ArtJack

Claude Code QA agent with memory, plus a read-only MCP server (verdict-qa-mcp on PyPI): baseline → delta runs (NEW/REGRESSED), a fact harness so the model judges while the system measures, hash-chained run history, flaky quarantine with expiry, and an exit-code release gate for CI. Ships six scored eval fixtures — misses published.

1 plugin · ArtJack/verdict

Your agent says it shipped. Git decides. A percent done you can defend: one markdown table of what must exist, each row held to a commit on main or a merged PR. Refuses the commit that would make the numbers lie. Python stdlib only.

1 plugin · ArtJack/gitdone