brett-buskirk/AgentGate - AI PR Guardrails
Guardrail checks for AI-agent-generated pull requests
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Activelast commit Jul 9, 2026
- License
- MIT
Pinned Snippet
uses: brett-buskirk/agent-gate@c8b789e11475c5e5506b0ed8ec9a1a47aa318918 # v1.2.1tags can be moved; commit SHAs can't. why a SHA?
Inputs
| name | description | required | default |
|---|---|---|---|
| github-token | GitHub token for API access | yes | ${{ github.token }} |
| config-path | Path to .agentgate.yml config file | no | .agentgate.yml |
| fail-on | Minimum severity that fails the check (error | warning | never). Overrides config file. | no | — |
| anthropic-api-key | Anthropic API key for the optional LLM intent check (or set the ANTHROPIC_API_KEY env var). | no | — |
Outputs
| name | description |
|---|---|
| verdict | Overall verdict: pass | warn | fail |
| finding-count | Total number of findings |