merkuryo/AI Code Reviewer
Automatically reviews PRs using Claude or a local Ollama LLM
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Stale
- License
- None
Inputs
| name | description | required | default |
|---|---|---|---|
| llm_provider | LLM provider to use: claude or ollama | no | ollama |
| anthropic_api_key | Anthropic API key (required if llm_provider=claude) | no | "" |
| ollama_url | Ollama API URL | no | http://localhost:11434 |
| model | Model name (e.g. llama3.2, claude-3-5-haiku-20241022) | no | llama3.2 |
| max_files | Maximum number of files to review (controls costs) | no | 10 |
| focus_areas | Comma-separated focus areas: bugs,security,performance,style | no | bugs,security,performance |
| post_comments | Whether to post review comments on the PR | no | true |
| min_severity | Minimum severity to comment on: low, medium, high, critical | no | medium |
Outputs
| name | description |
|---|---|
| review_summary | Summary of the review findings |
| issues_found | Number of issues found |