andreagriffiths11/proof-agent-verify

Adversarial verification for AI-generated code using GitHub Models API or custom providers (Anthropic, Azure OpenAI, OpenAI-compatible)

View on GitHub

Trust Signals

Scorecard Score
not yet scored
Maintenance Recency
Stale
License
None
namedescriptionrequireddefault
github-tokenGitHub token for authentication (defaults to GITHUB_TOKEN)no${{ github.token }}
models-tokenGitHub token with models:read permission (required for GitHub Models API access). If not provided, falls back to github-token.no""
base-refGit ref to compare against (e.g., origin/main, HEAD~1)noorigin/main
forceForce verification even if threshold not met (bypasses file count checks)nofalse
block-on-failExit with code 1 if verification fails (blocks PR merge)notrue
post-commentPost verification result as PR commentnotrue
comment-modeComment format: full, summary, or collapse (default: collapse)nocollapse
max-comment-lengthMaximum comment length in characters (default: 2000)no2000
namedescription
verdictVerification verdict (PASS, FAIL, or PARTIAL)
summaryVerification summary (full text)