ardesolo-alt/LLM Pre-Merge Review
Dual-vendor (Claude Haiku + GPT-4o) code reviewer. Runs on every PR, posts structured findings as a single comment. Non-blocking.
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Stale
- License
- None
Inputs
| name | description | required | default |
|---|---|---|---|
| incident-ledger | Path (from repo root) to a markdown file cataloging past incidents. Fed to both reviewers so recurring bug patterns get flagged. Optional — omit if you do not keep one. | no | "" |
| reviewer-prompt | Path (from repo root) to a custom reviewer prompt. If omitted, uses the default prompt bundled with this action. | no | "" |
| diff-max-chars | Cap on diff size in characters (truncated beyond). Default 80000 (~20k tokens). | no | 80000 |
| anthropic-model | Anthropic model name. Default claude-haiku-4-5-20251001. | no | claude-haiku-4-5-20251001 |
| openai-model | OpenAI primary model. Default gpt-4o. Falls back to gpt-4o-mini then gpt-4-turbo. | no | gpt-4o |
Outputs
no outputs