hzj-jeff-07/AI-CodeGuard Security Scan
AI-powered code security scanner combining Tree-sitter static analysis with optional LLM confirmation. Outputs SARIF for GitHub Code Scanning.
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Stale
- License
- None
Inputs
| name | description | required | default |
|---|---|---|---|
| paths | Files or directories to scan (space-separated) | no | . |
| config | Path to .codeguard.yml config file | no | — |
| output-file | Path to write the SARIF report | no | ai-codeguard.sarif |
| dry-run | Run Stage 1 only (no LLM calls). Set to "false" to enable Stage 2. | no | true |
| fix | Generate fix suggestions during Stage 2 (requires Stage 2 enabled). | no | false |
| api-key | CODEGUARD_API_KEY for Stage 2 (Anthropic / OpenAI). Required only when dry-run=false. | no | — |
| severity | Minimum severity to report: low, medium, high, critical | no | — |
| fail-on-findings | Exit with non-zero code if critical or high severity findings exist | no | true |
| node-version | Node.js version to use | no | 20 |
Outputs
| name | description |
|---|---|
| sarif-file | Path to the generated SARIF report |
| findings-count | Number of findings reported |