cumartesiolsun/AI PR Review (LM Studio / OpenAI)
Reviews PR diff using OpenAI-compatible Chat Completions (works with LM Studio and OpenAI cloud).
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Stale
- License
- None
Inputs
| name | description | required | default |
|---|---|---|---|
| base_url | OpenAI-compatible base URL (e.g. https://api.openai.com/v1 or http://host:11055/v1) | yes | — |
| api_key | API key/token | yes | — |
| model | Model name (e.g. openai/gpt-oss-20b or qwen3-coder) | yes | — |
| language | Review language | no | Turkish |
| max_files | Max files to review (avoid huge PRs) | no | 25 |
| max_chars | Max diff chars total | no | 120000 |
| fail_on_issues | If true, fail workflow when critical issues found | no | false |
| extra_instructions | Extra reviewer instructions | no | "" |
| comment_marker | Unique marker for sticky comments (use different values for multi-job setups) | no | AI_PR_REVIEW_ACTION |
| review_mode | Review mode: 'summary' for single PR comment, 'inline' for per-file inline comments | no | summary |
| timeout_ms | Request timeout in milliseconds for LLM calls | no | 60000 |
| max_tokens | Max output tokens for LLM response | no | 1024 |
| chunk_size_chars | Chunk size (chars) for splitting large diffs | no | 12000 |
| max_diff_chars_per_file | Max diff chars per file before trimming | no | 50000 |
Outputs
no outputs