linhan-dev/gh-agent-pi
Run a Pi coding agent on GitHub issues and pull requests.
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Stale
- License
- None
Inputs
| name | description | required | default |
|---|---|---|---|
| base_url | Custom model service base URL. | yes | — |
| provider_api | Pi model provider API adapter. | yes | — |
| provider_auth_header | Whether the Pi provider should set an Authorization header. | yes | — |
| model | Model name. | yes | — |
| model_reasoning | Whether the model supports reasoning. | yes | — |
| model_input | Comma-separated model input modes. | yes | — |
| model_context_window | Model context window size. | yes | — |
| model_max_tokens | Model max output tokens. | yes | — |
| thinking_level | Agent thinking level. | yes | — |
| tools | Comma-separated enabled agent tools. | yes | — |
| compaction_enabled | Whether session compaction is enabled. | yes | — |
| retry_enabled | Whether model retry is enabled. | yes | — |
| retry_max_retries | Maximum model retry attempts. | yes | — |
| no_skills | Disable Pi skills. | yes | — |
| no_context_files | Disable Pi context files. | yes | — |
| llm_key | Model API key. | yes | — |
| timeout | Agent timeout in seconds. | yes | — |
Outputs
| name | description |
|---|---|
| success | Whether the agent completed successfully. |
| response | Agent response text. |