get integration tests report
           REPORT_FILE_HAIKU=$(find evaluation/evaluation_outputs/outputs/integration_tests/CodeActAgent/*haiku*_maxiter_10_N* -name "report.md" -type f | head -n 1)
@@ -88,6 +89,7 @@ jobs:
           LLM_MODEL: "litellm_proxy/deepseek-chat"
           LLM_API_KEY: ${{ secrets.LLM_API_KEY }}
           LLM_BASE_URL: ${{ secrets.LLM_BASE_URL }}
+          MAX_ITERATIONS: 10
         run: |
           echo "[llm.eval]" > config.toml
           echo "model = \"$LLM_MODEL\"" >> config.toml
@@ -99,7 +101,7 @@ jobs:
         env:
           SANDBOX_FORCE_REBUILD_RUNTIME: True
         run: |
-          poetry run ./evaluation/integration_tests/scripts/run_infer.sh llm.eval HEAD CodeActAgent '' $N_PROCESSES '' 'deepseek_run'
+          poetry run ./evaluation/integration_tests/scripts/run_infer.sh llm.eval HEAD CodeActAgent '' 10 $N_PROCESSES '' 'deepseek_run'
 
           # get integration tests report
           REPORT_FILE_DEEPSEEK=$(find evaluation/evaluation_outputs/outputs/integration_tests/CodeActAgent/deepseek*_maxiter_10_N* -name "report.md" -type f | head -n 1)
@@ -109,11 +111,75 @@ jobs:
           echo >> $GITHUB_ENV
           echo "EOF" >> $GITHUB_ENV
 
+      # -------------------------------------------------------------
+      # Run DelegatorAgent tests for Haiku, limited to t01 and t02
+      - name: Wait a little bit (again)
+        run: sleep 5
+
+      - name: Configure config.toml for testing DelegatorAgent (Haiku)
+        env:
+          LLM_MODEL: "litellm_proxy/claude-3-5-haiku-20241022"
+          LLM_API_KEY: ${{ secrets.LLM_API_KEY }}
+          LLM_BASE_URL: ${{ secrets.LLM_BASE_URL }}
+          MAX_ITERATIONS: 30
+        run: |
+          echo "[llm.eval]" > config.toml
+          echo "model = \"$LLM_MODEL\"" >> config.toml
+          echo "api_key = \"$LLM_API_KEY\"" >> config.toml
+          echo "base_url = \"$LLM_BASE_URL\"" >> config.toml
+          echo "temperature = 0.0" >> config.toml
+
+      - name: Run integration test evaluation for DelegatorAgent (Haiku)
+        env:
+          SANDBOX_FORCE_REBUILD_RUNTIME: True
+        run: |
+          poetry run ./evaluation/integration_tests/scripts/run_infer.sh llm.eval HEAD DelegatorAgent '' 30 $N_PROCESSES "t01_fix_simple_typo,t02_add_bash_hello" 'delegator_haiku_run'
+
+          # Find and export the delegator test results
+          REPORT_FILE_DELEGATOR_HAIKU=$(find evaluation/evaluation_outputs/outputs/integration_tests/DelegatorAgent/*haiku*_maxiter_30_N* -name "report.md" -type f | head -n 1)
+          echo "REPORT_FILE_DELEGATOR_HAIKU: $REPORT_FILE_DELEGATOR_HAIKU"
+          echo "INTEGRATION_TEST_REPORT_DELEGATOR_HAIKU<<EOF" >> $GITHUB_ENV
+          cat $REPORT_FILE_DELEGATOR_HAIKU >> $GITHUB_ENV
+          echo >> $GITHUB_ENV
+          echo "EOF" >> $GITHUB_ENV
+
+      # -------------------------------------------------------------
+      # Run DelegatorAgent tests for DeepSeek, limited to t01 and t02
+      - name: Wait a little bit (again)
+        run: sleep 5
+
+      - name: Configure config.toml for testing DelegatorAgent (DeepSeek)
+        env:
+          LLM_MODEL: "litellm_proxy/deepseek-chat"
+          LLM_API_KEY: ${{ secrets.LLM_API_KEY }}
+          LLM_BASE_URL: ${{ secrets.LLM_BASE_URL }}
+          MAX_ITERATIONS: 30
+        run: |
+          echo "[llm.eval]" > config.toml
+          echo "model = \"$LLM_MODEL\"" >> config.toml
+          echo "api_key = \"$LLM_API_KEY\"" >> config.toml
+          echo "base_url = \"$LLM_BASE_URL\"" >> config.toml
+          echo "temperature = 0.0" >> config.toml
+
+      - name: Run integration test evaluation for DelegatorAgent (DeepSeek)
+        env:
+          SANDBOX_FORCE_REBUILD_RUNTIME: True
+        run: |
+          poetry run ./evaluation/integration_tests/scripts/run_infer.sh llm.eval HEAD DelegatorAgent '' 30 $N_PROCESSES "t01_fix_simple_typo,t02_add_bash_hello" 'delegator_deepseek_run'
+
+          # Find and export the delegator test results
+          REPORT_FILE_DELEGATOR_DEEPSEEK=$(find evaluation/evaluation_outputs/outputs/integration_tests/DelegatorAgent/deepseek*_maxiter_30_N* -name "report.md" -type f | head -n 1)
+          echo "REPORT_FILE_DELEGATOR_DEEPSEEK: $REPORT_FILE_DELEGATOR_DEEPSEEK"
+          echo "INTEGRATION_TEST_REPORT_DELEGATOR_DEEPSEEK<<EOF" >> $GITHUB_ENV
+          cat $REPORT_FILE_DELEGATOR_DEEPSEEK >> $GITHUB_ENV
+          echo >> $GITHUB_ENV
+          echo "EOF" >> $GITHUB_ENV
+
       - name: Create archive of evaluation outputs
         run: |
           TIMESTAMP=$(date +'%y-%m-%d-%H-%M')
           cd evaluation/evaluation_outputs/outputs  # Change to the outputs directory
-          tar -czvf ../../../integration_tests_${TIMESTAMP}.tar.gz integration_tests/CodeActAgent/*  # Only include the actual result directories
+          tar -czvf ../../../integration_tests_${TIMESTAMP}.tar.gz integration_tests/CodeActAgent/* integration_tests/DelegatorAgent/*  # Only include the actual result directories
 
       - name: Upload evaluation results as artifact
         uses: actions/upload-artifact@v4
@@ -154,5 +220,11 @@ jobs:
               **Integration Tests Report (DeepSeek)**
               DeepSeek LLM Test Results:
               ${{ env.INTEGRATION_TEST_REPORT_DEEPSEEK }}
+              ---
+                **Integration Tests Report Delegator (Haiku)**
+              ${{ env.INTEGRATION_TEST_REPORT_DELEGATOR_HAIKU }}
+              ---
+              **Integration Tests Report Delegator (DeepSeek)**
+              ${{ env.INTEGRATION_TEST_REPORT_DELEGATOR_DEEPSEEK }}
               ---
               Download testing outputs (includes both Haiku and DeepSeek results): [Download](${{ steps.upload_results_artifact.outputs.artifact-url }})

config.template.toml

@@ -34,7 +34,7 @@ workspace_base = "./workspace"
 
 # Path to store trajectories, can be a folder or a file
 # If it's a folder, the session id will be used as the file name
-#trajectories_path="./trajectories"
+#save_trajectory_path="./trajectories"
 
 # File store path
 #file_store_path = "/tmp/file_store"

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

@@ -1,5 +1,3 @@
-
-
 # Options de configuration
 
 Ce guide détaille toutes les options de configuration disponibles pour OpenHands, vous aidant à personnaliser son comportement et à l'intégrer avec d'autres services.
@@ -94,7 +92,7 @@ Les options de configuration de base sont définies dans la section `[core]` du
   - Description : Désactiver la couleur dans la sortie du terminal
 
 **Trajectoires**
-- `trajectories_path`
+- `save_trajectory_path`
   - Type : `str`
   - Valeur par défaut : `"./trajectories"`
   - Description : Chemin pour stocker les trajectoires (peut être un dossier ou un fichier). Si c'est un dossier, les trajectoires seront enregistrées dans un fichier nommé avec l'ID de session et l'extension .json, dans ce dossier.
@@ -184,6 +182,10 @@ Les options de configuration LLM (Large Language Model) sont définies dans la s
 
 Pour les utiliser avec la commande docker, passez `-e LLM_<option>`. Exemple : `-e LLM_NUM_RETRIES`.
 
+:::note
+Pour les configurations de développement, vous pouvez également définir des configurations LLM personnalisées. Voir [Configurations LLM personnalisées](./llms/custom-llm-configs) pour plus de détails.
+:::
+
 **Informations d'identification AWS**
 - `aws_access_key_id`
   - Type : `str`
@@ -368,4 +370,26 @@ Les options de configuration de l'agent sont définies dans les sections `[agent
 - `codeact_enable_llm_editor`
   - Type : `bool`
   - Valeur par défaut : `false`
-  - Description : Si l'éditeur LLM est activé dans l'espace d'action (foncti
+  - Description : Si l'éditeur LLM est activé dans l'espace d'action (fonctionne uniquement avec l'appel de fonction)
+
+**Utilisation du micro-agent**
+- `use_microagents`
+  - Type : `bool`
+  - Valeur par défaut : `true`
+  - Description : Indique si l'utilisation des micro-agents est activée ou non
+
+- `disabled_microagents`
+  - Type : `list of str`
+  - Valeur par défaut : `None`
+  - Description : Liste des micro-agents à désactiver
+
+### Exécution
+- `timeout`
+  - Type : `int`
+  - Valeur par défaut : `120`
+  - Description : Délai d'expiration du bac à sable, en secondes
+
+- `user_id`
+  - Type : `int`
+  - Valeur par défaut : `1000`
+  - Description : ID de l'utilisateur du bac à sable

docs/i18n/fr/docusaurus-plugin-content-docs/current/usage/llms/custom-llm-configs.md

@@ -0,0 +1,106 @@
+# Configurations LLM personnalisées
+
+OpenHands permet de définir plusieurs configurations LLM nommées dans votre fichier `config.toml`. Cette fonctionnalité vous permet d'utiliser différentes configurations LLM pour différents usages, comme utiliser un modèle moins coûteux pour les tâches qui ne nécessitent pas de réponses de haute qualité, ou utiliser différents modèles avec différents paramètres pour des agents spécifiques.
+
+## Comment ça fonctionne
+
+Les configurations LLM nommées sont définies dans le fichier `config.toml` en utilisant des sections qui commencent par `llm.`. Par exemple :
+
+```toml
+# Configuration LLM par défaut
+[llm]
+model = "gpt-4"
+api_key = "votre-clé-api"
+temperature = 0.0
+
+# Configuration LLM personnalisée pour un modèle moins coûteux
+[llm.gpt3]
+model = "gpt-3.5-turbo"
+api_key = "votre-clé-api"
+temperature = 0.2
+
+# Une autre configuration personnalisée avec des paramètres différents
+[llm.haute-creativite]
+model = "gpt-4"
+api_key = "votre-clé-api"
+temperature = 0.8
+top_p = 0.9
+```
+
+Chaque configuration nommée hérite de tous les paramètres de la section `[llm]` par défaut et peut remplacer n'importe lequel de ces paramètres. Vous pouvez définir autant de configurations personnalisées que nécessaire.
+
+## Utilisation des configurations personnalisées
+
+### Avec les agents
+
+Vous pouvez spécifier quelle configuration LLM un agent doit utiliser en définissant le paramètre `llm_config` dans la section de configuration de l'agent :
+
+```toml
+[agent.RepoExplorerAgent]
+# Utiliser la configuration GPT-3 moins coûteuse pour cet agent
+llm_config = 'gpt3'
+
+[agent.CodeWriterAgent]
+# Utiliser la configuration haute créativité pour cet agent
+llm_config = 'haute-creativite'
+```
+
+### Options de configuration
+
+Chaque configuration LLM nommée prend en charge toutes les mêmes options que la configuration LLM par défaut. Celles-ci incluent :
+
+- Sélection du modèle (`model`)
+- Configuration de l'API (`api_key`, `base_url`, etc.)
+- Paramètres du modèle (`temperature`, `top_p`, etc.)
+- Paramètres de nouvelle tentative (`num_retries`, `retry_multiplier`, etc.)
+- Limites de jetons (`max_input_tokens`, `max_output_tokens`)
+- Et toutes les autres options de configuration LLM
+
+Pour une liste complète des options disponibles, consultez la section Configuration LLM dans la documentation des [Options de configuration](../configuration-options).
+
+## Cas d'utilisation
+
+Les configurations LLM personnalisées sont particulièrement utiles dans plusieurs scénarios :
+
+- **Optimisation des coûts** : Utiliser des modèles moins coûteux pour les tâches qui ne nécessitent pas de réponses de haute qualité, comme l'exploration de dépôt ou les opérations simples sur les fichiers.
+- **Réglage spécifique aux tâches** : Configurer différentes valeurs de température et de top_p pour les tâches qui nécessitent différents niveaux de créativité ou de déterminisme.
+- **Différents fournisseurs** : Utiliser différents fournisseurs LLM ou points d'accès API pour différentes tâches.
+- **Tests et développement** : Basculer facilement entre différentes configurations de modèles pendant le développement et les tests.
+
+## Exemple : Optimisation des coûts
+
+Un exemple pratique d'utilisation des configurations LLM personnalisées pour optimiser les coûts :
+
+```toml
+# Configuration par défaut utilisant GPT-4 pour des réponses de haute qualité
+[llm]
+model = "gpt-4"
+api_key = "votre-clé-api"
+temperature = 0.0
+
+# Configuration moins coûteuse pour l'exploration de dépôt
+[llm.repo-explorer]
+model = "gpt-3.5-turbo"
+temperature = 0.2
+
+# Configuration pour la génération de code
+[llm.code-gen]
+model = "gpt-4"
+temperature = 0.0
+max_output_tokens = 2000
+
+[agent.RepoExplorerAgent]
+llm_config = 'repo-explorer'
+
+[agent.CodeWriterAgent]
+llm_config = 'code-gen'
+```
+
+Dans cet exemple :
+- L'exploration de dépôt utilise un modèle moins coûteux car il s'agit principalement de comprendre et de naviguer dans le code
+- La génération de code utilise GPT-4 avec une limite de jetons plus élevée pour générer des blocs de code plus importants
+- La configuration par défaut reste disponible pour les autres tâches
+
+:::note
+Les configurations LLM personnalisées ne sont disponibles que lors de l'utilisation d'OpenHands en mode développement, via `main.py` ou `cli.py`. Lors de l'exécution via `docker run`, veuillez utiliser les options de configuration standard.
+:::

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

@@ -91,7 +91,7 @@
   - 描述: 禁用终端输出中的颜色
 
 **轨迹**
-- `trajectories_path`
+- `save_trajectory_path`
   - 类型: `str`
   - 默认值: `"./trajectories"`
   - 描述: 存储轨迹的路径(可以是文件夹或文件)。如果是文件夹,轨迹将保存在该文件夹中以会话 ID 命名的 .json 文件中。

docs/modules/usage/configuration-options.md

@@ -50,7 +50,7 @@ The core configuration options are defined in the `[core]` section of the `confi
   - Description: Disable color in terminal output
 
 ### Trajectories
-- `trajectories_path`
+- `save_trajectory_path`
   - Type: `str`
   - Default: `"./trajectories"`
   - Description: Path to store trajectories (can be a folder or a file). If it's a folder, the trajectories will be saved in a file named with the session id name and .json extension, in that folder.
@@ -140,7 +140,11 @@ The LLM (Large Language Model) configuration options are defined in the `[llm]`
 
 To use these with the docker command, pass in `-e LLM_<option>`. Example: `-e LLM_NUM_RETRIES`.
 
-### AWS Credentials
+:::note
+For development setups, you can also define custom named LLM configurations. See [Custom LLM Configurations](./llms/custom-llm-configs) for details.
+:::
+
+**AWS Credentials**
 - `aws_access_key_id`
   - Type: `str`
   - Default: `""`

docs/modules/usage/llms/custom-llm-configs.md

@@ -0,0 +1,106 @@
+# Custom LLM Configurations
+
+OpenHands supports defining multiple named LLM configurations in your `config.toml` file. This feature allows you to use different LLM configurations for different purposes, such as using a cheaper model for tasks that don't require high-quality responses, or using different models with different parameters for specific agents.
+
+## How It Works
+
+Named LLM configurations are defined in the `config.toml` file using sections that start with `llm.`. For example:
+
+```toml
+# Default LLM configuration
+[llm]
+model = "gpt-4"
+api_key = "your-api-key"
+temperature = 0.0
+
+# Custom LLM configuration for a cheaper model
+[llm.gpt3]
+model = "gpt-3.5-turbo"
+api_key = "your-api-key"
+temperature = 0.2
+
+# Another custom configuration with different parameters
+[llm.high-creativity]
+model = "gpt-4"
+api_key = "your-api-key"
+temperature = 0.8
+top_p = 0.9
+```
+
+Each named configuration inherits all settings from the default `[llm]` section and can override any of those settings. You can define as many custom configurations as needed.
+
+## Using Custom Configurations
+
+### With Agents
+
+You can specify which LLM configuration an agent should use by setting the `llm_config` parameter in the agent's configuration section:
+
+```toml
+[agent.RepoExplorerAgent]
+# Use the cheaper GPT-3 configuration for this agent
+llm_config = 'gpt3'
+
+[agent.CodeWriterAgent]
+# Use the high creativity configuration for this agent
+llm_config = 'high-creativity'
+```
+
+### Configuration Options
+
+Each named LLM configuration supports all the same options as the default LLM configuration. These include:
+
+- Model selection (`model`)
+- API configuration (`api_key`, `base_url`, etc.)
+- Model parameters (`temperature`, `top_p`, etc.)
+- Retry settings (`num_retries`, `retry_multiplier`, etc.)
+- Token limits (`max_input_tokens`, `max_output_tokens`)
+- And all other LLM configuration options
+
+For a complete list of available options, see the LLM Configuration section in the [Configuration Options](../configuration-options) documentation.
+
+## Use Cases
+
+Custom LLM configurations are particularly useful in several scenarios:
+
+- **Cost Optimization**: Use cheaper models for tasks that don't require high-quality responses, like repository exploration or simple file operations.
+- **Task-Specific Tuning**: Configure different temperature and top_p values for tasks that require different levels of creativity or determinism.
+- **Different Providers**: Use different LLM providers or API endpoints for different tasks.
+- **Testing and Development**: Easily switch between different model configurations during development and testing.
+
+## Example: Cost Optimization
+
+A practical example of using custom LLM configurations to optimize costs:
+
+```toml
+# Default configuration using GPT-4 for high-quality responses
+[llm]
+model = "gpt-4"
+api_key = "your-api-key"
+temperature = 0.0
+
+# Cheaper configuration for repository exploration
+[llm.repo-explorer]
+model = "gpt-3.5-turbo"
+temperature = 0.2
+
+# Configuration for code generation
+[llm.code-gen]
+model = "gpt-4"
+temperature = 0.0
+max_output_tokens = 2000
+
+[agent.RepoExplorerAgent]
+llm_config = 'repo-explorer'
+
+[agent.CodeWriterAgent]
+llm_config = 'code-gen'
+```
+
+In this example:
+- Repository exploration uses a cheaper model since it mainly involves understanding and navigating code
+- Code generation uses GPT-4 with a higher token limit for generating larger code blocks
+- The default configuration remains available for other tasks
+
+:::note
+Custom LLM configurations are only available when using OpenHands in development mode, via `main.py` or `cli.py`. When running via `docker run`, please use the standard configuration options.
+:::

evaluation/benchmarks/the_agent_company/run_infer.py

@@ -39,7 +39,7 @@ def get_config(
         run_as_openhands=False,
         max_budget_per_task=4,
         max_iterations=100,
-        trajectories_path=os.path.join(
+        save_trajectory_path=os.path.join(
             mount_path_on_host, f'traj_{task_short_name}.json'
         ),
         sandbox=SandboxConfig(

evaluation/integration_tests/run_infer.py

@@ -8,13 +8,15 @@ from evaluation.integration_tests.tests.base import BaseIntegrationTest, TestRes
 from evaluation.utils.shared import (
     EvalMetadata,
     EvalOutput,
-    codeact_user_response,
     make_metadata,
     prepare_dataset,
     reset_logger_for_multiprocessing,
     run_evaluation,
     update_llm_config_for_completions_logging,
 )
+from evaluation.utils.shared import (
+    codeact_user_response as fake_user_response,
+)
 from openhands.controller.state.state import State
 from openhands.core.config import (
     AgentConfig,
@@ -31,7 +33,8 @@ from openhands.runtime.base import Runtime
 from openhands.utils.async_utils import call_async_from_sync
 
 FAKE_RESPONSES = {
-    'CodeActAgent': codeact_user_response,
+    'CodeActAgent': fake_user_response,
+    'DelegatorAgent': fake_user_response,
 }
 
 
@@ -219,7 +222,7 @@ if __name__ == '__main__':
 
     df = pd.read_json(output_file, lines=True, orient='records')
 
-    # record success and reason for failure for the final report
+    # record success and reason
     df['success'] = df['test_result'].apply(lambda x: x['success'])
     df['reason'] = df['test_result'].apply(lambda x: x['reason'])
     logger.info('-' * 100)
@@ -234,15 +237,27 @@ if __name__ == '__main__':
     logger.info('-' * 100)
 
     # record cost for each instance, with 3 decimal places
-    df['cost'] = df['metrics'].apply(lambda x: round(x['accumulated_cost'], 3))
+    # we sum up all the "costs" from the metrics array
+    df['cost'] = df['metrics'].apply(
+        lambda m: round(sum(c['cost'] for c in m['costs']), 3)
+        if m and 'costs' in m
+        else 0.0
+    )
+
+    # capture the top-level error if present, per instance
+    df['error_message'] = df.get('error', None)
+
     logger.info(f'Total cost: USD {df["cost"].sum():.2f}')
 
     report_file = os.path.join(metadata.eval_output_dir, 'report.md')
     with open(report_file, 'w') as f:
         f.write(
-            f'Success rate: {df["success"].mean():.2%} ({df["success"].sum()}/{len(df)})\n'
+            f'Success rate: {df["success"].mean():.2%}'
+            f' ({df["success"].sum()}/{len(df)})\n'
         )
         f.write(f'\nTotal cost: USD {df["cost"].sum():.2f}\n')
         f.write(
-            df[['instance_id', 'success', 'reason', 'cost']].to_markdown(index=False)
+            df[
+                ['instance_id', 'success', 'reason', 'cost', 'error_message']
+            ].to_markdown(index=False)
         )

evaluation/integration_tests/scripts/run_infer.sh

@@ -7,8 +7,9 @@ MODEL_CONFIG=$1
 COMMIT_HASH=$2
 AGENT=$3
 EVAL_LIMIT=$4
-NUM_WORKERS=$5
-EVAL_IDS=$6
+MAX_ITERATIONS=$5
+NUM_WORKERS=$6
+EVAL_IDS=$7
 
 if [ -z "$NUM_WORKERS" ]; then
   NUM_WORKERS=1
@@ -43,7 +44,7 @@ fi
 COMMAND="poetry run python evaluation/integration_tests/run_infer.py \
   --agent-cls $AGENT \
   --llm-config $MODEL_CONFIG \
-  --max-iterations 10 \
+  --max-iterations ${MAX_ITERATIONS:-10} \
   --eval-num-workers $NUM_WORKERS \
   --eval-note $EVAL_NOTE"
 

frontend/__tests__/api/github.test.ts

@@ -1,47 +0,0 @@
-import { describe, expect, it, vi } from "vitest";
-import { retrieveLatestGitHubCommit } from "../../src/api/github";
-
-describe("retrieveLatestGitHubCommit", () => {
-  const { githubGetMock } = vi.hoisted(() => ({
-    githubGetMock: vi.fn(),
-  }));
-
-  vi.mock("../../src/api/github-axios-instance", () => ({
-    github: {
-      get: githubGetMock,
-    },
-  }));
-
-  it("should return the latest commit when repository has commits", async () => {
-    const mockCommit = {
-      sha: "123abc",
-      commit: {
-        message: "Initial commit",
-      },
-    };
-
-    githubGetMock.mockResolvedValueOnce({
-      data: [mockCommit],
-    });
-
-    const result = await retrieveLatestGitHubCommit("user/repo");
-    expect(result).toEqual(mockCommit);
-  });
-
-  it("should return null when repository is empty", async () => {
-    const error = new Error("Repository is empty");
-    (error as any).response = { status: 409 };
-    githubGetMock.mockRejectedValueOnce(error);
-
-    const result = await retrieveLatestGitHubCommit("user/empty-repo");
-    expect(result).toBeNull();
-  });
-
-  it("should throw error for other error cases", async () => {
-    const error = new Error("Network error");
-    (error as any).response = { status: 500 };
-    githubGetMock.mockRejectedValueOnce(error);
-
-    await expect(retrieveLatestGitHubCommit("user/repo")).rejects.toThrow();
-  });
-});

frontend/__tests__/components/browser.test.tsx

@@ -1,11 +1,10 @@
 import { describe, it, expect, afterEach, vi } from "vitest";
-import * as router from "react-router";
 
 // Mock useParams before importing components
 vi.mock("react-router", async () => {
   const actual = await vi.importActual("react-router");
   return {
-    ...actual as object,
+    ...(actual as object),
     useParams: () => ({ conversationId: "test-conversation-id" }),
   };
 });
@@ -14,7 +13,7 @@ vi.mock("react-router", async () => {
 vi.mock("react-i18next", async () => {
   const actual = await vi.importActual("react-i18next");
   return {
-    ...actual as object,
+    ...(actual as object),
     useTranslation: () => ({
       t: (key: string) => key,
       i18n: {
@@ -28,7 +27,6 @@ import { screen } from "@testing-library/react";
 import { renderWithProviders } from "../../test-utils";
 import { BrowserPanel } from "#/components/features/browser/browser";
 
-
 describe("Browser", () => {
   afterEach(() => {
     vi.clearAllMocks();
@@ -45,7 +43,7 @@ describe("Browser", () => {
     });
 
     // i18n empty message key
-    expect(screen.getByText("BROWSER$EMPTY_MESSAGE")).toBeInTheDocument();
+    expect(screen.getByText("BROWSER$NO_PAGE_LOADED")).toBeInTheDocument();
   });
 
   it("renders the url and a screenshot", () => {

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

@@ -84,12 +84,10 @@ describe("ChatInput", () => {
     expect(onSubmitMock).not.toHaveBeenCalled();
   });
 
-  it("should render a placeholder", () => {
-    render(
-      <ChatInput placeholder="Enter your message" onSubmit={onSubmitMock} />,
-    );
+  it("should render a placeholder with translation key", () => {
+    render(<ChatInput onSubmit={onSubmitMock} />);
 
-    const textarea = screen.getByPlaceholderText("Enter your message");
+    const textarea = screen.getByPlaceholderText("SUGGESTIONS$WHAT_TO_BUILD");
     expect(textarea).toBeInTheDocument();
   });
 

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

@@ -2,36 +2,42 @@ import { describe, expect, it } from "vitest";
 import { screen } from "@testing-library/react";
 import { renderWithProviders } from "test-utils";
 import { ExpandableMessage } from "#/components/features/chat/expandable-message";
-import { vi } from 'vitest';
+import { vi } from "vitest";
 
-vi.mock('react-i18next', async () => {
-  const actual = await vi.importActual('react-i18next');
+vi.mock("react-i18next", async () => {
+  const actual = await vi.importActual("react-i18next");
   return {
     ...actual,
     useTranslation: () => ({
-      t: (key:string) => key,
+      t: (key: string) => key,
       i18n: {
         changeLanguage: () => new Promise(() => {}),
-        language: 'en',
+        language: "en",
         exists: () => true,
       },
     }),
-  }
+  };
 });
 
 describe("ExpandableMessage", () => {
   it("should render with neutral border for non-action messages", () => {
     renderWithProviders(<ExpandableMessage message="Hello" type="thought" />);
     const element = screen.getByText("Hello");
-    const container = element.closest("div.flex.gap-2.items-center.justify-start");
+    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 with neutral border for error messages", () => {
-    renderWithProviders(<ExpandableMessage message="Error occurred" type="error" />);
+    renderWithProviders(
+      <ExpandableMessage message="Error occurred" type="error" />,
+    );
     const element = screen.getByText("Error occurred");
-    const container = element.closest("div.flex.gap-2.items-center.justify-start");
+    const container = element.closest(
+      "div.flex.gap-2.items-center.justify-start",
+    );
     expect(container).toHaveClass("border-danger");
     expect(screen.queryByTestId("status-icon")).not.toBeInTheDocument();
   });
@@ -43,10 +49,12 @@ describe("ExpandableMessage", () => {
         message="Command executed successfully"
         type="action"
         success={true}
-      />
+      />,
     );
     const element = screen.getByText("OBSERVATION_MESSAGE$RUN");
-    const container = element.closest("div.flex.gap-2.items-center.justify-start");
+    const container = element.closest(
+      "div.flex.gap-2.items-center.justify-start",
+    );
     expect(container).toHaveClass("border-neutral-300");
     const icon = screen.getByTestId("status-icon");
     expect(icon).toHaveClass("fill-success");
@@ -59,10 +67,12 @@ describe("ExpandableMessage", () => {
         message="Command failed"
         type="action"
         success={false}
-      />
+      />,
     );
     const element = screen.getByText("OBSERVATION_MESSAGE$RUN");
-    const container = element.closest("div.flex.gap-2.items-center.justify-start");
+    const container = element.closest(
+      "div.flex.gap-2.items-center.justify-start",
+    );
     expect(container).toHaveClass("border-neutral-300");
     const icon = screen.getByTestId("status-icon");
     expect(icon).toHaveClass("fill-danger");
@@ -74,10 +84,12 @@ describe("ExpandableMessage", () => {
         id="OBSERVATION_MESSAGE$RUN"
         message="Running command"
         type="action"
-      />
+      />,
     );
     const element = screen.getByText("OBSERVATION_MESSAGE$RUN");
-    const container = element.closest("div.flex.gap-2.items-center.justify-start");
+    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();
   });

frontend/__tests__/components/features/conversation-panel/conversation-panel.test.tsx

@@ -71,7 +71,7 @@ describe("ConversationPanel", () => {
 
     renderConversationPanel();
 
-    const emptyState = await screen.findByText("No conversations found");
+    const emptyState = await screen.findByText("CONVERSATION$NO_CONVERSATIONS");
     expect(emptyState).toBeInTheDocument();
   });
 

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

@@ -3,7 +3,6 @@ import { describe, expect, it, vi } from "vitest";
 import { renderWithProviders } from "test-utils";
 import { GitHubRepositorySelector } from "#/components/features/github/github-repo-selector";
 import OpenHands from "#/api/open-hands";
-import * as GitHubAPI from "#/api/github";
 
 describe("GitHubRepositorySelector", () => {
   const onInputChangeMock = vi.fn();
@@ -20,7 +19,7 @@ describe("GitHubRepositorySelector", () => {
     );
 
     expect(
-      screen.getByPlaceholderText("Select a GitHub project"),
+      screen.getByPlaceholderText("LANDING$SELECT_REPO"),
     ).toBeInTheDocument();
   });
 
@@ -60,8 +59,8 @@ describe("GitHubRepositorySelector", () => {
     ];
 
     const searchPublicRepositoriesSpy = vi.spyOn(
-      GitHubAPI,
-      "searchPublicRepositories",
+      OpenHands,
+      "searchGitHubRepositories",
     );
     searchPublicRepositoriesSpy.mockResolvedValue(mockSearchedRepos);
 

frontend/__tests__/components/features/sidebar/sidebar.test.tsx

@@ -128,7 +128,7 @@ describe("Sidebar", () => {
       await user.click(norskOption);
 
       const tokenInput =
-        within(accountSettingsModal).getByLabelText(/github token/i);
+        within(accountSettingsModal).getByLabelText(/GITHUB\$TOKEN_LABEL/i);
       await user.type(tokenInput, "new-token");
 
       const saveButton =
@@ -151,7 +151,11 @@ describe("Sidebar", () => {
 
       const settingsModal = screen.getByTestId("ai-config-modal");
 
-      const apiKeyInput = within(settingsModal).getByLabelText(/api key/i);
+      // Click the advanced options switch to show the API key input
+      const advancedOptionsSwitch = within(settingsModal).getByTestId("advanced-option-switch");
+      await user.click(advancedOptionsSwitch);
+
+      const apiKeyInput = within(settingsModal).getByLabelText(/API\$KEY/i);
       await user.type(apiKeyInput, "SET");
 
       const saveButton = within(settingsModal).getByTestId(
@@ -162,7 +166,7 @@ describe("Sidebar", () => {
       expect(saveSettingsSpy).toHaveBeenCalledWith({
         ...MOCK_USER_PREFERENCES.settings,
         llm_api_key: undefined,
-        llm_base_url: undefined,
+        llm_base_url: "",
         security_analyzer: undefined,
       });
     });

frontend/__tests__/components/feedback-form.test.tsx

@@ -1,11 +1,10 @@
-import { afterEach, beforeEach, describe, expect, it, vi } from "vitest";
-import * as router from "react-router";
+import { afterEach, describe, expect, it, vi } from "vitest";
 
 // Mock useParams before importing components
 vi.mock("react-router", async () => {
   const actual = await vi.importActual("react-router");
   return {
-    ...actual as object,
+    ...(actual as object),
     useParams: () => ({ conversationId: "test-conversation-id" }),
   };
 });
@@ -14,6 +13,7 @@ import { screen } from "@testing-library/react";
 import userEvent from "@testing-library/user-event";
 import { renderWithProviders } from "test-utils";
 import { FeedbackForm } from "#/components/features/feedback/feedback-form";
+import { I18nKey } from "#/i18n/declaration";
 
 describe("FeedbackForm", () => {
   const user = userEvent.setup();
@@ -28,20 +28,20 @@ describe("FeedbackForm", () => {
       <FeedbackForm polarity="positive" onClose={onCloseMock} />,
     );
 
-    screen.getByLabelText("Email");
-    screen.getByLabelText("Private");
-    screen.getByLabelText("Public");
+    screen.getByLabelText(I18nKey.FEEDBACK$EMAIL_LABEL);
+    screen.getByLabelText(I18nKey.FEEDBACK$PRIVATE_LABEL);
+    screen.getByLabelText(I18nKey.FEEDBACK$PUBLIC_LABEL);
 
-    screen.getByRole("button", { name: "Submit" });
-    screen.getByRole("button", { name: "Cancel" });
+    screen.getByRole("button", { name: I18nKey.FEEDBACK$CONTRIBUTE_LABEL });
+    screen.getByRole("button", { name: I18nKey.FEEDBACK$CANCEL_LABEL });
   });
 
   it("should switch between private and public permissions", async () => {
     renderWithProviders(
       <FeedbackForm polarity="positive" onClose={onCloseMock} />,
     );
-    const privateRadio = screen.getByLabelText("Private");
-    const publicRadio = screen.getByLabelText("Public");
+    const privateRadio = screen.getByLabelText(I18nKey.FEEDBACK$PRIVATE_LABEL);
+    const publicRadio = screen.getByLabelText(I18nKey.FEEDBACK$PUBLIC_LABEL);
 
     expect(privateRadio).toBeChecked(); // private is the default value
     expect(publicRadio).not.toBeChecked();
@@ -59,7 +59,9 @@ describe("FeedbackForm", () => {
     renderWithProviders(
       <FeedbackForm polarity="positive" onClose={onCloseMock} />,
     );
-    await user.click(screen.getByRole("button", { name: "Cancel" }));
+    await user.click(
+      screen.getByRole("button", { name: I18nKey.FEEDBACK$CANCEL_LABEL }),
+    );
 
     expect(onCloseMock).toHaveBeenCalled();
   });

frontend/__tests__/components/interactive-chat-box.test.tsx

@@ -157,7 +157,7 @@ describe("InteractiveChatBox", () => {
     expect(onChange).not.toHaveBeenCalledWith("");
 
     // Submit the message with image
-    const submitButton = screen.getByRole("button", { name: "Send" });
+    const submitButton = screen.getByRole("button", { name: "BUTTON$SEND" });
     await user.click(submitButton);
 
     // Verify onSubmit was called with the message and image

frontend/__tests__/components/landing-translations.test.tsx

@@ -0,0 +1,190 @@
+import { render, screen } from "@testing-library/react";
+import { test, expect, describe, vi } from "vitest";
+import { useTranslation } from "react-i18next";
+import translations from "../../src/i18n/translation.json";
+import { UserAvatar } from "../../src/components/features/sidebar/user-avatar";
+
+vi.mock("@nextui-org/react", () => ({
+  Tooltip: ({ content, children }: { content: string; children: React.ReactNode }) => (
+    <div>
+      {children}
+      <div>{content}</div>
+    </div>
+  ),
+}));
+
+const supportedLanguages = ['en', 'ja', 'zh-CN', 'zh-TW', 'ko-KR', 'de', 'no', 'it', 'pt', 'es', 'ar', 'fr', 'tr'];
+
+// Helper function to check if a translation exists for all supported languages
+function checkTranslationExists(key: string) {
+  const missingTranslations: string[] = [];
+
+  const translationEntry = (translations as Record<string, Record<string, string>>)[key];
+  if (!translationEntry) {
+    throw new Error(`Translation key "${key}" does not exist in translation.json`);
+  }
+
+  for (const lang of supportedLanguages) {
+    if (!translationEntry[lang]) {
+      missingTranslations.push(lang);
+    }
+  }
+
+  return missingTranslations;
+}
+
+// Helper function to find duplicate translation keys
+function findDuplicateKeys(obj: Record<string, any>) {
+  const seen = new Set<string>();
+  const duplicates = new Set<string>();
+
+  // Only check top-level keys as these are our translation keys
+  for (const key in obj) {
+    if (seen.has(key)) {
+      duplicates.add(key);
+    } else {
+      seen.add(key);
+    }
+  }
+
+  return Array.from(duplicates);
+}
+
+vi.mock("react-i18next", () => ({
+  useTranslation: () => ({
+    t: (key: string) => {
+      const translationEntry = (translations as Record<string, Record<string, string>>)[key];
+      return translationEntry?.ja || key;
+    },
+  }),
+}));
+
+describe("Landing page translations", () => {
+  test("should render Japanese translations correctly", () => {
+    // Mock a simple component that uses the translations
+    const TestComponent = () => {
+      const { t } = useTranslation();
+      return (
+        <div>
+          <UserAvatar onClick={() => {}} />
+          <div data-testid="main-content">
+            <h1>{t("LANDING$TITLE")}</h1>
+            <button>{t("VSCODE$OPEN")}</button>
+            <button>{t("SUGGESTIONS$INCREASE_TEST_COVERAGE")}</button>
+            <button>{t("SUGGESTIONS$AUTO_MERGE_PRS")}</button>
+            <button>{t("SUGGESTIONS$FIX_README")}</button>
+            <button>{t("SUGGESTIONS$CLEAN_DEPENDENCIES")}</button>
+          </div>
+          <div data-testid="tabs">
+            <span>{t("WORKSPACE$TERMINAL_TAB_LABEL")}</span>
+            <span>{t("WORKSPACE$BROWSER_TAB_LABEL")}</span>
+            <span>{t("WORKSPACE$JUPYTER_TAB_LABEL")}</span>
+            <span>{t("WORKSPACE$CODE_EDITOR_TAB_LABEL")}</span>
+          </div>
+          <div data-testid="workspace-label">{t("WORKSPACE$TITLE")}</div>
+          <button data-testid="new-project">{t("PROJECT$NEW_PROJECT")}</button>
+          <div data-testid="status">
+            <span>{t("TERMINAL$WAITING_FOR_CLIENT")}</span>
+            <span>{t("STATUS$CONNECTED")}</span>
+            <span>{t("STATUS$CONNECTED_TO_SERVER")}</span>
+          </div>
+          <div data-testid="time">
+            <span>{`5 ${t("TIME$MINUTES_AGO")}`}</span>
+            <span>{`2 ${t("TIME$HOURS_AGO")}`}</span>
+            <span>{`3 ${t("TIME$DAYS_AGO")}`}</span>
+          </div>
+        </div>
+      );
+    };
+
+    render(<TestComponent />);
+
+    // Check main content translations
+    expect(screen.getByText("開発を始めましょう！")).toBeInTheDocument();
+    expect(screen.getByText("VS Codeで開く")).toBeInTheDocument();
+    expect(screen.getByText("テストカバレッジを向上させる")).toBeInTheDocument();
+    expect(screen.getByText("Dependabot PRを自動マージ")).toBeInTheDocument();
+    expect(screen.getByText("READMEを改善")).toBeInTheDocument();
+    expect(screen.getByText("依存関係を整理")).toBeInTheDocument();
+
+    // Check user avatar tooltip
+    const userAvatar = screen.getByTestId("user-avatar");
+    userAvatar.focus();
+    expect(screen.getByText("アカウント設定")).toBeInTheDocument();
+
+    // Check tab labels
+    const tabs = screen.getByTestId("tabs");
+    expect(tabs).toHaveTextContent("ターミナル");
+    expect(tabs).toHaveTextContent("ブラウザ");
+    expect(tabs).toHaveTextContent("Jupyter");
+    expect(tabs).toHaveTextContent("コードエディタ");
+
+    // Check workspace label and new project button
+    expect(screen.getByTestId("workspace-label")).toHaveTextContent("ワークスペース");
+    expect(screen.getByTestId("new-project")).toHaveTextContent("新規プロジェクト");
+
+    // Check status messages
+    const status = screen.getByTestId("status");
+    expect(status).toHaveTextContent("クライアントの準備を待機中");
+    expect(status).toHaveTextContent("接続済み");
+    expect(status).toHaveTextContent("サーバーに接続済み");
+
+    // Check account settings menu
+    expect(screen.getByText("アカウント設定")).toBeInTheDocument();
+
+    // Check time-related translations
+    const time = screen.getByTestId("time");
+    expect(time).toHaveTextContent("5 分前");
+    expect(time).toHaveTextContent("2 時間前");
+    expect(time).toHaveTextContent("3 日前");
+  });
+
+  test("all translation keys should have translations for all supported languages", () => {
+    // Test all translation keys used in the component
+    const translationKeys = [
+      "LANDING$TITLE",
+      "VSCODE$OPEN",
+      "SUGGESTIONS$INCREASE_TEST_COVERAGE",
+      "SUGGESTIONS$AUTO_MERGE_PRS",
+      "SUGGESTIONS$FIX_README",
+      "SUGGESTIONS$CLEAN_DEPENDENCIES",
+      "WORKSPACE$TERMINAL_TAB_LABEL",
+      "WORKSPACE$BROWSER_TAB_LABEL",
+      "WORKSPACE$JUPYTER_TAB_LABEL",
+      "WORKSPACE$CODE_EDITOR_TAB_LABEL",
+      "WORKSPACE$TITLE",
+      "PROJECT$NEW_PROJECT",
+      "TERMINAL$WAITING_FOR_CLIENT",
+      "STATUS$CONNECTED",
+      "STATUS$CONNECTED_TO_SERVER",
+      "TIME$MINUTES_AGO",
+      "TIME$HOURS_AGO",
+      "TIME$DAYS_AGO"
+    ];
+
+    // Check all keys and collect missing translations
+    const missingTranslationsMap = new Map<string, string[]>();
+    translationKeys.forEach(key => {
+      const missing = checkTranslationExists(key);
+      if (missing.length > 0) {
+        missingTranslationsMap.set(key, missing);
+      }
+    });
+
+    // If any translations are missing, throw an error with all missing translations
+    if (missingTranslationsMap.size > 0) {
+      const errorMessage = Array.from(missingTranslationsMap.entries())
+        .map(([key, langs]) => `\n- "${key}" is missing translations for: ${langs.join(', ')}`)
+        .join('');
+      throw new Error(`Missing translations:${errorMessage}`);
+    }
+  });
+
+  test("translation file should not have duplicate keys", () => {
+    const duplicates = findDuplicateKeys(translations);
+
+    if (duplicates.length > 0) {
+      throw new Error(`Found duplicate translation keys: ${duplicates.join(', ')}`);
+    }
+  });
+});

frontend/__tests__/components/modals/settings/model-selector.test.tsx

@@ -1,7 +1,23 @@
-import { describe, it, expect } from "vitest";
+import { describe, it, expect, vi } from "vitest";
 import { render, screen } from "@testing-library/react";
 import userEvent from "@testing-library/user-event";
 import { ModelSelector } from "#/components/shared/modals/settings/model-selector";
+import { I18nKey } from "#/i18n/declaration";
+
+// Mock react-i18next
+vi.mock("react-i18next", () => ({
+  useTranslation: () => ({
+    t: (key: string) => {
+      const translations: { [key: string]: string } = {
+        LLM$PROVIDER: "LLM Provider",
+        LLM$MODEL: "LLM Model",
+        LLM$SELECT_PROVIDER_PLACEHOLDER: "Select a provider",
+        LLM$SELECT_MODEL_PLACEHOLDER: "Select a model",
+      };
+      return translations[key] || key;
+    },
+  }),
+}));
 
 describe("ModelSelector", () => {
   const models = {

frontend/__tests__/components/suggestion-item.test.tsx

@@ -2,6 +2,20 @@ import { render, screen } from "@testing-library/react";
 import userEvent from "@testing-library/user-event";
 import { afterEach, describe, expect, it, vi } from "vitest";
 import { SuggestionItem } from "#/components/features/suggestions/suggestion-item";
+import { I18nKey } from "#/i18n/declaration";
+
+vi.mock("react-i18next", () => ({
+  useTranslation: () => ({
+    t: (key: string) => {
+      const translations: Record<string, string> = {
+        "SUGGESTIONS$TODO_APP": "ToDoリストアプリを開発する",
+        "LANDING$BUILD_APP_BUTTON": "プルリクエストを表示するアプリを開発する",
+        "SUGGESTIONS$HACKER_NEWS": "Hacker Newsのトップ記事を表示するbashスクリプトを作成する",
+      };
+      return translations[key] || key;
+    },
+  }),
+}));
 
 describe("SuggestionItem", () => {
   const suggestionItem = { label: "suggestion1", value: "a long text value" };
@@ -18,6 +32,19 @@ describe("SuggestionItem", () => {
     expect(screen.getByText(/suggestion1/i)).toBeInTheDocument();
   });
 
+  it("should render a translated suggestion when using I18nKey", async () => {
+    const translatedSuggestion = {
+      label: I18nKey.SUGGESTIONS$TODO_APP,
+      value: "todo app value",
+    };
+
+    const { container } = render(<SuggestionItem suggestion={translatedSuggestion} onClick={onClick} />);
+    console.log('Rendered HTML:', container.innerHTML);
+
+
+    expect(screen.getByText("ToDoリストアプリを開発する")).toBeInTheDocument();
+  });
+
   it("should call onClick when clicking a suggestion", async () => {
     const user = userEvent.setup();
     render(<SuggestionItem suggestion={suggestionItem} onClick={onClick} />);

frontend/__tests__/components/user-avatar.test.tsx

@@ -14,7 +14,7 @@ describe("UserAvatar", () => {
     render(<UserAvatar onClick={onClickMock} />);
     expect(screen.getByTestId("user-avatar")).toBeInTheDocument();
     expect(
-      screen.getByLabelText("user avatar placeholder"),
+      screen.getByLabelText("USER$AVATAR_PLACEHOLDER"),
     ).toBeInTheDocument();
   });
 
@@ -38,7 +38,7 @@ describe("UserAvatar", () => {
 
     expect(screen.getByAltText("user avatar")).toBeInTheDocument();
     expect(
-      screen.queryByLabelText("user avatar placeholder"),
+      screen.queryByLabelText("USER$AVATAR_PLACEHOLDER"),
     ).not.toBeInTheDocument();
   });
 
@@ -46,13 +46,13 @@ describe("UserAvatar", () => {
     const { rerender } = render(<UserAvatar onClick={onClickMock} />);
     expect(screen.queryByTestId("loading-spinner")).not.toBeInTheDocument();
     expect(
-      screen.getByLabelText("user avatar placeholder"),
+      screen.getByLabelText("USER$AVATAR_PLACEHOLDER"),
     ).toBeInTheDocument();
 
     rerender(<UserAvatar onClick={onClickMock} isLoading />);
     expect(screen.getByTestId("loading-spinner")).toBeInTheDocument();
     expect(
-      screen.queryByLabelText("user avatar placeholder"),
+      screen.queryByLabelText("USER$AVATAR_PLACEHOLDER"),
     ).not.toBeInTheDocument();
 
     rerender(

frontend/__tests__/i18n/duplicate-keys.test.ts

@@ -0,0 +1,76 @@
+import { describe, expect, it } from 'vitest';
+import fs from 'fs';
+import path from 'path';
+
+describe('translation.json', () => {
+  it('should not have duplicate translation keys', () => {
+    // Read the translation.json file
+    const translationPath = path.join(__dirname, '../../src/i18n/translation.json');
+    const translationContent = fs.readFileSync(translationPath, 'utf-8');
+
+    // First, let's check for exact string matches of key definitions
+    const keyRegex = /"([^"]+)": {/g;
+    const matches = translationContent.matchAll(keyRegex);
+    const keyOccurrences = new Map<string, number>();
+    const duplicateKeys: string[] = [];
+
+    for (const match of matches) {
+      const key = match[1];
+      const count = (keyOccurrences.get(key) || 0) + 1;
+      keyOccurrences.set(key, count);
+      if (count > 1) {
+        duplicateKeys.push(key);
+      }
+    }
+
+    // Remove duplicates from duplicateKeys array
+    const uniqueDuplicates = [...new Set(duplicateKeys)];
+
+    // If there are duplicates, create a helpful error message
+    if (uniqueDuplicates.length > 0) {
+      const errorMessage = `Found duplicate translation keys:\n${uniqueDuplicates
+        .map((key) => `  - "${key}" appears ${keyOccurrences.get(key)} times`)
+        .join('\n')}`;
+      throw new Error(errorMessage);
+    }
+
+    // Expect no duplicates (this will pass if we reach here)
+    expect(uniqueDuplicates).toHaveLength(0);
+  });
+
+  it('should have consistent translations for each key', () => {
+    // Read the translation.json file
+    const translationPath = path.join(__dirname, '../../src/i18n/translation.json');
+    const translationContent = fs.readFileSync(translationPath, 'utf-8');
+    const translations = JSON.parse(translationContent);
+
+    // Create a map to store English translations for each key
+    const englishTranslations = new Map<string, string>();
+    const inconsistentKeys: string[] = [];
+
+    // Check each key's English translation
+    Object.entries(translations).forEach(([key, value]: [string, any]) => {
+      if (typeof value === 'object' && value.en !== undefined) {
+        const currentEn = value.en.toLowerCase();
+        const existingEn = englishTranslations.get(key)?.toLowerCase();
+
+        if (existingEn !== undefined && existingEn !== currentEn) {
+          inconsistentKeys.push(key);
+        } else {
+          englishTranslations.set(key, value.en);
+        }
+      }
+    });
+
+    // If there are inconsistencies, create a helpful error message
+    if (inconsistentKeys.length > 0) {
+      const errorMessage = `Found inconsistent translations for keys:\n${inconsistentKeys
+        .map((key) => `  - "${key}" has multiple different English translations`)
+        .join('\n')}`;
+      throw new Error(errorMessage);
+    }
+
+    // Expect no inconsistencies
+    expect(inconsistentKeys).toHaveLength(0);
+  });
+});

frontend/__tests__/i18n/translations.test.tsx

@@ -0,0 +1,20 @@
+import { screen } from '@testing-library/react';
+import { describe, expect, it } from 'vitest';
+import i18n from '../../src/i18n';
+import { AccountSettingsContextMenu } from '../../src/components/features/context-menu/account-settings-context-menu';
+import { renderWithProviders } from '../../test-utils';
+
+describe('Translations', () => {
+  it('should render translated text', () => {
+    i18n.changeLanguage('en');
+    renderWithProviders(
+      <AccountSettingsContextMenu
+        onClickAccountSettings={() => {}}
+        onLogout={() => {}}
+        onClose={() => {}}
+        isLoggedIn={true}
+      />
+    );
+    expect(screen.getByTestId('account-settings-context-menu')).toBeInTheDocument();
+  });
+});

frontend/__tests__/routes/_oh.test.tsx

@@ -1,5 +1,4 @@
 import { afterEach, beforeAll, describe, expect, it, vi } from "vitest";
-import * as router from "react-router";
 import { createRoutesStub } from "react-router";
 import { screen, waitFor, within } from "@testing-library/react";
 import { renderWithProviders } from "test-utils";

frontend/__tests__/utils/check-hardcoded-strings.test.tsx

@@ -0,0 +1,40 @@
+import { render, screen } from "@testing-library/react";
+import { test, expect, describe, vi } from "vitest";
+import { InteractiveChatBox } from "#/components/features/chat/interactive-chat-box";
+import { ChatInput } from "#/components/features/chat/chat-input";
+
+// Mock react-i18next
+vi.mock("react-i18next", () => ({
+  useTranslation: () => ({
+    t: (key: string) => key,
+  }),
+}));
+
+describe("Check for hardcoded English strings", () => {
+  test("InteractiveChatBox should not have hardcoded English strings", () => {
+    const { container } = render(
+      <InteractiveChatBox
+        onSubmit={() => {}}
+        onStop={() => {}}
+      />
+    );
+
+    // Get all text content
+    const text = container.textContent;
+
+    // List of English strings that should be translated
+    const hardcodedStrings = [
+      "What do you want to build?",
+    ];
+
+    // Check each string
+    hardcodedStrings.forEach(str => {
+      expect(text).not.toContain(str);
+    });
+  });
+
+  test("ChatInput should use translation key for placeholder", () => {
+    render(<ChatInput onSubmit={() => {}} />);
+    screen.getByPlaceholderText("SUGGESTIONS$WHAT_TO_BUILD");
+  });
+});

frontend/__tests__/utils/i18n-test-utils.tsx

@@ -0,0 +1,29 @@
+import { ReactNode } from "react";
+import { I18nextProvider } from "react-i18next";
+
+const mockI18n = {
+  language: "ja",
+  t: (key: string) => {
+    const translations: Record<string, string> = {
+      "SUGGESTIONS$TODO_APP": "ToDoリストアプリを開発する",
+      "LANDING$BUILD_APP_BUTTON": "プルリクエストを表示するアプリを開発する",
+      "SUGGESTIONS$HACKER_NEWS": "Hacker Newsのトップ記事を表示するbashスクリプトを作成する",
+      "LANDING$TITLE": "一緒に開発を始めましょう！",
+      "OPEN_IN_VSCODE": "VS Codeで開く",
+      "INCREASE_TEST_COVERAGE": "テストカバレッジを向上",
+      "AUTO_MERGE_PRS": "PRを自動マージ",
+      "FIX_README": "READMEを修正",
+      "CLEAN_DEPENDENCIES": "依存関係を整理"
+    };
+    return translations[key] || key;
+  },
+  exists: () => true,
+  changeLanguage: () => new Promise(() => {}),
+  use: () => mockI18n,
+};
+
+export function I18nTestProvider({ children }: { children: ReactNode }) {
+  return (
+    <I18nextProvider i18n={mockI18n as any}>{children}</I18nextProvider>
+  );
+}

frontend/__tests__/utils/test-config.tsx

@@ -0,0 +1,20 @@
+import { vi } from "vitest";
+import OpenHands from "#/api/open-hands";
+
+export const setupTestConfig = () => {
+  const getConfigSpy = vi.spyOn(OpenHands, "getConfig");
+  getConfigSpy.mockResolvedValue({
+    APP_MODE: "oss",
+    GITHUB_CLIENT_ID: "test-id",
+    POSTHOG_CLIENT_KEY: "test-key",
+  });
+};
+
+export const setupSaasTestConfig = () => {
+  const getConfigSpy = vi.spyOn(OpenHands, "getConfig");
+  getConfigSpy.mockResolvedValue({
+    APP_MODE: "saas",
+    GITHUB_CLIENT_ID: "test-id",
+    POSTHOG_CLIENT_KEY: "test-key",
+  });
+};

frontend/src/api/github-axios-instance.ts

@@ -41,6 +41,7 @@ export const isGitHubErrorReponse = <T extends object | Array<unknown>>(
 
 // Axios interceptor to handle token refresh
 const setupAxiosInterceptors = (
+  appMode: string,
   refreshToken: () => Promise<boolean>,
   logout: () => void,
 ) => {
@@ -74,18 +75,21 @@ const setupAxiosInterceptors = (
         !originalRequest._retry // Prevent infinite retry loops
       ) {
         originalRequest._retry = true;
-        try {
-          const refreshed = await refreshToken();
-          if (refreshed) {
-            return await github(originalRequest);
-          }
 
-          logout();
-          return await Promise.reject(new Error("Failed to refresh token"));
-        } catch (refreshError) {
-          // If token refresh fails, evict the user
-          logout();
-          return Promise.reject(refreshError);
+        if (appMode === "saas") {
+          try {
+            const refreshed = await refreshToken();
+            if (refreshed) {
+              return await github(originalRequest);
+            }
+
+            logout();
+            return await Promise.reject(new Error("Failed to refresh token"));
+          } catch (refreshError) {
+            // If token refresh fails, evict the user
+            logout();
+            return Promise.reject(refreshError);
+          }
         }
       }
 

frontend/src/api/github.ts

@@ -1,19 +1,6 @@
 import { extractNextPageFromLink } from "#/utils/extract-next-page-from-link";
-import { github } from "./github-axios-instance";
 import { openHands } from "./open-hands-axios";
 
-/**
- * Given the user, retrieves app installations IDs for OpenHands Github App
- * Uses user access token for Github App
- */
-export const retrieveGitHubAppInstallations = async (): Promise<number[]> => {
-  const response = await github.get<GithubAppInstallation>(
-    "/user/installations",
-  );
-
-  return response.data.installations.map((installation) => installation.id);
-};
-
 /**
  * Retrieves repositories where OpenHands Github App has been installed
  * @param installationIndex Pagination cursor position for app installation IDs
@@ -82,72 +69,3 @@ export const retrieveGitHubUserRepositories = async (
 
   return { data: response.data, nextPage };
 };
-
-/**
- * Given a GitHub token, retrieves the authenticated user
- * @returns The authenticated user or an error response
- */
-export const retrieveGitHubUser = async () => {
-  const response = await github.get<GitHubUser>("/user");
-
-  const { data } = response;
-
-  const user: GitHubUser = {
-    id: data.id,
-    login: data.login,
-    avatar_url: data.avatar_url,
-    company: data.company,
-    name: data.name,
-    email: data.email,
-  };
-
-  return user;
-};
-
-export const searchPublicRepositories = async (
-  query: string,
-  per_page = 5,
-  sort: "" | "updated" | "stars" | "forks" = "stars",
-  order: "desc" | "asc" = "desc",
-): Promise<GitHubRepository[]> => {
-  const response = await github.get<{ items: GitHubRepository[] }>(
-    "/search/repositories",
-    {
-      params: {
-        q: query,
-        per_page,
-        sort,
-        order,
-      },
-    },
-  );
-  return response.data.items;
-};
-
-export const retrieveLatestGitHubCommit = async (
-  repository: string,
-): Promise<GitHubCommit | null> => {
-  try {
-    const response = await github.get<GitHubCommit[]>(
-      `/repos/${repository}/commits`,
-      {
-        params: {
-          per_page: 1,
-        },
-      },
-    );
-    return response.data[0] || null;
-  } catch (error) {
-    if (!error || typeof error !== "object") {
-      throw new Error("Unknown error occurred");
-    }
-    const axiosError = error as { response?: { status: number } };
-    if (axiosError.response?.status === 409) {
-      // Repository is empty, no commits yet
-      return null;
-    }
-    throw new Error(
-      error instanceof Error ? error.message : "Unknown error occurred",
-    );
-  }
-};

frontend/src/api/open-hands.ts

@@ -315,6 +315,45 @@ class OpenHands {
     const data = await openHands.post("/api/settings", settings);
     return data.status === 200;
   }
+
+  static async getGitHubUser(): Promise<GitHubUser> {
+    const response = await openHands.get<GitHubUser>("/api/github/user");
+
+    const { data } = response;
+
+    const user: GitHubUser = {
+      id: data.id,
+      login: data.login,
+      avatar_url: data.avatar_url,
+      company: data.company,
+      name: data.name,
+      email: data.email,
+    };
+
+    return user;
+  }
+
+  static async getGitHubUserInstallationIds(): Promise<number[]> {
+    const response = await openHands.get<number[]>("/api/github/installations");
+    return response.data;
+  }
+
+  static async searchGitHubRepositories(
+    query: string,
+    per_page = 5,
+  ): Promise<GitHubRepository[]> {
+    const response = await openHands.get<{ items: GitHubRepository[] }>(
+      "/api/github/search/repositories",
+      {
+        params: {
+          query,
+          per_page,
+        },
+      },
+    );
+
+    return response.data.items;
+  }
 }
 
 export default OpenHands;

frontend/src/components/features/browser/empty-browser-message.tsx

@@ -8,7 +8,7 @@ export function EmptyBrowserMessage() {
   return (
     <div className="flex flex-col items-center h-full justify-center">
       <IoIosGlobe size={100} />
-      {t(I18nKey.BROWSER$EMPTY_MESSAGE)}
+      {t(I18nKey.BROWSER$NO_PAGE_LOADED)}
     </div>
   );
 }

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

@@ -1,5 +1,7 @@
 import React from "react";
 import TextareaAutosize from "react-textarea-autosize";
+import { useTranslation } from "react-i18next";
+import { I18nKey } from "#/i18n/declaration";
 import { cn } from "#/utils/utils";
 import { SubmitButton } from "#/components/shared/buttons/submit-button";
 import { StopButton } from "#/components/shared/buttons/stop-button";
@@ -8,7 +10,6 @@ interface ChatInputProps {
   name?: string;
   button?: "submit" | "stop";
   disabled?: boolean;
-  placeholder?: string;
   showButton?: boolean;
   value?: string;
   maxRows?: number;
@@ -26,7 +27,6 @@ export function ChatInput({
   name,
   button = "submit",
   disabled,
-  placeholder,
   showButton = true,
   value,
   maxRows = 4,
@@ -39,6 +39,7 @@ export function ChatInput({
   className,
   buttonClassName,
 }: ChatInputProps) {
+  const { t } = useTranslation();
   const textareaRef = React.useRef<HTMLTextAreaElement>(null);
   const [isDraggingOver, setIsDraggingOver] = React.useState(false);
 
@@ -117,7 +118,7 @@ export function ChatInput({
       <TextareaAutosize
         ref={textareaRef}
         name={name}
-        placeholder={placeholder}
+        placeholder={t(I18nKey.SUGGESTIONS$WHAT_TO_BUILD)}
         onKeyDown={handleKeyPress}
         onChange={handleChange}
         onFocus={onFocus}

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

@@ -1,4 +1,6 @@
+import { useTranslation } from "react-i18next";
 import { Suggestions } from "#/components/features/suggestions/suggestions";
+import { I18nKey } from "#/i18n/declaration";
 import BuildIt from "#/icons/build-it.svg?react";
 import { SUGGESTIONS } from "#/utils/suggestions";
 
@@ -7,12 +9,14 @@ interface ChatSuggestionsProps {
 }
 
 export function ChatSuggestions({ onSuggestionsClick }: ChatSuggestionsProps) {
+  const { t } = useTranslation();
+
   return (
     <div className="flex flex-col gap-6 h-full px-4 items-center justify-center">
       <div className="flex flex-col items-center p-4 bg-neutral-700 rounded-xl w-full">
         <BuildIt width={45} height={54} />
         <span className="font-semibold text-[20px] leading-6 -tracking-[0.01em] gap-1">
-          Let&apos;s start building!
+          {t(I18nKey.LANDING$TITLE)}
         </span>
       </div>
       <Suggestions

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

@@ -68,7 +68,6 @@ export function InteractiveChatBox({
         <ChatInput
           disabled={isDisabled}
           button={mode}
-          placeholder="What do you want to build?"
           onChange={onChange}
           onSubmit={handleSubmit}
           onStop={onStop}

frontend/src/components/features/controls/agent-control-bar.tsx

@@ -1,4 +1,6 @@
 import { useSelector } from "react-redux";
+import { useTranslation } from "react-i18next";
+import { I18nKey } from "#/i18n/declaration";
 import PauseIcon from "#/assets/pause";
 import PlayIcon from "#/assets/play";
 import { generateAgentStateChangeEvent } from "#/services/agent-state-service";
@@ -9,6 +11,7 @@ import { IGNORE_TASK_STATE_MAP } from "#/ignore-task-state-map.constant";
 import { ActionButton } from "#/components/shared/buttons/action-button";
 
 export function AgentControlBar() {
+  const { t } = useTranslation();
   const { send } = useWsClient();
   const { curAgentState } = useSelector((state: RootState) => state.agent);
 
@@ -27,8 +30,8 @@ export function AgentControlBar() {
         }
         content={
           curAgentState === AgentState.PAUSED
-            ? "Resume the agent task"
-            : "Pause the current task"
+            ? t(I18nKey.AGENT$RESUME_TASK)
+            : t(I18nKey.AGENT$PAUSE_TASK)
         }
         action={
           curAgentState === AgentState.PAUSED

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

@@ -1,5 +1,7 @@
 import React from "react";
 import { NavLink, useParams } from "react-router";
+import { useTranslation } from "react-i18next";
+import { I18nKey } from "#/i18n/declaration";
 import { ConversationCard } from "./conversation-card";
 import { useUserConversations } from "#/hooks/query/use-user-conversations";
 import { useDeleteConversation } from "#/hooks/mutation/use-delete-conversation";
@@ -15,6 +17,7 @@ interface ConversationPanelProps {
 }
 
 export function ConversationPanel({ onClose }: ConversationPanelProps) {
+  const { t } = useTranslation();
   const { conversationId: cid } = useParams();
   const endSession = useEndSession();
   const ref = useClickOutsideElement<HTMLDivElement>(onClose);
@@ -78,7 +81,9 @@ export function ConversationPanel({ onClose }: ConversationPanelProps) {
       )}
       {conversations?.length === 0 && (
         <div className="flex flex-col items-center justify-center h-full">
-          <p className="text-neutral-400">No conversations found</p>
+          <p className="text-neutral-400">
+            {t(I18nKey.CONVERSATION$NO_CONVERSATIONS)}
+          </p>
         </div>
       )}
       {conversations?.map((project) => (

frontend/src/components/features/conversation-panel/new-conversation-button.tsx

@@ -1,8 +1,12 @@
+import { useTranslation } from "react-i18next";
+import { I18nKey } from "#/i18n/declaration";
+
 interface NewConversationButtonProps {
   onClick: () => void;
 }
 
 export function NewConversationButton({ onClick }: NewConversationButtonProps) {
+  const { t } = useTranslation();
   return (
     <button
       data-testid="new-conversation-button"
@@ -10,7 +14,7 @@ export function NewConversationButton({ onClick }: NewConversationButtonProps) {
       onClick={onClick}
       className="font-bold bg-[#4465DB] px-2 py-1 rounded"
     >
-      + New Project
+      + {t(I18nKey.PROJECT$NEW)}
     </button>
   );
 }

frontend/src/components/features/feedback/feedback-form.tsx

@@ -1,5 +1,7 @@
 import React from "react";
 import hotToast from "react-hot-toast";
+import { useTranslation } from "react-i18next";
+import { I18nKey } from "#/i18n/declaration";
 import { Feedback } from "#/api/open-hands.types";
 import { useSubmitFeedback } from "#/hooks/mutation/use-submit-feedback";
 import { ModalButton } from "#/components/shared/buttons/modal-button";
@@ -13,8 +15,9 @@ interface FeedbackFormProps {
 }
 
 export function FeedbackForm({ onClose, polarity }: FeedbackFormProps) {
+  const { t } = useTranslation();
   const copiedToClipboardToast = () => {
-    hotToast("Password copied to clipboard", {
+    hotToast(t(I18nKey.FEEDBACK$PASSWORD_COPIED_MESSAGE), {
       icon: "📋",
       position: "bottom-right",
     });
@@ -41,10 +44,13 @@ export function FeedbackForm({ onClose, polarity }: FeedbackFormProps) {
           target="_blank"
           rel="noreferrer"
         >
-          Go to shared feedback
+          {t(I18nKey.FEEDBACK$GO_TO_FEEDBACK)}
         </a>
         <span onClick={() => onPressToast(password)} className="cursor-pointer">
-          Password: {password} <span className="text-gray-500">(copy)</span>
+          {t(I18nKey.FEEDBACK$PASSWORD)}: {password}{" "}
+          <span className="text-gray-500">
+            ({t(I18nKey.FEEDBACK$COPY_LABEL)})
+          </span>
         </span>
       </div>,
       { duration: 10000 },
@@ -86,12 +92,14 @@ export function FeedbackForm({ onClose, polarity }: FeedbackFormProps) {
   return (
     <form onSubmit={handleSubmit} className="flex flex-col gap-6 w-full">
       <label className="flex flex-col gap-2">
-        <span className="text-xs text-neutral-400">Email</span>
+        <span className="text-xs text-neutral-400">
+          {t(I18nKey.FEEDBACK$EMAIL_LABEL)}
+        </span>
         <input
           required
           name="email"
           type="email"
-          placeholder="Please enter your email"
+          placeholder={t(I18nKey.FEEDBACK$EMAIL_PLACEHOLDER)}
           className="bg-[#27272A] px-3 py-[10px] rounded"
         />
       </label>
@@ -104,11 +112,11 @@ export function FeedbackForm({ onClose, polarity }: FeedbackFormProps) {
             type="radio"
             defaultChecked
           />
-          Private
+          {t(I18nKey.FEEDBACK$PRIVATE_LABEL)}
         </label>
         <label className="flex gap-2 cursor-pointer">
           <input name="permissions" value="public" type="radio" />
-          Public
+          {t(I18nKey.FEEDBACK$PUBLIC_LABEL)}
         </label>
       </div>
 
@@ -116,12 +124,12 @@ export function FeedbackForm({ onClose, polarity }: FeedbackFormProps) {
         <ModalButton
           disabled={isPending}
           type="submit"
-          text="Submit"
+          text={t(I18nKey.FEEDBACK$CONTRIBUTE_LABEL)}
           className="bg-[#4465DB] grow"
         />
         <ModalButton
           disabled={isPending}
-          text="Cancel"
+          text={t(I18nKey.FEEDBACK$CANCEL_LABEL)}
           onClick={onClose}
           className="bg-[#737373] grow"
         />

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

@@ -1,4 +1,5 @@
 import React from "react";
+import { useTranslation } from "react-i18next";
 import {
   Autocomplete,
   AutocompleteItem,
@@ -6,6 +7,7 @@ import {
 } from "@nextui-org/react";
 import { useDispatch } from "react-redux";
 import posthog from "posthog-js";
+import { I18nKey } from "#/i18n/declaration";
 import { setSelectedRepository } from "#/state/initial-query-slice";
 import { useConfig } from "#/hooks/query/use-config";
 import { sanitizeQuery } from "#/utils/sanitize-query";
@@ -23,6 +25,7 @@ export function GitHubRepositorySelector({
   userRepositories,
   publicRepositories,
 }: GitHubRepositorySelectorProps) {
+  const { t } = useTranslation();
   const { data: config } = useConfig();
   const [selectedKey, setSelectedKey] = React.useState<string | null>(null);
 
@@ -49,14 +52,14 @@ export function GitHubRepositorySelector({
     dispatch(setSelectedRepository(null));
   };
 
-  const emptyContent = "No results found.";
+  const emptyContent = t(I18nKey.GITHUB$NO_RESULTS);
 
   return (
     <Autocomplete
       data-testid="github-repo-selector"
       name="repo"
       aria-label="GitHub Repository"
-      placeholder="Select a GitHub project"
+      placeholder={t(I18nKey.LANDING$SELECT_REPO)}
       isVirtualized={false}
       selectedKey={selectedKey}
       inputProps={{
@@ -86,12 +89,12 @@ export function GitHubRepositorySelector({
               rel="noreferrer noopener"
               onClick={(e) => e.stopPropagation()}
             >
-              Add more repositories...
+              {t(I18nKey.GITHUB$ADD_MORE_REPOS)}
             </a>
           </AutocompleteItem> // eslint-disable-next-line @typescript-eslint/no-explicit-any
         ) as any)}
       {userRepositories.length > 0 && (
-        <AutocompleteSection showDivider title="Your Repos">
+        <AutocompleteSection showDivider title={t(I18nKey.GITHUB$YOUR_REPOS)}>
           {userRepositories.map((repo) => (
             <AutocompleteItem
               data-testid="github-repo-item"
@@ -106,7 +109,7 @@ export function GitHubRepositorySelector({
         </AutocompleteSection>
       )}
       {publicRepositories.length > 0 && (
-        <AutocompleteSection showDivider title="Public Repos">
+        <AutocompleteSection showDivider title={t(I18nKey.GITHUB$PUBLIC_REPOS)}>
           {publicRepositories.map((repo) => (
             <AutocompleteItem
               data-testid="github-repo-item"

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

@@ -1,4 +1,6 @@
 import React from "react";
+import { useTranslation } from "react-i18next";
+import { I18nKey } from "#/i18n/declaration";
 import { SuggestionBox } from "#/components/features/suggestions/suggestion-box";
 import GitHubLogo from "#/assets/branding/github-logo.svg?react";
 import { GitHubRepositorySelector } from "./github-repo-selector";
@@ -23,6 +25,7 @@ export function GitHubRepositoriesSuggestionBox({
   gitHubAuthUrl,
   user,
 }: GitHubRepositoriesSuggestionBoxProps) {
+  const { t } = useTranslation();
   const [connectToGitHubModalOpen, setConnectToGitHubModalOpen] =
     React.useState(false);
   const [searchQuery, setSearchQuery] = React.useState<string>("");
@@ -53,18 +56,18 @@ export function GitHubRepositoriesSuggestionBox({
   return (
     <>
       <SuggestionBox
-        title="Open a Repo"
+        title={t(I18nKey.LANDING$OPEN_REPO)}
         content={
           isLoggedIn ? (
             <GitHubRepositorySelector
               onInputChange={setSearchQuery}
               onSelect={handleSubmit}
-              publicRepositories={searchedRepos}
+              publicRepositories={searchedRepos || []}
               userRepositories={repositories}
             />
           ) : (
             <ModalButton
-              text="Connect to GitHub"
+              text={t(I18nKey.GITHUB$CONNECT)}
               icon={<GitHubLogo width={20} height={20} />}
               className="bg-[#791B80] w-full"
               onClick={handleConnectToGitHub}

frontend/src/components/features/images/attach-image-label.tsx

@@ -1,10 +1,13 @@
+import { useTranslation } from "react-i18next";
+import { I18nKey } from "#/i18n/declaration";
 import Clip from "#/icons/clip.svg?react";
 
 export function AttachImageLabel() {
+  const { t } = useTranslation();
   return (
     <div className="flex self-start items-center text-[#A3A3A3] text-xs leading-[18px] -tracking-[0.08px] cursor-pointer">
       <Clip width={16} height={16} />
-      Attach images
+      {t(I18nKey.LANDING$ATTACH_IMAGES)}
     </div>
   );
 }

frontend/src/components/features/jupyter/jupyter-cell-output.tsx

@@ -1,6 +1,8 @@
 import Markdown from "react-markdown";
 import SyntaxHighlighter from "react-syntax-highlighter";
 import { atomOneDark } from "react-syntax-highlighter/dist/esm/styles/hljs";
+import { useTranslation } from "react-i18next";
+import { I18nKey } from "#/i18n/declaration";
 import { JupyterLine } from "#/utils/parse-cell-content";
 
 interface JupyterCellOutputProps {
@@ -8,9 +10,12 @@ interface JupyterCellOutputProps {
 }
 
 export function JupyterCellOutput({ lines }: JupyterCellOutputProps) {
+  const { t } = useTranslation();
   return (
     <div className="rounded-lg bg-gray-800 dark:bg-gray-900 p-2 text-xs">
-      <div className="mb-1 text-gray-400">STDOUT/STDERR</div>
+      <div className="mb-1 text-gray-400">
+        {t(I18nKey.JUPYTER$OUTPUT_LABEL)}
+      </div>
       <pre
         className="scrollbar-custom scrollbar-thumb-gray-500 hover:scrollbar-thumb-gray-400 dark:scrollbar-thumb-white/10 dark:hover:scrollbar-thumb-white/20 overflow-auto px-5 max-h-[60vh] bg-gray-800"
         style={{ padding: 0, marginBottom: 0, fontSize: "0.75rem" }}

frontend/src/components/features/sidebar/user-avatar.tsx

@@ -1,3 +1,5 @@
+import { useTranslation } from "react-i18next";
+import { I18nKey } from "#/i18n/declaration";
 import { LoadingSpinner } from "#/components/shared/loading-spinner";
 import DefaultUserAvatar from "#/icons/default-user.svg?react";
 import { cn } from "#/utils/utils";
@@ -11,11 +13,12 @@ interface UserAvatarProps {
 }
 
 export function UserAvatar({ onClick, avatarUrl, isLoading }: UserAvatarProps) {
+  const { t } = useTranslation();
   return (
     <TooltipButton
       testId="user-avatar"
-      tooltip="Account settings"
-      ariaLabel="Account settings"
+      tooltip={t(I18nKey.USER$ACCOUNT_SETTINGS)}
+      ariaLabel={t(I18nKey.USER$ACCOUNT_SETTINGS)}
       onClick={onClick}
       className={cn(
         "w-8 h-8 rounded-full flex items-center justify-center border-2 border-gray-200",
@@ -25,7 +28,7 @@ export function UserAvatar({ onClick, avatarUrl, isLoading }: UserAvatarProps) {
       {!isLoading && avatarUrl && <Avatar src={avatarUrl} />}
       {!isLoading && !avatarUrl && (
         <DefaultUserAvatar
-          aria-label="user avatar placeholder"
+          aria-label={t(I18nKey.USER$AVATAR_PLACEHOLDER)}
           width={20}
           height={20}
         />

frontend/src/components/features/suggestions/import-project-suggestion-box.tsx

@@ -1,3 +1,5 @@
+import { useTranslation } from "react-i18next";
+import { I18nKey } from "#/i18n/declaration";
 import { SuggestionBox } from "./suggestion-box";
 
 interface ImportProjectSuggestionBoxProps {
@@ -7,13 +9,14 @@ interface ImportProjectSuggestionBoxProps {
 export function ImportProjectSuggestionBox({
   onChange,
 }: ImportProjectSuggestionBoxProps) {
+  const { t } = useTranslation();
   return (
     <SuggestionBox
-      title="+ Import Project"
+      title={t(I18nKey.LANDING$IMPORT_PROJECT)}
       content={
         <label htmlFor="import-project" className="w-full flex justify-center">
           <span className="border-2 border-dashed border-neutral-600 rounded px-2 py-1 cursor-pointer">
-            Upload a .zip
+            {t(I18nKey.LANDING$UPLOAD_ZIP)}
           </span>
           <input
             hidden

frontend/src/components/features/suggestions/suggestion-bubble.tsx

@@ -1,8 +1,10 @@
+import { useTranslation } from "react-i18next";
 import { RefreshButton } from "#/components/shared/buttons/refresh-button";
 import Lightbulb from "#/icons/lightbulb.svg?react";
+import { I18nKey } from "#/i18n/declaration";
 
 interface SuggestionBubbleProps {
-  suggestion: string;
+  suggestion: { key: string; value: string };
   onClick: () => void;
   onRefresh: () => void;
 }
@@ -12,6 +14,7 @@ export function SuggestionBubble({
   onClick,
   onRefresh,
 }: SuggestionBubbleProps) {
+  const { t } = useTranslation();
   const handleRefresh = (e: React.MouseEvent<HTMLButtonElement>) => {
     e.stopPropagation();
     onRefresh();
@@ -24,7 +27,7 @@ export function SuggestionBubble({
     >
       <div className="flex items-center gap-2">
         <Lightbulb width={18} height={18} />
-        <span className="text-sm">{suggestion}</span>
+        <span className="text-sm">{t(suggestion.key as I18nKey)}</span>
       </div>
       <RefreshButton onClick={handleRefresh} />
     </div>

frontend/src/components/features/suggestions/suggestion-item.tsx

@@ -1,4 +1,7 @@
-export type Suggestion = { label: string; value: string };
+import { useTranslation } from "react-i18next";
+import { I18nKey } from "#/i18n/declaration";
+
+export type Suggestion = { label: I18nKey | string; value: string };
 
 interface SuggestionItemProps {
   suggestion: Suggestion;
@@ -6,6 +9,7 @@ interface SuggestionItemProps {
 }
 
 export function SuggestionItem({ suggestion, onClick }: SuggestionItemProps) {
+  const { t } = useTranslation();
   return (
     <li className="list-none border border-neutral-600 rounded-xl hover:bg-neutral-700 flex-1">
       <button
@@ -14,7 +18,7 @@ export function SuggestionItem({ suggestion, onClick }: SuggestionItemProps) {
         onClick={() => onClick(suggestion.value)}
         className="text-[16px] leading-6 -tracking-[0.01em] text-center w-full p-3 font-semibold"
       >
-        {suggestion.label}
+        {t(suggestion.label)}
       </button>
     </li>
   );

frontend/src/components/features/terminal/terminal-status-label.tsx

@@ -1,9 +1,12 @@
 import { useSelector } from "react-redux";
+import { useTranslation } from "react-i18next";
 import { cn } from "#/utils/utils";
 import { AgentState } from "#/types/agent-state";
 import { RootState } from "#/store";
+import { I18nKey } from "#/i18n/declaration";
 
 export function TerminalStatusLabel() {
+  const { t } = useTranslation();
   const { curAgentState } = useSelector((state: RootState) => state.agent);
 
   return (
@@ -17,7 +20,7 @@ export function TerminalStatusLabel() {
             : "bg-green-500",
         )}
       />
-      Terminal
+      {t(I18nKey.WORKSPACE$TERMINAL_TAB_LABEL)}
     </div>
   );
 }

frontend/src/components/features/waitlist/tos-checkbox.tsx

@@ -1,20 +1,24 @@
+import { useTranslation } from "react-i18next";
+import { I18nKey } from "#/i18n/declaration";
+
 interface TOSCheckboxProps {
   onChange: () => void;
 }
 
 export function TOSCheckbox({ onChange }: TOSCheckboxProps) {
+  const { t } = useTranslation();
   return (
     <label className="flex items-center gap-2">
       <input type="checkbox" onChange={onChange} />
       <span>
-        I accept the{" "}
+        {t(I18nKey.TOS$ACCEPT)}{" "}
         <a
           href="https://www.all-hands.dev/tos"
           target="_blank"
           rel="noopener noreferrer"
           className="underline underline-offset-2 text-blue-500 hover:text-blue-700"
         >
-          terms of service
+          {t(I18nKey.TOS$TERMS)}
         </a>
       </span>
     </label>

frontend/src/components/layout/beta-badge.tsx

@@ -1,7 +1,11 @@
+import { useTranslation } from "react-i18next";
+import { I18nKey } from "#/i18n/declaration";
+
 export function BetaBadge() {
+  const { t } = useTranslation();
   return (
     <span className="text-[11px] leading-5 text-root-primary bg-neutral-400 px-1 rounded-xl">
-      Beta
+      {t(I18nKey.BADGE$BETA)}
     </span>
   );
 }

frontend/src/components/layout/served-app-label.tsx

@@ -1,12 +1,15 @@
+import { useTranslation } from "react-i18next";
 import { useActiveHost } from "#/hooks/query/use-active-host";
+import { I18nKey } from "#/i18n/declaration";
 
 export function ServedAppLabel() {
+  const { t } = useTranslation();
   const { activeHost } = useActiveHost();
 
   return (
     <div className="flex items-center justify-between w-full">
       <div className="flex items-center gap-2">
-        <div className="flex items-center gap-2">App</div>
+        <div className="flex items-center gap-2">{t(I18nKey.APP$TITLE)}</div>
         <span className="border rounded-md text- px-1 font-bold">BETA</span>
       </div>
       {activeHost && <div className="w-2 h-2 bg-green-500 rounded-full" />}

frontend/src/components/shared/action-tooltip.tsx

@@ -22,7 +22,11 @@ export function ActionTooltip({ type, onClick }: ActionTooltipProps) {
       <button
         data-testid={`action-${type}-button`}
         type="button"
-        aria-label={type === "confirm" ? "Confirm action" : "Reject action"}
+        aria-label={
+          type === "confirm"
+            ? t(I18nKey.ACTION$CONFIRM)
+            : t(I18nKey.ACTION$REJECT)
+        }
         className="bg-neutral-700 rounded-full p-1 hover:bg-neutral-800"
         onClick={onClick}
       >

frontend/src/components/shared/buttons/docs-button.tsx

@@ -1,11 +1,14 @@
+import { useTranslation } from "react-i18next";
 import DocsIcon from "#/icons/docs.svg?react";
+import { I18nKey } from "#/i18n/declaration";
 import { TooltipButton } from "./tooltip-button";
 
 export function DocsButton() {
+  const { t } = useTranslation();
   return (
     <TooltipButton
-      tooltip="Documentation"
-      ariaLabel="Documentation"
+      tooltip={t(I18nKey.SIDEBAR$DOCS)}
+      ariaLabel={t(I18nKey.SIDEBAR$DOCS)}
       href="https://docs.all-hands.dev"
     >
       <DocsIcon width={28} height={28} />

frontend/src/components/shared/buttons/exit-project-button.tsx

@@ -1,3 +1,5 @@
+import { useTranslation } from "react-i18next";
+import { I18nKey } from "#/i18n/declaration";
 import NewProjectIcon from "#/icons/new-project.svg?react";
 import { TooltipButton } from "./tooltip-button";
 
@@ -6,10 +8,12 @@ interface ExitProjectButtonProps {
 }
 
 export function ExitProjectButton({ onClick }: ExitProjectButtonProps) {
+  const { t } = useTranslation();
+  const startNewProject = t(I18nKey.PROJECT$START_NEW);
   return (
     <TooltipButton
-      tooltip="Start new project"
-      ariaLabel="Start new project"
+      tooltip={startNewProject}
+      ariaLabel={startNewProject}
       onClick={onClick}
       testId="new-project-button"
     >

frontend/src/components/shared/buttons/open-vscode-button.tsx

@@ -1,3 +1,5 @@
+import { useTranslation } from "react-i18next";
+import { I18nKey } from "#/i18n/declaration";
 import { cn } from "#/utils/utils";
 import VSCodeIcon from "#/assets/vscode-alt.svg?react";
 
@@ -10,6 +12,9 @@ export function OpenVSCodeButton({
   isDisabled,
   onClick,
 }: OpenVSCodeButtonProps) {
+  const { t } = useTranslation();
+  const buttonText = t(I18nKey.VSCODE$OPEN);
+
   return (
     <button
       type="button"
@@ -21,10 +26,10 @@ export function OpenVSCodeButton({
           ? "bg-neutral-600 cursor-not-allowed"
           : "bg-[#4465DB] hover:bg-[#3451C7]",
       )}
-      aria-label="Open in VS Code"
+      aria-label={buttonText}
     >
       <VSCodeIcon width={20} height={20} />
-      Open in VS Code
+      {buttonText}
     </button>
   );
 }

frontend/src/components/shared/buttons/settings-button.tsx

@@ -1,16 +1,19 @@
 import { FaCog } from "react-icons/fa";
+import { useTranslation } from "react-i18next";
 import { TooltipButton } from "./tooltip-button";
+import { I18nKey } from "#/i18n/declaration";
 
 interface SettingsButtonProps {
   onClick: () => void;
 }
 
 export function SettingsButton({ onClick }: SettingsButtonProps) {
+  const { t } = useTranslation();
   return (
     <TooltipButton
       testId="settings-button"
-      tooltip="Settings"
-      ariaLabel="Settings"
+      tooltip={t(I18nKey.SETTINGS$TITLE)}
+      ariaLabel={t(I18nKey.SETTINGS$TITLE)}
       onClick={onClick}
     >
       <FaCog size={24} />

frontend/src/components/shared/buttons/stop-button.tsx

@@ -1,13 +1,17 @@
+import { useTranslation } from "react-i18next";
+import { I18nKey } from "#/i18n/declaration";
+
 interface StopButtonProps {
   isDisabled?: boolean;
   onClick?: () => void;
 }
 
 export function StopButton({ isDisabled, onClick }: StopButtonProps) {
+  const { t } = useTranslation();
   return (
     <button
       data-testid="stop-button"
-      aria-label="Stop"
+      aria-label={t(I18nKey.BUTTON$STOP)}
       disabled={isDisabled}
       onClick={onClick}
       type="button"

frontend/src/components/shared/buttons/submit-button.tsx

@@ -1,4 +1,6 @@
+import { useTranslation } from "react-i18next";
 import ArrowSendIcon from "#/icons/arrow-send.svg?react";
+import { I18nKey } from "#/i18n/declaration";
 
 interface SubmitButtonProps {
   isDisabled?: boolean;
@@ -6,9 +8,10 @@ interface SubmitButtonProps {
 }
 
 export function SubmitButton({ isDisabled, onClick }: SubmitButtonProps) {
+  const { t } = useTranslation();
   return (
     <button
-      aria-label="Send"
+      aria-label={t(I18nKey.BUTTON$SEND)}
       disabled={isDisabled}
       onClick={onClick}
       type="submit"

frontend/src/components/shared/download-progress.tsx

@@ -1,3 +1,6 @@
+import { useTranslation } from "react-i18next";
+import { I18nKey } from "#/i18n/declaration";
+
 export interface DownloadProgressState {
   filesTotal: number;
   filesDownloaded: number;
@@ -16,6 +19,7 @@ export function DownloadProgress({
   progress,
   onCancel,
 }: DownloadProgressProps) {
+  const { t } = useTranslation();
   const formatBytes = (bytes: number) => {
     const units = ["B", "KB", "MB", "GB"];
     let size = bytes;
@@ -33,12 +37,12 @@ export function DownloadProgress({
         <div className="mb-4">
           <h3 className="text-lg font-semibold mb-2 text-white">
             {progress.isDiscoveringFiles
-              ? "Preparing Download..."
-              : "Downloading Files"}
+              ? t(I18nKey.DOWNLOAD$PREPARING)
+              : t(I18nKey.DOWNLOAD$DOWNLOADING)}
           </h3>
           <p className="text-sm text-gray-400 truncate">
             {progress.isDiscoveringFiles
-              ? `Found ${progress.filesTotal} files...`
+              ? t(I18nKey.DOWNLOAD$FOUND_FILES, { count: progress.filesTotal })
               : progress.currentFile}
           </p>
         </div>
@@ -64,8 +68,11 @@ export function DownloadProgress({
         <div className="flex justify-between text-sm text-gray-400">
           <span>
             {progress.isDiscoveringFiles
-              ? `Scanning workspace...`
-              : `${progress.filesDownloaded} of ${progress.filesTotal} files`}
+              ? t(I18nKey.DOWNLOAD$SCANNING)
+              : t(I18nKey.DOWNLOAD$FILES_PROGRESS, {
+                  downloaded: progress.filesDownloaded,
+                  total: progress.filesTotal,
+                })}
           </span>
           {!progress.isDiscoveringFiles && (
             <span>{formatBytes(progress.bytesDownloadedPerSecond)}/s</span>
@@ -78,7 +85,7 @@ export function DownloadProgress({
             onClick={onCancel}
             className="px-4 py-2 text-sm text-gray-400 hover:text-white transition-colors"
           >
-            Cancel
+            {t(I18nKey.DOWNLOAD$CANCEL)}
           </button>
         </div>
       </div>

frontend/src/components/shared/hero-heading.tsx

@@ -1,24 +1,26 @@
+import { useTranslation } from "react-i18next";
 import BuildIt from "#/icons/build-it.svg?react";
+import { I18nKey } from "#/i18n/declaration";
 
 export function HeroHeading() {
+  const { t } = useTranslation();
   return (
     <div className="w-[304px] text-center flex flex-col gap-4 items-center py-4">
       <BuildIt width={88} height={104} />
       <h1 className="text-[38px] leading-[32px] -tracking-[0.02em]">
-        Let&apos;s Start Building!
+        {t(I18nKey.LANDING$TITLE)}
       </h1>
       <p className="mx-4 text-sm flex flex-col gap-2">
-        OpenHands makes it easy to build and maintain software using a simple
-        prompt.{" "}
+        {t(I18nKey.LANDING$SUBTITLE)}{" "}
         <span className="">
-          Not sure how to start?{" "}
+          {t(I18nKey.LANDING$START_HELP)}{" "}
           <a
             rel="noopener noreferrer"
             target="_blank"
             href="https://docs.all-hands.dev/modules/usage/getting-started"
             className="text-hyperlink underline underline-offset-[3px]"
           >
-            Read this
+            {t(I18nKey.LANDING$START_HELP_LINK)}
           </a>
         </span>
       </p>

frontend/src/components/shared/inputs/api-key-input.tsx

@@ -22,14 +22,14 @@ export function APIKeyInput({ isDisabled, isSet }: APIKeyInputProps) {
           {!isSet && (
             <FaExclamationCircle className="text-[#FF3860] inline-block" />
           )}
-          {t(I18nKey.SETTINGS_FORM$API_KEY_LABEL)}
+          {t(I18nKey.API$KEY)}
         </label>
       </Tooltip>
       <Input
         isDisabled={isDisabled}
         id="api-key"
         name="api-key"
-        aria-label="API Key"
+        aria-label={t(I18nKey.API$KEY)}
         type="password"
         defaultValue=""
         classNames={{
@@ -37,14 +37,14 @@ export function APIKeyInput({ isDisabled, isSet }: APIKeyInputProps) {
         }}
       />
       <p className="text-sm text-[#A3A3A3]">
-        {t(I18nKey.SETTINGS_FORM$DONT_KNOW_API_KEY_LABEL)}{" "}
+        {t(I18nKey.API$DONT_KNOW_KEY)}{" "}
         <a
           href="https://docs.all-hands.dev/modules/usage/llms"
           rel="noreferrer noopener"
           target="_blank"
           className="underline underline-offset-2"
         >
-          {t(I18nKey.SETTINGS_FORM$CLICK_HERE_FOR_INSTRUCTIONS_LABEL)}
+          {t(I18nKey.COMMON$CLICK_FOR_INSTRUCTIONS)}
         </a>
       </p>
     </fieldset>

frontend/src/components/shared/modals/account-settings/account-settings-form.tsx

@@ -64,7 +64,7 @@ export function AccountSettingsForm({
     <ModalBody testID="account-settings-form">
       <form className="flex flex-col w-full gap-6" onSubmit={handleSubmit}>
         <div className="w-full flex flex-col gap-2">
-          <BaseModalTitle title="Account Settings" />
+          <BaseModalTitle title={t(I18nKey.ACCOUNT_SETTINGS$TITLE)} />
 
           {config?.APP_MODE === "saas" && config?.APP_SLUG && (
             <a
@@ -73,12 +73,12 @@ export function AccountSettingsForm({
               rel="noreferrer noopener"
               className="underline"
             >
-              Configure Github Repositories
+              {t(I18nKey.GITHUB$CONFIGURE_REPOS)}
             </a>
           )}
           <FormFieldset
             id="language"
-            label="Language"
+            label={t(I18nKey.LANGUAGE$LABEL)}
             defaultSelectedKey={selectedLanguage}
             isClearable={false}
             items={AvailableLanguages.map(({ label, value: key }) => ({
@@ -91,32 +91,32 @@ export function AccountSettingsForm({
             <>
               <CustomInput
                 name="ghToken"
-                label="GitHub Token"
+                label={t(I18nKey.GITHUB$TOKEN_LABEL)}
                 type="password"
                 defaultValue={gitHubToken ?? ""}
               />
               <BaseModalDescription>
-                {t(I18nKey.CONNECT_TO_GITHUB_MODAL$GET_YOUR_TOKEN)}{" "}
+                {t(I18nKey.GITHUB$GET_TOKEN)}{" "}
                 <a
                   href="https://github.com/settings/tokens/new?description=openhands-app&scopes=repo,user,workflow"
                   target="_blank"
                   rel="noreferrer noopener"
                   className="text-[#791B80] underline"
                 >
-                  {t(I18nKey.CONNECT_TO_GITHUB_MODAL$HERE)}
+                  {t(I18nKey.COMMON$HERE)}
                 </a>
               </BaseModalDescription>
             </>
           )}
           {gitHubError && (
             <p className="text-danger text-xs">
-              {t(I18nKey.ACCOUNT_SETTINGS_MODAL$GITHUB_TOKEN_INVALID)}
+              {t(I18nKey.GITHUB$TOKEN_INVALID)}
             </p>
           )}
           {gitHubToken && !gitHubError && (
             <ModalButton
               variant="text-like"
-              text={t(I18nKey.ACCOUNT_SETTINGS_MODAL$DISCONNECT)}
+              text={t(I18nKey.BUTTON$DISCONNECT)}
               onClick={() => {
                 logout();
                 onClose();
@@ -132,7 +132,7 @@ export function AccountSettingsForm({
             type="checkbox"
             defaultChecked={analyticsConsent === "true"}
           />
-          Enable analytics
+          {t(I18nKey.ANALYTICS$ENABLE)}
         </label>
 
         <div className="flex flex-col gap-2 w-full">
@@ -140,11 +140,11 @@ export function AccountSettingsForm({
             testId="save-settings"
             type="submit"
             intent="account"
-            text={t(I18nKey.ACCOUNT_SETTINGS_MODAL$SAVE)}
+            text={t(I18nKey.BUTTON$SAVE)}
             className="bg-[#4465DB]"
           />
           <ModalButton
-            text={t(I18nKey.ACCOUNT_SETTINGS_MODAL$CLOSE)}
+            text={t(I18nKey.BUTTON$CLOSE)}
             onClick={onClose}
             className="bg-[#737373]"
           />

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

@@ -4,6 +4,8 @@ import {
   AutocompleteSection,
 } from "@nextui-org/react";
 import React from "react";
+import { useTranslation } from "react-i18next";
+import { I18nKey } from "#/i18n/declaration";
 import { mapProvider } from "#/utils/map-provider";
 import { VERIFIED_MODELS, VERIFIED_PROVIDERS } from "#/utils/verified-models";
 import { extractModelAndProvider } from "#/utils/extract-model-and-provider";
@@ -60,12 +62,14 @@ export function ModelSelector({
     setLitellmId(null);
   };
 
+  const { t } = useTranslation();
+
   return (
     <div data-testid="model-selector" className="flex flex-col gap-2">
       <div className="flex flex-row gap-3">
         <fieldset className="flex flex-col gap-2">
           <label htmlFor="agent" className="font-[500] text-[#A3A3A3] text-xs">
-            LLM Provider
+            {t(I18nKey.LLM$PROVIDER)}
           </label>
           <Autocomplete
             data-testid="llm-provider"
@@ -73,8 +77,8 @@ export function ModelSelector({
             isVirtualized={false}
             name="llm-provider"
             isDisabled={isDisabled}
-            aria-label="LLM Provider"
-            placeholder="Select a provider"
+            aria-label={t(I18nKey.LLM$PROVIDER)}
+            placeholder={t(I18nKey.LLM$SELECT_PROVIDER_PLACEHOLDER)}
             isClearable={false}
             onSelectionChange={(e) => {
               if (e?.toString()) handleChangeProvider(e.toString());
@@ -115,15 +119,15 @@ export function ModelSelector({
 
         <fieldset className="flex flex-col gap-2">
           <label htmlFor="agent" className="font-[500] text-[#A3A3A3] text-xs">
-            LLM Model
+            {t(I18nKey.LLM$MODEL)}
           </label>
           <Autocomplete
             data-testid="llm-model"
             isRequired
             isVirtualized={false}
             name="llm-model"
-            aria-label="LLM Model"
-            placeholder="Select a model"
+            aria-label={t(I18nKey.LLM$MODEL)}
+            placeholder={t(I18nKey.LLM$SELECT_MODEL_PLACEHOLDER)}
             isClearable={false}
             onSelectionChange={(e) => {
               if (e?.toString()) handleChangeModel(e.toString());

frontend/src/components/shared/modals/settings/runtime-size-selector.tsx

@@ -1,5 +1,6 @@
 import { useTranslation } from "react-i18next";
 import { Select, SelectItem } from "@nextui-org/react";
+import { I18nKey } from "#/i18n/declaration";
 
 interface RuntimeSizeSelectorProps {
   isDisabled: boolean;
@@ -18,15 +19,18 @@ export function RuntimeSizeSelector({
         htmlFor="runtime-size"
         className="font-[500] text-[#A3A3A3] text-xs"
       >
-        {t("SETTINGS_FORM$RUNTIME_SIZE_LABEL")}
+        {t(I18nKey.SETTINGS_FORM$RUNTIME_SIZE_LABEL)}
       </label>
       <Select
         data-testid="runtime-size"
         id="runtime-size"
         name="runtime-size"
         defaultSelectedKeys={[String(defaultValue || 1)]}
+        selectedKeys={[String(defaultValue || 1)]}
         isDisabled={isDisabled}
-        aria-label={t("SETTINGS_FORM$RUNTIME_SIZE_LABEL")}
+        selectionMode="single"
+        disallowEmptySelection
+        aria-label={t(I18nKey.SETTINGS_FORM$RUNTIME_SIZE_LABEL)}
         classNames={{
           trigger: "bg-[#27272A] rounded-md text-sm px-3 py-[10px]",
         }}

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

@@ -2,11 +2,11 @@ import { useLocation } from "react-router";
 import { useTranslation } from "react-i18next";
 import React from "react";
 import posthog from "posthog-js";
+import { I18nKey } from "#/i18n/declaration";
 import { organizeModelsAndProviders } from "#/utils/organize-models-and-providers";
 import { getDefaultSettings, Settings } from "#/services/settings";
 import { extractModelAndProvider } from "#/utils/extract-model-and-provider";
 import { DangerModal } from "../confirmation-modals/danger-modal";
-import { I18nKey } from "#/i18n/declaration";
 import { extractSettings, saveSettingsView } from "#/utils/settings-utils";
 import { useEndSession } from "#/hooks/use-end-session";
 import { ModalButton } from "../../buttons/modal-button";
@@ -209,18 +209,18 @@ export function SettingsForm({
               testId="save-settings-button"
               disabled={disabled}
               type="submit"
-              text={t(I18nKey.SETTINGS_FORM$SAVE_LABEL)}
+              text={t(I18nKey.BUTTON$SAVE)}
               className="bg-[#4465DB] w-full"
             />
             <ModalButton
-              text={t(I18nKey.SETTINGS_FORM$CLOSE_LABEL)}
+              text={t(I18nKey.BUTTON$CLOSE)}
               className="bg-[#737373] w-full"
               onClick={onClose}
             />
           </div>
           <ModalButton
             disabled={disabled}
-            text={t(I18nKey.SETTINGS_FORM$RESET_TO_DEFAULTS_LABEL)}
+            text={t(I18nKey.BUTTON$RESET_TO_DEFAULTS)}
             variant="text-like"
             className="text-danger self-start"
             onClick={() => {
@@ -234,17 +234,15 @@ export function SettingsForm({
         <ModalBackdrop>
           <DangerModal
             testId="reset-defaults-modal"
-            title={t(I18nKey.SETTINGS_FORM$ARE_YOU_SURE_LABEL)}
-            description={t(
-              I18nKey.SETTINGS_FORM$ALL_INFORMATION_WILL_BE_DELETED_MESSAGE,
-            )}
+            title={t(I18nKey.MODAL$CONFIRM_RESET_TITLE)}
+            description={t(I18nKey.MODAL$CONFIRM_RESET_MESSAGE)}
             buttons={{
               danger: {
-                text: t(I18nKey.SETTINGS_FORM$RESET_TO_DEFAULTS_LABEL),
+                text: t(I18nKey.BUTTON$RESET_TO_DEFAULTS),
                 onClick: handleConfirmResetSettings,
               },
               cancel: {
-                text: t(I18nKey.SETTINGS_FORM$CANCEL_LABEL),
+                text: t(I18nKey.BUTTON$CANCEL),
                 onClick: () => setConfirmResetDefaultsModalOpen(false),
               },
             }}
@@ -254,17 +252,15 @@ export function SettingsForm({
       {confirmEndSessionModalOpen && (
         <ModalBackdrop>
           <DangerModal
-            title={t(I18nKey.SETTINGS_FORM$END_SESSION_LABEL)}
-            description={t(
-              I18nKey.SETTINGS_FORM$CHANGING_WORKSPACE_WARNING_MESSAGE,
-            )}
+            title={t(I18nKey.MODAL$END_SESSION_TITLE)}
+            description={t(I18nKey.MODAL$END_SESSION_MESSAGE)}
             buttons={{
               danger: {
-                text: t(I18nKey.SETTINGS_FORM$END_SESSION_LABEL),
+                text: t(I18nKey.BUTTON$END_SESSION),
                 onClick: handleConfirmEndSession,
               },
               cancel: {
-                text: t(I18nKey.SETTINGS_FORM$CANCEL_LABEL),
+                text: t(I18nKey.BUTTON$CANCEL),
                 onClick: () => setConfirmEndSessionModalOpen(false),
               },
             }}

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

@@ -1,5 +1,7 @@
+import { useTranslation } from "react-i18next";
 import { useAIConfigOptions } from "#/hooks/query/use-ai-config-options";
 import { Settings } from "#/services/settings";
+import { I18nKey } from "#/i18n/declaration";
 import { LoadingSpinner } from "../../loading-spinner";
 import { ModalBackdrop } from "../modal-backdrop";
 import { SettingsForm } from "./settings-form";
@@ -11,6 +13,7 @@ interface SettingsModalProps {
 
 export function SettingsModal({ onClose, settings }: SettingsModalProps) {
   const aiConfigOptions = useAIConfigOptions();
+  const { t } = useTranslation();
 
   return (
     <ModalBackdrop onClose={onClose}>
@@ -22,14 +25,12 @@ export function SettingsModal({ onClose, settings }: SettingsModalProps) {
           <p className="text-danger text-xs">{aiConfigOptions.error.message}</p>
         )}
         <span className="text-xl leading-6 font-semibold -tracking-[0.01em">
-          AI Provider Configuration
+          {t(I18nKey.AI_SETTINGS$TITLE)}
         </span>
         <p className="text-xs text-[#A3A3A3]">
-          To continue, connect an OpenAI, Anthropic, or other LLM account
-        </p>
-        <p className="text-xs text-danger">
-          Changing settings during an active session will end the session
+          {t(I18nKey.SETTINGS$DESCRIPTION)}
         </p>
+        <p className="text-xs text-danger">{t(I18nKey.SETTINGS$WARNING)}</p>
         {aiConfigOptions.isLoading && (
           <div className="flex justify-center">
             <LoadingSpinner size="small" />

frontend/src/components/shared/task-form.tsx

@@ -23,14 +23,13 @@ export function TaskForm({ ref }: TaskFormProps) {
   const dispatch = useDispatch();
   const navigation = useNavigation();
 
-  const { selectedRepository, files } = useSelector(
-    (state: RootState) => state.initialQuery,
-  );
+  const { files } = useSelector((state: RootState) => state.initialQuery);
 
   const [text, setText] = React.useState("");
-  const [suggestion, setSuggestion] = React.useState(
-    getRandomKey(SUGGESTIONS["non-repo"]),
-  );
+  const [suggestion, setSuggestion] = React.useState(() => {
+    const key = getRandomKey(SUGGESTIONS["non-repo"]);
+    return { key, value: SUGGESTIONS["non-repo"][key] };
+  });
   const [inputIsFocused, setInputIsFocused] = React.useState(false);
   const { mutate: createConversation, isPending } = useCreateConversation();
 
@@ -38,26 +37,16 @@ export function TaskForm({ ref }: TaskFormProps) {
     const suggestions = SUGGESTIONS["non-repo"];
     // remove current suggestion to avoid refreshing to the same suggestion
     const suggestionCopy = { ...suggestions };
-    delete suggestionCopy[suggestion];
+    delete suggestionCopy[suggestion.key];
 
     const key = getRandomKey(suggestionCopy);
-    setSuggestion(key);
+    setSuggestion({ key, value: suggestions[key] });
   };
 
   const onClickSuggestion = () => {
-    const suggestions = SUGGESTIONS["non-repo"];
-    const value = suggestions[suggestion];
-    setText(value);
+    setText(suggestion.value);
   };
 
-  const placeholder = React.useMemo(() => {
-    if (selectedRepository) {
-      return `What would you like to change in ${selectedRepository}?`;
-    }
-
-    return "What do you want to build?";
-  }, [selectedRepository]);
-
   const handleSubmit = (event: React.FormEvent<HTMLFormElement>) => {
     event.preventDefault();
     const formData = new FormData(event.currentTarget);
@@ -105,7 +94,6 @@ export function TaskForm({ ref }: TaskFormProps) {
                   dispatch(addFile(base64));
                 });
               }}
-              placeholder={placeholder}
               value={text}
               maxRows={15}
               showButton={!!text}

frontend/src/context/auth-context.tsx

@@ -87,7 +87,12 @@ function AuthProvider({ children }: React.PropsWithChildren) {
 
     setGitHubToken(storedGitHubToken);
     setUserId(userId);
-    setupGithubAxiosInterceptors(refreshToken, logout);
+    const setupIntercepter = async () => {
+      const config = await OpenHands.getConfig();
+      setupGithubAxiosInterceptors(config.APP_MODE, refreshToken, logout);
+    };
+
+    setupIntercepter();
   }, []);
 
   const value = React.useMemo(

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

@@ -1,7 +1,7 @@
 import { useQuery } from "@tanstack/react-query";
 import { useAuth } from "#/context/auth-context";
 import { useConfig } from "./use-config";
-import { retrieveGitHubAppInstallations } from "#/api/github";
+import OpenHands from "#/api/open-hands";
 
 export const useAppInstallations = () => {
   const { data: config } = useConfig();
@@ -9,10 +9,7 @@ export const useAppInstallations = () => {
 
   return useQuery({
     queryKey: ["installations", gitHubToken, config?.GITHUB_CLIENT_ID],
-    queryFn: async () => {
-      const data = await retrieveGitHubAppInstallations();
-      return data;
-    },
+    queryFn: OpenHands.getGitHubUserInstallationIds,
     enabled:
       !!gitHubToken &&
       !!config?.GITHUB_CLIENT_ID &&

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

@@ -1,9 +1,9 @@
 import { useQuery } from "@tanstack/react-query";
 import React from "react";
 import posthog from "posthog-js";
-import { retrieveGitHubUser } from "#/api/github";
 import { useAuth } from "#/context/auth-context";
 import { useConfig } from "./use-config";
+import OpenHands from "#/api/open-hands";
 
 export const useGitHubUser = () => {
   const { gitHubToken, setUserId } = useAuth();
@@ -11,7 +11,7 @@ export const useGitHubUser = () => {
 
   const user = useQuery({
     queryKey: ["user", gitHubToken],
-    queryFn: retrieveGitHubUser,
+    queryFn: OpenHands.getGitHubUser,
     enabled: !!gitHubToken && !!config?.APP_MODE,
     retry: false,
   });

frontend/src/hooks/query/use-latest-repo-commit.ts

@@ -1,17 +0,0 @@
-import { useQuery } from "@tanstack/react-query";
-import { retrieveLatestGitHubCommit } from "#/api/github";
-import { useAuth } from "#/context/auth-context";
-
-interface UseLatestRepoCommitConfig {
-  repository: string | null;
-}
-
-export const useLatestRepoCommit = (config: UseLatestRepoCommitConfig) => {
-  const { gitHubToken } = useAuth();
-
-  return useQuery({
-    queryKey: ["latest_commit", gitHubToken, config.repository],
-    queryFn: () => retrieveLatestGitHubCommit(config.repository!),
-    enabled: !!gitHubToken && !!config.repository,
-  });
-};

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

@@ -1,12 +1,11 @@
 import { useQuery } from "@tanstack/react-query";
-import { searchPublicRepositories } from "#/api/github";
+import OpenHands from "#/api/open-hands";
 
 export function useSearchRepositories(query: string) {
   return useQuery({
     queryKey: ["repositories", query],
-    queryFn: () => searchPublicRepositories(query, 3),
+    queryFn: () => OpenHands.searchGitHubRepositories(query, 3),
     enabled: !!query,
     select: (data) => data.map((repo) => ({ ...repo, is_public: true })),
-    initialData: [],
   });
 }

frontend/src/i18n/index.ts

@@ -5,6 +5,7 @@ import { initReactI18next } from "react-i18next";
 
 export const AvailableLanguages = [
   { label: "English", value: "en" },
+  { label: "日本語", value: "ja" },
   { label: "简体中文", value: "zh-CN" },
   { label: "繁體中文", value: "zh-TW" },
   { label: "한국어", value: "ko-KR" },

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

@@ -1,6 +1,8 @@
 import React from "react";
 import { useDispatch } from "react-redux";
+import { useTranslation } from "react-i18next";
 import posthog from "posthog-js";
+import { I18nKey } from "#/i18n/declaration";
 import { setImportedProjectZip } from "#/state/initial-query-slice";
 import { convertZipToBase64 } from "#/utils/convert-zip-to-base64";
 import { useGitHubUser } from "#/hooks/query/use-github-user";
@@ -13,6 +15,7 @@ import { HeroHeading } from "#/components/shared/hero-heading";
 import { TaskForm } from "#/components/shared/task-form";
 
 function Home() {
+  const { t } = useTranslation();
   const { gitHubToken } = useAuth();
   const dispatch = useDispatch();
   const formRef = React.useRef<HTMLFormElement>(null);
@@ -59,12 +62,12 @@ function Home() {
       {latestConversation && (
         <div className="flex gap-4 w-full text-center mt-8">
           <p className="text-center w-full">
-            Or&nbsp;
+            {t(I18nKey.LANDING$OR)}&nbsp;
             <a
               className="underline"
               href={`/conversations/${latestConversation}`}
             >
-              jump back to your most recent conversation
+              {t(I18nKey.LANDING$RECENT_CONVERSATION)}
             </a>
           </p>
         </div>

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

@@ -1,6 +1,5 @@
 import React from "react";
 import { useDispatch, useSelector } from "react-redux";
-import { useAuth } from "#/context/auth-context";
 import {
   useWsClient,
   WsClientProviderStatus,
@@ -14,17 +13,12 @@ import { AgentState } from "#/types/agent-state";
 
 export const useWSStatusChange = () => {
   const { send, status } = useWsClient();
-  const { gitHubToken } = useAuth();
   const { curAgentState } = useSelector((state: RootState) => state.agent);
   const dispatch = useDispatch();
 
   const statusRef = React.useRef<WsClientProviderStatus | null>(null);
 
-  const { selectedRepository } = useSelector(
-    (state: RootState) => state.initialQuery,
-  );
-
-  const { files, importedProjectZip, initialQuery } = useSelector(
+  const { files, initialQuery } = useSelector(
     (state: RootState) => state.initialQuery,
   );
 
@@ -33,30 +27,15 @@ export const useWSStatusChange = () => {
     send(createChatMessage(query, base64Files, timestamp));
   };
 
-  const dispatchInitialQuery = (query: string, additionalInfo: string) => {
-    if (additionalInfo) {
-      sendInitialQuery(`${query}\n\n[${additionalInfo}]`, files);
-    } else {
-      sendInitialQuery(query, files);
-    }
-
+  const dispatchInitialQuery = (query: string) => {
+    sendInitialQuery(query, files);
     dispatch(clearFiles()); // reset selected files
     dispatch(clearInitialQuery()); // reset initial query
   };
 
   const handleAgentInit = () => {
-    let additionalInfo = "";
-
-    if (gitHubToken && selectedRepository) {
-      additionalInfo = `Repository ${selectedRepository} has been cloned to /workspace. Please check the /workspace for files.`;
-    } else if (importedProjectZip) {
-      // if there's an uploaded project zip, add it to the chat
-      additionalInfo =
-        "Files have been uploaded. Please check the /workspace for files.";
-    }
-
     if (initialQuery) {
-      dispatchInitialQuery(initialQuery, additionalInfo);
+      dispatchInitialQuery(initialQuery);
     }
   };
   React.useEffect(() => {

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

@@ -4,6 +4,8 @@ import { Outlet } from "react-router";
 import { useDispatch, useSelector } from "react-redux";
 import { FaServer } from "react-icons/fa";
 import toast from "react-hot-toast";
+import { useTranslation } from "react-i18next";
+import { I18nKey } from "#/i18n/declaration";
 import {
   ConversationProvider,
   useConversation,
@@ -39,6 +41,7 @@ import { MULTI_CONVERSATION_UI } from "#/utils/feature-flags";
 
 function AppContent() {
   useConversationConfig();
+  const { t } = useTranslation();
   const { gitHubToken } = useAuth();
   const { data: settings } = useSettings();
   const { conversationId } = useConversation();
@@ -126,7 +129,11 @@ function AppContent() {
               <Container
                 className="h-full"
                 labels={[
-                  { label: "Workspace", to: "", icon: <CodeIcon /> },
+                  {
+                    label: t(I18nKey.WORKSPACE$TITLE),
+                    to: "",
+                    icon: <CodeIcon />,
+                  },
                   { label: "Jupyter", to: "jupyter", icon: <ListIcon /> },
                   {
                     label: <ServedAppLabel />,
@@ -136,7 +143,7 @@ function AppContent() {
                   {
                     label: (
                       <div className="flex items-center gap-1">
-                        Browser
+                        {t(I18nKey.BROWSER$TITLE)}
                         {updateCount > 0 && <CountBadge count={updateCount} />}
                       </div>
                     ),

frontend/src/utils/suggestions/non-repo-suggestions.ts

@@ -1,22 +1,18 @@
-const KEY_1 = "Build an app to view pull requests";
-const VALUE_1 = `I want to create a React app to view all of the open pull
-requests that exist on all of my team's github repos. Here
-are some details:
+import { I18nKey } from "#/i18n/declaration";
 
-1. Please initialize the app using vite and react-ts.
-2. You can test the app on the https://github.com/OpenDevin/
-   github org
-3. I have provided a github token in the environment (the
-   variable name is $GITHUB_TOKEN)
-4. It should have a dropdown that allows me to select a
-   single repo within the org.
-5. There should be tests written using vitest.
+const KEY_1 = I18nKey.SUGGESTIONS$HACKER_NEWS;
+const VALUE_1 = `Please write a bash script which displays the top story on Hacker News. It should show the title, the link, and the number of points.
+The script should only use tools that are widely available on unix systems, like curl and grep.`;
 
-When things are working, initialize a github repo, create
-a .gitignore file, and commit the changes.`;
+const KEY_2 = I18nKey.SUGGESTIONS$HELLO_WORLD;
+const VALUE_2 = `I want to create a Hello World app in Javascript that:
+* Displays Hello World in the middle.
+* Has a button that when clicked, changes the greeting with a bouncing animation to fun versions of Hello.
+* Has a counter for how many times the button has been clicked.
+* Has another button that changes the app's background color.`;
 
-const KEY_2 = "Build a todo list application";
-const VALUE_2 = `I want to create a VueJS app that allows me to:
+const KEY_3 = I18nKey.SUGGESTIONS$TODO_APP;
+const VALUE_3 = `I want to create a VueJS app that allows me to:
 * See all the items on my todo list
 * add a new item to the list
 * mark an item as done
@@ -24,14 +20,7 @@ const VALUE_2 = `I want to create a VueJS app that allows me to:
 * change the text of an item
 * set a due date on the item
 
-This should be a client-only app with no backend. The list should persist in localStorage.
-
-Please add tests for all of the above and make sure they pass`;
-
-const KEY_3 = "Write a bash script that shows the top story on Hacker News";
-const VALUE_3 = `Please write a bash script which displays the top story on Hacker News. It should show the title, the link, and the number of points.
-
-The script should only use tools that are widely available on unix systems, like curl and grep.`;
+This should be a client-only app with no backend. The list should persist in localStorage.`;
 
 export const NON_REPO_SUGGESTIONS: Record<string, string> = {
   [KEY_1]: VALUE_1,

frontend/src/utils/suggestions/repo-suggestions.ts

@@ -1,4 +1,4 @@
-const KEY_1 = "Increase my test coverage";
+const KEY_1 = "INCREASE_TEST_COVERAGE";
 const VALUE_1 = `I want to increase the test coverage of the repository in the current directory.
 
 Please investigate the repo to figure out what language is being used, and where tests are located, if there are any.
@@ -9,10 +9,10 @@ If there are existing tests, find a function or method which lacks adequate unit
 
 Make sure the tests pass before you finish.`;
 
-const KEY_2 = "Auto-merge Dependabot PRs";
+const KEY_2 = "AUTO_MERGE_PRS";
 const VALUE_2 = `Please add a GitHub action to this repository which automatically merges pull requests from Dependabot so long as the tests are passing.`;
 
-const KEY_3 = "Fix up my README";
+const KEY_3 = "FIX_README";
 const VALUE_3 = `Please look at the README and make the following improvements, if they make sense:
 * correct any typos that you find
 * add missing language annotations on codeblocks
@@ -22,7 +22,7 @@ const VALUE_3 = `Please look at the README and make the following improvements,
 
 If there are no obvious ways to improve the README, make at least one small change to make the wording clearer or friendlier`;
 
-const KEY_4 = "Clean up my dependencies";
+const KEY_4 = "CLEAN_DEPENDENCIES";
 const VALUE_4 = `Examine the dependencies of the current codebase. Make sure you can run the code and any tests.
 
 Then run any commands necessary to update all dependencies to the latest versions, and make sure the code continues to run correctly and the tests pass. If changes need to be made to the codebase, go ahead and make those changes. You can look up documentation for new versions using the browser if you need to.
@@ -31,10 +31,10 @@ If a particular dependency update is causing trouble (e.g. breaking changes that
 
 Additionally, if you're able to prune any dependencies that are obviously unused, please do so. You may use third party tools to check for unused dependencies.`;
 
-const KEY_5 = "Add best practices docs for contributors";
+const KEY_5 = "ADD_DOCS";
 const VALUE_5 = `Investigate the documentation in the root of the current repo. Please add a CODE_OF_CONDUCT.md and CONTRIBUTORS.md with good defaults if they are not present. Use information in the README to inform the CONTRIBUTORS doc. If there is no LICENSE currently in the repo, please add the Apache 2.0 license. Add links to all these documents into the README`;
 
-const KEY_6 = "Add/improve a Dockerfile";
+const KEY_6 = "ADD_DOCKERFILE";
 const VALUE_6 = `Investigate the current repo to understand the installation instructions. Then create a Dockerfile that runs the application, using best practices like arguments and multi-stage builds wherever appropriate.
 
 If there is an existing Dockerfile, and there are ways to improve it according to best practices, do so.`;

openhands/agenthub/codeact_agent/codeact_agent.py

@@ -107,12 +107,7 @@ class CodeActAgent(Agent):
             f'TOOLS loaded for CodeActAgent: {json.dumps(self.tools, indent=2, ensure_ascii=False).replace("\\n", "\n")}'
         )
         self.prompt_manager = PromptManager(
-            microagent_dir=os.path.join(
-                os.path.dirname(os.path.dirname(openhands.__file__)),
-                'microagents',
-            )
-            if self.config.use_microagents
-            else None,
+            microagent_dir=None,  # Will be set in step() when we have access to the runtime
             prompt_dir=os.path.join(os.path.dirname(__file__), 'prompts'),
             disabled_microagents=self.config.disabled_microagents,
         )
@@ -369,6 +364,14 @@ class CodeActAgent(Agent):
         - MessageAction(content) - Message action to run (e.g. ask for clarification)
         - AgentFinishAction() - end the interaction
         """
+        # Initialize the prompt_manager with microagents from the runtime
+        if self.config.use_microagents and 'runtime' in state.inputs:
+            # Load microagents from the runtime
+            runtime = state.inputs['runtime']
+            microagents = runtime.get_microagents_from_selected_repo(None)  # None means current workspace
+            
+            # Load the microagents into the prompt manager
+            self.prompt_manager.load_microagents(microagents)
         # Continue with pending actions if any
         if self.pending_actions:
             return self.pending_actions.popleft()

openhands/agenthub/codeact_agent/prompts/system_prompt.j2

@@ -5,9 +5,14 @@ You are OpenHands agent, a helpful AI assistant that can interact with a compute
 * The assistant MUST NOT include comments in the code unless they are necessary to describe non-obvious behavior.
 {{ runtime_info }}
 </IMPORTANT>
-{% if repo_instructions -%}
+{% if repository_info %}
+<REPOSITORY_INFO>
+At the user's request, repository {{ repository_info.repo_name }} has been cloned to directory {{ repository_info.repo_directory }}.
+</REPOSITORY_INFO>
+{% endif %}
+{% if repository_instructions -%}
 <REPOSITORY_INSTRUCTIONS>
-{{ repo_instructions }}
+{{ repository_instructions }}
 </REPOSITORY_INSTRUCTIONS>
 {% endif %}
 {% if runtime_info and runtime_info.available_hosts -%}

openhands/agenthub/micro/agent.py

@@ -50,6 +50,10 @@ class MicroAgent(Agent):
         # history is in reverse order, let's fix it
         processed_history.reverse()
 
+        # everything starts with a message
+        # the first message is already in the prompt as the task
+        # TODO: so we don't need to include it in the history
+
         return json.dumps(processed_history, **kwargs)
 
     def __init__(self, llm: LLM, config: AgentConfig):

openhands/controller/agent_controller.py

@@ -112,12 +112,16 @@ class AgentController:
         self.id = sid
         self.agent = agent
         self.headless_mode = headless_mode
+        self.is_delegate = is_delegate
 
-        # subscribe to the event stream
+        # the event stream must be set before maybe subscribing to it
         self.event_stream = event_stream
-        self.event_stream.subscribe(
-            EventStreamSubscriber.AGENT_CONTROLLER, self.on_event, self.id
-        )
+
+        # subscribe to the event stream if this is not a delegate
+        if not self.is_delegate:
+            self.event_stream.subscribe(
+                EventStreamSubscriber.AGENT_CONTROLLER, self.on_event, self.id
+            )
 
         # state from the previous session, state from a parent agent, or a fresh state
         self.set_initial_state(
@@ -165,7 +169,11 @@ class AgentController:
         )
 
         # unsubscribe from the event stream
-        self.event_stream.unsubscribe(EventStreamSubscriber.AGENT_CONTROLLER, self.id)
+        # only the root parent controller subscribes to the event stream
+        if not self.is_delegate:
+            self.event_stream.unsubscribe(
+                EventStreamSubscriber.AGENT_CONTROLLER, self.id
+            )
         self._closed = True
 
     def log(self, level: str, message: str, extra: dict | None = None) -> None:
@@ -226,9 +234,21 @@ class AgentController:
             await self._react_to_exception(reported)
 
     def should_step(self, event: Event) -> bool:
+        # it might be the delegate's day in the sun
+        if self.delegate is not None:
+            return False
+
         if isinstance(event, Action):
             if isinstance(event, MessageAction) and event.source == EventSource.USER:
                 return True
+            if (
+                isinstance(event, MessageAction)
+                and self.get_agent_state() != AgentState.AWAITING_USER_INPUT
+            ):
+                # TODO: this is fragile, but how else to check if eligible?
+                return True
+            if isinstance(event, AgentDelegateAction):
+                return True
             return False
         if isinstance(event, Observation):
             if isinstance(event, NullObservation) or isinstance(
@@ -244,12 +264,35 @@ class AgentController:
         Args:
             event (Event): The incoming event to process.
         """
+
+        # If we have a delegate that is not finished or errored, forward events to it
+        if self.delegate is not None:
+            delegate_state = self.delegate.get_agent_state()
+            if delegate_state not in (
+                AgentState.FINISHED,
+                AgentState.ERROR,
+                AgentState.REJECTED,
+            ):
+                # Forward the event to delegate and skip parent processing
+                asyncio.get_event_loop().run_until_complete(
+                    self.delegate._on_event(event)
+                )
+                return
+            else:
+                # delegate is done or errored, so end it
+                self.end_delegate()
+                return
+
+        # continue parent processing only if there's no active delegate
         asyncio.get_event_loop().run_until_complete(self._on_event(event))
 
     async def _on_event(self, event: Event) -> None:
         if hasattr(event, 'hidden') and event.hidden:
             return
 
+        # Give others a little chance
+        await asyncio.sleep(0.01)
+
         # if the event is not filtered out, add it to the history
         if not any(isinstance(event, filter_type) for filter_type in self.filter_out):
             self.state.history.append(event)
@@ -263,17 +306,22 @@ class AgentController:
             self.step()
 
     async def _handle_action(self, action: Action) -> None:
-        """Handles actions from the event stream.
-
-        Args:
-            action (Action): The action to handle.
-        """
+        """Handles an Action from the agent or delegate."""
         if isinstance(action, ChangeAgentStateAction):
             await self.set_agent_state_to(action.agent_state)  # type: ignore
         elif isinstance(action, MessageAction):
             await self._handle_message_action(action)
         elif isinstance(action, AgentDelegateAction):
             await self.start_delegate(action)
+            assert self.delegate is not None
+            # Post a MessageAction with the task for the delegate
+            if 'task' in action.inputs:
+                self.event_stream.add_event(
+                    MessageAction(content='TASK: ' + action.inputs['task']),
+                    EventSource.USER,
+                )
+                await self.delegate.set_agent_state_to(AgentState.RUNNING)
+            return
 
         elif isinstance(action, AgentFinishAction):
             self.state.outputs = action.outputs
@@ -491,7 +539,7 @@ class AgentController:
             f'start delegate, creating agent {delegate_agent.name} using LLM {llm}',
         )
 
-        self.event_stream.unsubscribe(EventStreamSubscriber.AGENT_CONTROLLER, self.id)
+        # Create the delegate with is_delegate=True so it does NOT subscribe directly
         self.delegate = AgentController(
             sid=self.id + '-delegate',
             agent=delegate_agent,
@@ -504,7 +552,57 @@ class AgentController:
             is_delegate=True,
             headless_mode=self.headless_mode,
         )
-        await self.delegate.set_agent_state_to(AgentState.RUNNING)
+
+    def end_delegate(self) -> None:
+        """Ends the currently active delegate (e.g., if it is finished or errored)
+        so that this controller can resume normal operation.
+        """
+        if self.delegate is None:
+            return
+
+        delegate_state = self.delegate.get_agent_state()
+
+        # update iteration that is shared across agents
+        self.state.iteration = self.delegate.state.iteration
+
+        # close the delegate controller before adding new events
+        asyncio.get_event_loop().run_until_complete(self.delegate.close())
+
+        if delegate_state in (AgentState.FINISHED, AgentState.REJECTED):
+            # retrieve delegate result
+            delegate_outputs = (
+                self.delegate.state.outputs if self.delegate.state else {}
+            )
+
+            # prepare delegate result observation
+            # TODO: replace this with AI-generated summary (#2395)
+            formatted_output = ', '.join(
+                f'{key}: {value}' for key, value in delegate_outputs.items()
+            )
+            content = (
+                f'{self.delegate.agent.name} finishes task with {formatted_output}'
+            )
+
+            # emit the delegate result observation
+            obs = AgentDelegateObservation(outputs=delegate_outputs, content=content)
+            self.event_stream.add_event(obs, EventSource.AGENT)
+        else:
+            # delegate state is ERROR
+            # emit AgentDelegateObservation with error content
+            delegate_outputs = (
+                self.delegate.state.outputs if self.delegate.state else {}
+            )
+            content = (
+                f'{self.delegate.agent.name} encountered an error during execution.'
+            )
+
+            # emit the delegate result observation
+            obs = AgentDelegateObservation(outputs=delegate_outputs, content=content)
+            self.event_stream.add_event(obs, EventSource.AGENT)
+
+        # unset delegate so parent can resume normal handling
+        self.delegate = None
+        self.delegateAction = None
 
     async def _step(self) -> None:
         """Executes a single step of the parent or delegate agent. Detects stuck agents and limits on the number of iterations and the task budget."""
@@ -514,14 +612,6 @@ class AgentController:
         if self._pending_action:
             return
 
-        if self.delegate is not None:
-            assert self.delegate != self
-            # TODO this conditional will always be false, because the parent controllers are unsubscribed
-            # remove if it's still useless when delegation is reworked
-            if self.delegate.get_agent_state() != AgentState.PAUSED:
-                await self._delegate_step()
-            return
-
         self.log(
             'info',
             f'LEVEL {self.state.delegate_level} LOCAL STEP {self.state.local_iteration} GLOBAL STEP {self.state.iteration}',
@@ -611,68 +701,6 @@ class AgentController:
         log_level = 'info' if LOG_ALL_EVENTS else 'debug'
         self.log(log_level, str(action), extra={'msg_type': 'ACTION'})
 
-    async def _delegate_step(self) -> None:
-        """Executes a single step of the delegate agent."""
-        await self.delegate._step()  # type: ignore[union-attr]
-        assert self.delegate is not None
-        delegate_state = self.delegate.get_agent_state()
-        self.log('debug', f'Delegate state: {delegate_state}')
-        if delegate_state == AgentState.ERROR:
-            # update iteration that shall be shared across agents
-            self.state.iteration = self.delegate.state.iteration
-
-            # emit AgentDelegateObservation to mark delegate termination due to error
-            delegate_outputs = (
-                self.delegate.state.outputs if self.delegate.state else {}
-            )
-            content = (
-                f'{self.delegate.agent.name} encountered an error during execution.'
-            )
-            obs = AgentDelegateObservation(outputs=delegate_outputs, content=content)
-            self.event_stream.add_event(obs, EventSource.AGENT)
-
-            # close the delegate upon error
-            await self.delegate.close()
-
-            # resubscribe parent when delegate is finished
-            self.event_stream.subscribe(
-                EventStreamSubscriber.AGENT_CONTROLLER, self.on_event, self.id
-            )
-            self.delegate = None
-            self.delegateAction = None
-
-        elif delegate_state in (AgentState.FINISHED, AgentState.REJECTED):
-            self.log('debug', 'Delegate agent has finished execution')
-            # retrieve delegate result
-            outputs = self.delegate.state.outputs if self.delegate.state else {}
-
-            # update iteration that shall be shared across agents
-            self.state.iteration = self.delegate.state.iteration
-
-            # close delegate controller: we must close the delegate controller before adding new events
-            await self.delegate.close()
-
-            # resubscribe parent when delegate is finished
-            self.event_stream.subscribe(
-                EventStreamSubscriber.AGENT_CONTROLLER, self.on_event, self.id
-            )
-
-            # update delegate result observation
-            # TODO: replace this with AI-generated summary (#2395)
-            formatted_output = ', '.join(
-                f'{key}: {value}' for key, value in outputs.items()
-            )
-            content = (
-                f'{self.delegate.agent.name} finishes task with {formatted_output}'
-            )
-            obs = AgentDelegateObservation(outputs=outputs, content=content)
-
-            # clean up delegate status
-            self.delegate = None
-            self.delegateAction = None
-            self.event_stream.add_event(obs, EventSource.AGENT)
-        return
-
     async def _handle_traffic_control(
         self, limit_type: str, current_value: float, max_value: float
     ) -> bool:

openhands/core/config/app_config.py

@@ -27,7 +27,7 @@ class AppConfig:
         runtime: Runtime environment identifier.
         file_store: Type of file store to use.
         file_store_path: Path to the file store.
-        trajectories_path: Folder path to store trajectories.
+        save_trajectory_path: Either a folder path to store trajectories with auto-generated filenames, or a designated trajectory file path.
         workspace_base: Base path for the workspace. Defaults to `./workspace` as absolute path.
         workspace_mount_path: Path to mount the workspace. Defaults to `workspace_base`.
         workspace_mount_path_in_sandbox: Path to mount the workspace in sandbox. Defaults to `/workspace`.
@@ -54,7 +54,7 @@ class AppConfig:
     runtime: str = 'docker'
     file_store: str = 'local'
     file_store_path: str = '/tmp/openhands_file_store'
-    trajectories_path: str | None = None
+    save_trajectory_path: str | None = None
     workspace_base: str | None = None
     workspace_mount_path: str | None = None
     workspace_mount_path_in_sandbox: str = '/workspace'

openhands/core/config/llm_config.py

@@ -138,8 +138,19 @@ class LLMConfig:
         This function is used to create an LLMConfig object from a dictionary,
         with the exception of the 'draft_editor' key, which is a nested LLMConfig object.
         """
-        args = {k: v for k, v in llm_config_dict.items() if not isinstance(v, dict)}
-        if 'draft_editor' in llm_config_dict:
-            draft_editor_config = LLMConfig(**llm_config_dict['draft_editor'])
-            args['draft_editor'] = draft_editor_config
+        # Keep None values to preserve defaults, filter out other dicts
+        args = {
+            k: v
+            for k, v in llm_config_dict.items()
+            if not isinstance(v, dict) or v is None
+        }
+        if (
+            'draft_editor' in llm_config_dict
+            and llm_config_dict['draft_editor'] is not None
+        ):
+            if isinstance(llm_config_dict['draft_editor'], LLMConfig):
+                args['draft_editor'] = llm_config_dict['draft_editor']
+            else:
+                draft_editor_config = LLMConfig(**llm_config_dict['draft_editor'])
+                args['draft_editor'] = draft_editor_config
         return cls(**args)

openhands/core/config/sandbox_config.py

@@ -41,7 +41,7 @@ class SandboxConfig:
 
     remote_runtime_api_url: str = 'http://localhost:8000'
     local_runtime_url: str = 'http://localhost'
-    keep_runtime_alive: bool = True
+    keep_runtime_alive: bool = False
     rm_all_containers: bool = False
     api_key: str | None = None
     base_container_image: str = 'nikolaik/python-nodejs:python3.12-nodejs22'  # default to nikolaik/python-nodejs:python3.12-nodejs22 for eventstream runtime
@@ -60,7 +60,7 @@ class SandboxConfig:
     runtime_startup_env_vars: dict[str, str] = field(default_factory=dict)
     browsergym_eval_env: str | None = None
     platform: str | None = None
-    close_delay: int = 900
+    close_delay: int = 15
     remote_runtime_resource_factor: int = 1
     enable_gpu: bool = False
     docker_runtime_kwargs: str | None = None

openhands/core/config/utils.py

@@ -109,7 +109,6 @@ def load_from_toml(cfg: AppConfig, toml_file: str = 'config.toml'):
     except toml.TomlDecodeError as e:
         logger.openhands_logger.warning(
             f'Cannot parse config from toml, toml values have not been applied.\nError: {e}',
-            exc_info=False,
         )
         return
 
@@ -145,15 +144,48 @@ def load_from_toml(cfg: AppConfig, toml_file: str = 'config.toml'):
                     logger.openhands_logger.debug(
                         'Attempt to load default LLM config from config toml'
                     )
-                    llm_config = LLMConfig.from_dict(value)
-                    cfg.set_llm_config(llm_config, 'llm')
+                    # TODO clean up draft_editor
+                    # Extract generic LLM fields, keeping draft_editor
+                    generic_llm_fields = {}
+                    for k, v in value.items():
+                        if not isinstance(v, dict) or k == 'draft_editor':
+                            generic_llm_fields[k] = v
+                    generic_llm_config = LLMConfig.from_dict(generic_llm_fields)
+                    cfg.set_llm_config(generic_llm_config, 'llm')
+
+                    # Process custom named LLM configs
                     for nested_key, nested_value in value.items():
                         if isinstance(nested_value, dict):
                             logger.openhands_logger.debug(
-                                f'Attempt to load group {nested_key} from config toml as llm config'
+                                f'Processing custom LLM config "{nested_key}":'
                             )
-                            llm_config = LLMConfig.from_dict(nested_value)
-                            cfg.set_llm_config(llm_config, nested_key)
+                            # Apply generic LLM config with custom LLM overrides, e.g.
+                            # [llm]
+                            # model="..."
+                            # num_retries = 5
+                            # [llm.claude]
+                            # model="claude-3-5-sonnet"
+                            # results in num_retries APPLIED to claude-3-5-sonnet
+                            custom_fields = {}
+                            for k, v in nested_value.items():
+                                if not isinstance(v, dict) or k == 'draft_editor':
+                                    custom_fields[k] = v
+                            merged_llm_dict = generic_llm_config.__dict__.copy()
+                            merged_llm_dict.update(custom_fields)
+                            # TODO clean up draft_editor
+                            # Handle draft_editor with fallback values:
+                            # - If draft_editor is "null", use None
+                            # - If draft_editor is in custom fields, use that value
+                            # - If draft_editor is not specified, fall back to generic config value
+                            if 'draft_editor' in custom_fields:
+                                if custom_fields['draft_editor'] == 'null':
+                                    merged_llm_dict['draft_editor'] = None
+                            else:
+                                merged_llm_dict['draft_editor'] = (
+                                    generic_llm_config.draft_editor
+                                )
+                            custom_llm_config = LLMConfig.from_dict(merged_llm_dict)
+                            cfg.set_llm_config(custom_llm_config, nested_key)
                 elif key is not None and key.lower() == 'security':
                     logger.openhands_logger.debug(
                         'Attempt to load security config from config toml'
@@ -167,7 +199,6 @@ def load_from_toml(cfg: AppConfig, toml_file: str = 'config.toml'):
             except (TypeError, KeyError) as e:
                 logger.openhands_logger.warning(
                     f'Cannot parse [{key}] config from toml, values have not been applied.\nError: {e}',
-                    exc_info=False,
                 )
         else:
             logger.openhands_logger.warning(f'Unknown section [{key}] in {toml_file}')
@@ -204,7 +235,6 @@ def load_from_toml(cfg: AppConfig, toml_file: str = 'config.toml'):
     except (TypeError, KeyError) as e:
         logger.openhands_logger.warning(
             f'Cannot parse [sandbox] config from toml, values have not been applied.\nError: {e}',
-            exc_info=False,
         )
 
 
@@ -461,7 +491,11 @@ def setup_config_from_args(args: argparse.Namespace) -> AppConfig:
 
     # Override with command line arguments if provided
     if args.llm_config:
-        llm_config = get_llm_config_arg(args.llm_config)
+        # if we didn't already load it, get it from the toml file
+        if args.llm_config not in config.llms:
+            llm_config = get_llm_config_arg(args.llm_config)
+        else:
+            llm_config = config.llms[args.llm_config]
         if llm_config is None:
             raise ValueError(f'Invalid toml file, cannot read {args.llm_config}')
         config.set_llm_config(llm_config)

openhands/core/logger.py

@@ -52,6 +52,14 @@ LOG_COLORS: Mapping[str, ColorType] = {
 }
 
 
+class StackInfoFilter(logging.Filter):
+    def filter(self, record):
+        if record.levelno >= logging.ERROR:
+            record.stack_info = True
+            record.exc_info = True
+        return True
+
+
 class NoColorFormatter(logging.Formatter):
     """Formatter for non-colored logging in files."""
 
@@ -260,6 +268,9 @@ if LOG_LEVEL in logging.getLevelNamesMapping():
     current_log_level = logging.getLevelNamesMapping()[LOG_LEVEL]
 openhands_logger.setLevel(current_log_level)
 
+if DEBUG:
+    openhands_logger.addFilter(StackInfoFilter())
+
 if current_log_level == logging.DEBUG:
     LOG_TO_FILE = True
     openhands_logger.debug('DEBUG mode enabled.')

openhands/core/main.py

@@ -164,12 +164,12 @@ async def run_controller(
     state = controller.get_state()
 
     # save trajectories if applicable
-    if config.trajectories_path is not None:
-        # if trajectories_path is a folder, use session id as file name
-        if os.path.isdir(config.trajectories_path):
-            file_path = os.path.join(config.trajectories_path, sid + '.json')
+    if config.save_trajectory_path is not None:
+        # if save_trajectory_path is a folder, use session id as file name
+        if os.path.isdir(config.save_trajectory_path):
+            file_path = os.path.join(config.save_trajectory_path, sid + '.json')
         else:
-            file_path = config.trajectories_path
+            file_path = config.save_trajectory_path
         os.makedirs(os.path.dirname(file_path), exist_ok=True)
         histories = [event_to_trajectory(event) for event in state.history]
         with open(file_path, 'w') as f:

openhands/events/stream.py

@@ -65,6 +65,7 @@ class EventStream:
     _queue: queue.Queue[Event]
     _queue_thread: threading.Thread
     _queue_loop: asyncio.AbstractEventLoop | None
+    _thread_pools: dict[str, dict[str, ThreadPoolExecutor]]
     _thread_loops: dict[str, dict[str, asyncio.AbstractEventLoop]]
 
     def __init__(self, sid: str, file_store: FileStore):
@@ -72,8 +73,8 @@ class EventStream:
         self.file_store = file_store
         self._stop_flag = threading.Event()
         self._queue: queue.Queue[Event] = queue.Queue()
-        self._thread_pools: dict[str, dict[str, ThreadPoolExecutor]] = {}
-        self._thread_loops: dict[str, dict[str, asyncio.AbstractEventLoop]] = {}
+        self._thread_pools = {}
+        self._thread_loops = {}
         self._queue_loop = None
         self._queue_thread = threading.Thread(target=self._run_queue_loop)
         self._queue_thread.daemon = True
@@ -257,7 +258,7 @@ class EventStream:
     def add_event(self, event: Event, source: EventSource):
         if hasattr(event, '_id') and event.id is not None:
             raise ValueError(
-                'Event already has an ID. It was probably added back to the EventStream from inside a handler, trigging a loop.'
+                f'Event already has an ID:{event.id}. It was probably added back to the EventStream from inside a handler, triggering a loop.'
             )
         with self._lock:
             event._id = self._cur_id  # type: ignore [attr-defined]
@@ -285,6 +286,8 @@ class EventStream:
                 event = self._queue.get(timeout=0.1)
             except queue.Empty:
                 continue
+
+            # pass each event to each callback in order
             for key in sorted(self._subscribers.keys()):
                 callbacks = self._subscribers[key]
                 for callback_id in callbacks:
@@ -301,8 +304,6 @@ class EventStream:
             except Exception as e:
                 logger.error(
                     f'Error in event callback {callback_id} for subscriber {subscriber_id}: {str(e)}',
-                    exc_info=True,
-                    stack_info=True,
                 )
                 # Re-raise in the main thread so the error is not swallowed
                 raise e

openhands/llm/retry_mixin.py

@@ -46,5 +46,4 @@ class RetryMixin:
         exception = retry_state.outcome.exception()
         logger.error(
             f'{exception}. Attempt #{retry_state.attempt_number} | You can customize retry values in the configuration.',
-            exc_info=False,
         )

openhands/memory/condenser.py

@@ -268,7 +268,7 @@ class LLMSummarizingCondenser(Condenser):
             return [summary_event]
 
         except Exception as e:
-            logger.error('Error condensing events: %s', str(e), exc_info=False)
+            logger.error(f'Error condensing events: {str(e)}')
             raise e
 
 

openhands/microagent/microagent.py

@@ -8,6 +8,7 @@ from pydantic import BaseModel
 from openhands.core.exceptions import (
     MicroAgentValidationError,
 )
+from openhands.core.logger import openhands_logger as logger
 from openhands.microagent.types import MicroAgentMetadata, MicroAgentType
 
 
@@ -132,8 +133,10 @@ def load_microagents_from_dir(
 ]:
     """Load all microagents from the given directory.
 
+    Note, legacy repo instructions will not be loaded here.
+
     Args:
-        microagent_dir: Path to the microagents directory.
+        microagent_dir: Path to the microagents directory (e.g. .openhands/microagents)
 
     Returns:
         Tuple of (repo_agents, knowledge_agents, task_agents) dictionaries
@@ -145,20 +148,24 @@ def load_microagents_from_dir(
     knowledge_agents = {}
     task_agents = {}
 
-    # Load all agents
-    for file in microagent_dir.rglob('*.md'):
-        # skip README.md
-        if file.name == 'README.md':
-            continue
-        try:
-            agent = BaseMicroAgent.load(file)
-            if isinstance(agent, RepoMicroAgent):
-                repo_agents[agent.name] = agent
-            elif isinstance(agent, KnowledgeMicroAgent):
-                knowledge_agents[agent.name] = agent
-            elif isinstance(agent, TaskMicroAgent):
-                task_agents[agent.name] = agent
-        except Exception as e:
-            raise ValueError(f'Error loading agent from {file}: {e}')
+    # Load all agents from .openhands/microagents directory
+    logger.debug(f'Loading agents from {microagent_dir}')
+    if microagent_dir.exists():
+        for file in microagent_dir.rglob('*.md'):
+            logger.debug(f'Checking file {file}...')
+            # skip README.md
+            if file.name == 'README.md':
+                continue
+            try:
+                agent = BaseMicroAgent.load(file)
+                if isinstance(agent, RepoMicroAgent):
+                    repo_agents[agent.name] = agent
+                elif isinstance(agent, KnowledgeMicroAgent):
+                    knowledge_agents[agent.name] = agent
+                elif isinstance(agent, TaskMicroAgent):
+                    task_agents[agent.name] = agent
+                logger.debug(f'Loaded agent {agent.name} from {file}')
+            except Exception as e:
+                raise ValueError(f'Error loading agent from {file}: {e}')
 
     return repo_agents, knowledge_agents, task_agents

openhands/resolver/issue_definitions.py

@@ -45,6 +45,7 @@ class IssueHandlerInterface(ABC):
 
 class IssueHandler(IssueHandlerInterface):
     issue_type: ClassVar[str] = 'issue'
+    default_git_patch: ClassVar[str] = 'No changes made yet'
 
     def __init__(self, owner: str, repo: str, token: str, llm_config: LLMConfig):
         self.download_url = 'https://api.github.com/repos/{}/{}/issues'
@@ -276,7 +277,11 @@ class IssueHandler(IssueHandlerInterface):
             'r',
         ) as f:
             template = jinja2.Template(f.read())
-        prompt = template.render(issue_context=issue_context, last_message=last_message)
+        prompt = template.render(
+            issue_context=issue_context,
+            last_message=last_message,
+            git_patch=git_patch or self.default_git_patch,
+        )
 
         # Get the LLM response and check for 'success' and 'explanation' in the answer
         response = self.llm.completion(messages=[{'role': 'user', 'content': prompt}])
@@ -685,7 +690,7 @@ class PRHandler(IssueHandler):
             feedback=review_thread.comment,
             files_context=files_context,
             last_message=last_message,
-            git_patch=git_patch or 'No changes made yet',
+            git_patch=git_patch or self.default_git_patch,
         )
 
         return self._check_feedback_with_llm(prompt)
@@ -712,7 +717,7 @@ class PRHandler(IssueHandler):
             issue_context=issues_context,
             thread_context=thread_context,
             last_message=last_message,
-            git_patch=git_patch or 'No changes made yet',
+            git_patch=git_patch or self.default_git_patch,
         )
 
         return self._check_feedback_with_llm(prompt)
@@ -739,7 +744,7 @@ class PRHandler(IssueHandler):
             issue_context=issues_context,
             review_context=review_context,
             last_message=last_message,
-            git_patch=git_patch or 'No changes made yet',
+            git_patch=git_patch or self.default_git_patch,
         )
 
         return self._check_feedback_with_llm(prompt)

openhands/resolver/patching/patch.py

@@ -610,10 +610,14 @@ def parse_unified_diff(text):
                 # - Start at line 1 in the old file and show 6 lines
                 # - Start at line 1 in the new file and show 6 lines
                 old = int(h.group(1))  # Starting line in old file
-                old_len = int(h.group(2)) if len(h.group(2)) > 0 else 1  # Number of lines in old file
+                old_len = (
+                    int(h.group(2)) if len(h.group(2)) > 0 else 1
+                )  # Number of lines in old file
 
                 new = int(h.group(3))  # Starting line in new file
-                new_len = int(h.group(4)) if len(h.group(4)) > 0 else 1  # Number of lines in new file
+                new_len = (
+                    int(h.group(4)) if len(h.group(4)) > 0 else 1
+                )  # Number of lines in new file
 
                 h = None
                 break
@@ -622,7 +626,9 @@ def parse_unified_diff(text):
         for n in hunk:
             # Each line in a unified diff starts with a space (context), + (addition), or - (deletion)
             # The first character is the kind, the rest is the line content
-            kind = n[0] if len(n) > 0 else ' '  # Empty lines in the hunk are treated as context lines
+            kind = (
+                n[0] if len(n) > 0 else ' '
+            )  # Empty lines in the hunk are treated as context lines
             line = n[1:] if len(n) > 1 else ''
 
             # Process the line based on its kind

openhands/resolver/prompts/guess_success/issue-success-check.jinja

@@ -3,6 +3,9 @@ Given the following issue description and the last message from an AI agent atte
 Issue description:
 {{ issue_context }}
 
+Changes made (git patch):
+{{ git_patch }}
+
 Last message from AI agent:
 {{ last_message }}
 

openhands/runtime/action_execution_server.py

@@ -522,9 +522,7 @@ if __name__ == '__main__':
             observation = await client.run_action(action)
             return event_to_dict(observation)
         except Exception as e:
-            logger.error(
-                f'Error processing command: {str(e)}', exc_info=True, stack_info=True
-            )
+            logger.error(f'Error while running /execute_action: {str(e)}')
             raise HTTPException(
                 status_code=500,
                 detail=traceback.format_exc(),
@@ -716,7 +714,7 @@ if __name__ == '__main__':
             return sorted_entries
 
         except Exception as e:
-            logger.error(f'Error listing files: {e}', exc_info=True)
+            logger.error(f'Error listing files: {e}')
             return []
 
     logger.debug(f'Starting action execution API on port {args.port}')

openhands/runtime/base.py

@@ -4,10 +4,13 @@ import copy
 import json
 import os
 import random
+import shutil
 import string
+import tempfile
 from abc import abstractmethod
 from pathlib import Path
 from typing import Callable
+from zipfile import ZipFile
 
 from requests.exceptions import ConnectionError
 
@@ -37,9 +40,7 @@ from openhands.events.observation import (
 from openhands.events.serialization.action import ACTION_TYPE_TO_CLASS
 from openhands.microagent import (
     BaseMicroAgent,
-    KnowledgeMicroAgent,
-    RepoMicroAgent,
-    TaskMicroAgent,
+    load_microagents_from_dir,
 )
 from openhands.runtime.plugins import (
     JupyterRequirement,
@@ -125,7 +126,7 @@ class Runtime(FileEditRuntimeMixin):
     def setup_initial_env(self) -> None:
         if self.attach_to_existing:
             return
-        logger.debug(f'Adding env vars: {self.initial_env_vars}')
+        logger.debug(f'Adding env vars: {self.initial_env_vars.keys()}')
         self.add_env_vars(self.initial_env_vars)
         if self.config.sandbox.runtime_startup_env_vars:
             self.add_env_vars(self.config.sandbox.runtime_startup_env_vars)
@@ -172,7 +173,7 @@ class Runtime(FileEditRuntimeMixin):
         obs = self.run(CmdRunAction(cmd))
         if not isinstance(obs, CmdOutputObservation) or obs.exit_code != 0:
             raise RuntimeError(
-                f'Failed to add env vars [{env_vars}] to environment: {obs.content}'
+                f'Failed to add env vars [{env_vars.keys()}] to environment: {obs.content}'
             )
 
     def on_event(self, event: Event) -> None:
@@ -193,11 +194,7 @@ class Runtime(FileEditRuntimeMixin):
                 e, AgentRuntimeDisconnectedError
             ):
                 err_id = 'STATUS$ERROR_RUNTIME_DISCONNECTED'
-            logger.error(
-                'Unexpected error while running action',
-                exc_info=True,
-                stack_info=True,
-            )
+            self.log('error', f'Unexpected error while running action: {str(e)}')
             self.log('error', f'Problematic action: {str(event)}')
             self.send_error_message(err_id, str(e))
             self.close()
@@ -210,7 +207,7 @@ class Runtime(FileEditRuntimeMixin):
         source = event.source if event.source else EventSource.AGENT
         self.event_stream.add_event(observation, source)  # type: ignore[arg-type]
 
-    def clone_repo(self, github_token: str, selected_repository: str):
+    def clone_repo(self, github_token: str, selected_repository: str) -> str:
         if not github_token or not selected_repository:
             raise ValueError(
                 'github_token and selected_repository must be provided to clone a repository'
@@ -227,25 +224,42 @@ class Runtime(FileEditRuntimeMixin):
         )
         self.log('info', f'Cloning repo: {selected_repository}')
         self.run_action(action)
+        return dir_name
 
     def get_microagents_from_selected_repo(
         self, selected_repository: str | None
     ) -> list[BaseMicroAgent]:
+        """Load microagents from the selected repository.
+        If selected_repository is None, load microagents from the current workspace.
+
+        This is the main entry point for loading microagents.
+        """
+
         loaded_microagents: list[BaseMicroAgent] = []
-        dir_name = Path('.openhands') / 'microagents'
+        workspace_root = Path(self.config.workspace_mount_path_in_sandbox)
+        microagents_dir = workspace_root / '.openhands' / 'microagents'
+        repo_root = None
         if selected_repository:
-            dir_name = Path('/workspace') / selected_repository.split('/')[1] / dir_name
+            repo_root = workspace_root / selected_repository.split('/')[1]
+            microagents_dir = repo_root / '.openhands' / 'microagents'
+        self.log(
+            'info',
+            f'Selected repo: {selected_repository}, loading microagents from {microagents_dir} (inside runtime)',
+        )
 
         # Legacy Repo Instructions
         # Check for legacy .openhands_instructions file
-        obs = self.read(FileReadAction(path='.openhands_instructions'))
-        if isinstance(obs, ErrorObservation):
+        obs = self.read(
+            FileReadAction(path=str(workspace_root / '.openhands_instructions'))
+        )
+        if isinstance(obs, ErrorObservation) and repo_root is not None:
+            # If the instructions file is not found in the workspace root, try to load it from the repo root
             self.log(
                 'debug',
-                f'openhands_instructions not present, trying to load from {dir_name}',
+                f'.openhands_instructions not present, trying to load from repository {microagents_dir=}',
             )
             obs = self.read(
-                FileReadAction(path=str(dir_name / '.openhands_instructions'))
+                FileReadAction(path=str(repo_root / '.openhands_instructions'))
             )
 
         if isinstance(obs, FileReadObservation):
@@ -256,44 +270,40 @@ class Runtime(FileEditRuntimeMixin):
                 )
             )
 
-        # Check for local repository microagents
-        files = self.list_files(str(dir_name))
-        self.log('info', f'Found {len(files)} local microagents.')
-        if 'repo.md' in files:
-            obs = self.read(FileReadAction(path=str(dir_name / 'repo.md')))
-            if isinstance(obs, FileReadObservation):
-                self.log('info', 'repo.md microagent loaded.')
-                loaded_microagents.append(
-                    RepoMicroAgent.load(
-                        path=str(dir_name / 'repo.md'), file_content=obs.content
-                    )
-                )
+        # Load microagents from directory
+        files = self.list_files(str(microagents_dir))
+        if files:
+            self.log('info', f'Found {len(files)} files in microagents directory.')
+            zip_path = self.copy_from(str(microagents_dir))
+            microagent_folder = tempfile.mkdtemp()
 
-        if 'knowledge' in files:
-            knowledge_dir = dir_name / 'knowledge'
-            _knowledge_microagents_files = self.list_files(str(knowledge_dir))
-            for fname in _knowledge_microagents_files:
-                obs = self.read(FileReadAction(path=str(knowledge_dir / fname)))
-                if isinstance(obs, FileReadObservation):
-                    self.log('info', f'knowledge/{fname} microagent loaded.')
-                    loaded_microagents.append(
-                        KnowledgeMicroAgent.load(
-                            path=str(knowledge_dir / fname), file_content=obs.content
-                        )
-                    )
+            # Properly handle the zip file
+            with ZipFile(zip_path, 'r') as zip_file:
+                zip_file.extractall(microagent_folder)
+
+            # Add debug print of directory structure
+            self.log('debug', 'Microagent folder structure:')
+            for root, _, files in os.walk(microagent_folder):
+                relative_path = os.path.relpath(root, microagent_folder)
+                self.log('debug', f'Directory: {relative_path}/')
+                for file in files:
+                    self.log('debug', f'  File: {os.path.join(relative_path, file)}')
+
+            # Clean up the temporary zip file
+            zip_path.unlink()
+            # Load all microagents using the existing function
+            repo_agents, knowledge_agents, task_agents = load_microagents_from_dir(
+                microagent_folder
+            )
+            self.log(
+                'info',
+                f'Loaded {len(repo_agents)} repo agents, {len(knowledge_agents)} knowledge agents, and {len(task_agents)} task agents',
+            )
+            loaded_microagents.extend(repo_agents.values())
+            loaded_microagents.extend(knowledge_agents.values())
+            loaded_microagents.extend(task_agents.values())
+            shutil.rmtree(microagent_folder)
 
-        if 'tasks' in files:
-            tasks_dir = dir_name / 'tasks'
-            _tasks_microagents_files = self.list_files(str(tasks_dir))
-            for fname in _tasks_microagents_files:
-                obs = self.read(FileReadAction(path=str(tasks_dir / fname)))
-                if isinstance(obs, FileReadObservation):
-                    self.log('info', f'tasks/{fname} microagent loaded.')
-                    loaded_microagents.append(
-                        TaskMicroAgent.load(
-                            path=str(tasks_dir / fname), file_content=obs.content
-                        )
-                    )
         return loaded_microagents
 
     def run_action(self, action: Action) -> Observation: