montagne-dev/LampeCI
Supercharge your code reviews and PR documentation with intelligent, AI-powered generation using the Lampe SDK
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Activelast commit Mar 5, 2026
- License
- GPL 3.0
Pinned Snippet
uses: montagne-dev/lampe@20dc1215eea0df18861cd7053f7c4a4be52e9910 # v0.8.0tags can be moved; commit SHAs can't. why a SHA?
Inputs
| name | description | required | default |
|---|---|---|---|
| command | The Lampe CLI command to run (describe, review, healthcheck) | yes | describe |
| repo | Repository path (defaults to current directory) | no | . |
| title | PR title (for local runs) | no | Pull Request |
| pr_number | PR number | no | ${{ github.event.number }} |
| base_ref | Base ref for comparison | no | ${{ github.base_ref }} |
| head_ref | Head ref for comparison | no | ${{ github.head_ref }} |
| output | Output provider (auto|console|github|gitlab|bitbucket) | no | auto |
| variant | Generation variant (default|agentic) | no | default |
| files_exclude | Comma-separated list of file patterns to exclude | no | — |
| files_reinclude | Comma-separated list of file patterns to reinclude | no | — |
| max_tokens | Maximum tokens for truncation | no | 100000 |
| timeout_seconds | Timeout in seconds | no | — |
| verbose | Enable verbose output | no | false |
| review_depth | Review depth level (basic|standard|comprehensive). Model selection: basic=gpt-5-nano, standard=gpt-5, comprehensive=gpt-5.1 | no | standard |
| deepen_length | Git fetch deepen length for merge-base | no | 100 |
Outputs
no outputs