willisback/AI Code Review
Review a GitHub Pull Request using a self-hosted vLLM instance. Posts a structured Markdown comment directly on the PR.
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Stale
- License
- None
Inputs
| name | description | required | default |
|---|---|---|---|
| vllm-url | Base URL of the vLLM server (e.g. http://<your-host>:30000/v1) | yes | — |
| github-token | GitHub token for fetching PR diff and posting the review comment | yes | — |
| vllm-model | Override model ID. Auto-detected from /v1/models if empty. | no | "" |
| vllm-timeout | Total request timeout in seconds | no | 120 |
| vllm-retries | Number of retries on LLM request failure | no | 2 |
| vllm-api-key | API key for vLLM server authentication (optional, defaults to 'none') | no | "" |
Outputs
no outputs