Show HN: I pre-registered the rules,then measured AI agents shipping wrong fixes

Category: infrastructure

Tags: ai-agents, ci-cd, verification

Score: 6.8/10 (Innovation: 7, Technical: 6, Documentation: 7, Utility: 7)

Patchward provides an independent, deterministic verdict layer for AI coding agents that blocks or flags broken fixes before they ship. It pre-registered scoring rules and published reproducible results showing a significant reduction in false accepts. The project addresses a critical trust gap in AI-assisted CI/CD pipelines.

Target audience: backend devs, devops, security engineers

Repository: https://github.com/kolesnikov-arch/patchward · Python · NOASSERTION · 1 stars

View on Hacker News