Download agentrx/github/.env.example from nguyenthanhdo/failure-attribution-data: direct link, hf CLI and curl.
- Browser
- Download file 2.2 kB
-
https://huggingface.co/datasets/nguyenthanhdo/failure-attribution-data/resolve/main/agentrx/github/.env.example
- Command line
-
hf download hf://datasets/nguyenthanhdo/failure-attribution-data/agentrx/github/.env.example
-
curl -L -o .env.example https://huggingface.co/datasets/nguyenthanhdo/failure-attribution-data/resolve/main/agentrx/github/.env.example
2.2 kB
| # ============================================================ | |
| # AgentRx Environment Configuration | |
| # ============================================================ | |
| # Copy this file to .env and fill in your values: | |
| # ---------------------------------------------------------- | |
| # Default endpoint type: "copilot", "azure", or "trapi" (override per-command with --endpoint) | |
| # copilot — uses GitHub Copilot CLI (no Azure creds needed, just `gh auth login`) | |
| # azure — Azure OpenAI (run `az login` first) | |
| # trapi — Microsoft Research internal (run `az login` first) | |
| # ---------------------------------------------------------- | |
| AGENT_VERIFY_ENDPOINT_TYPE=copilot | |
| # ---------------------------------------------------------- | |
| # Copilot CLI endpoint (default) — no extra config needed | |
| # Just ensure `copilot` is on PATH and authenticated via `gh auth login` | |
| # ---------------------------------------------------------- | |
| # AGENT_VERIFY_COPILOT_TIMEOUT=600 # optional: timeout in seconds per LLM call | |
| # Supported models (run `copilot --help` for the current list): | |
| # claude-opus-4.6 (default) | |
| # claude-sonnet-4.5 | |
| # gpt-5 | |
| # gpt-5-mini | |
| # gpt-5-codex | |
| AGENT_VERIFY_COPILOT_MODEL=claude-opus-4.6 | |
| # ---------------------------------------------------------- | |
| # Azure OpenAI endpoint (use --endpoint azure) | |
| # ---------------------------------------------------------- | |
| AGENT_VERIFY_ENDPOINT= # e.g. https://my-resource.openai.azure.com/ | |
| AGENT_VERIFY_DEPLOYMENT=gpt-5 # model deployment name | |
| AGENT_VERIFY_MODEL_NAME=gpt-5 # model name for judge (must match deployment name for Azure) | |
| AGENT_VERIFY_API_VERSION=2024-12-01-preview | |
| AGENT_VERIFY_CLIENT_ID= # Managed Identity client ID (if applicable) | |
| # ---------------------------------------------------------- | |
| # TRAPI endpoint (Microsoft Research internal, use --endpoint trapi) | |
| # See https://aka.ms/trapi/models for instance and deployment names. | |
| # ---------------------------------------------------------- | |
| AGENT_VERIFY_TRAPI_INSTANCE= # e.g. my-instance/my-pool | |
| AGENT_VERIFY_TRAPI_DEPLOYMENT_NAME= # e.g. gpt-5 | |
| SCOPE= # Azure AD scope for TRAPI |