omniverse-dev-new-house/Security Review Agent
LLM-powered security review — analyzes PR diffs and posts findings as PR comments
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Stale
- License
- None
Inputs
| name | description | required | default |
|---|---|---|---|
| inference-hub-api-key | NVIDIA Inference Hub API key | yes | — |
| inference-hub-model | Model to use for analysis | — | aws/anthropic/bedrock-claude-opus-4-6 |
| review-type | Type of review: security, quality, or both | — | both |
| custom-instructions | Additional instructions for the reviewer (repo-specific context) | — | "" |
| max-comments | Maximum number of inline comments to post | — | 15 |
| github-token | GitHub token for posting PR comments | — | ${{ github.token }} |
Outputs
| name | description |
|---|---|
| findings-count | Total number of findings |
| summary | Brief summary of the review |