Harnix is an open-source CLI that scans repositories and scores how ready they are for reliable AI-assisted development workflows.
AI coding agents perform poorly when repositories lack clear agent instructions, onboarding docs, CI, tests, repo structure, and source-of-truth discipline. Teams often discover these gaps only after wasted agent sessions, inconsistent outputs, or regression-prone changes.
Built and published a TypeScript/Node.js CLI distributed through npm as harnix. It runs locally with `npx harnix scan .`, performs 8 built-in checks, calculates a weighted harness readiness score, and outputs console, Markdown, and self-contained HTML reports with prioritized recommendations.
Harnix reduces the repeated manual work of assessing whether a repository is ready for AI-assisted development. Instead of manually checking for agent instructions, README quality, documentation roots, CI, tests, repo structure, source-of-truth issues, and agent skills, a user runs one CLI command and receives a scored report with prioritized recommendations.
Automated repository harness-readiness assessment across 8 dimensions and replaced a manual repo-readiness review with a repeatable CLI scan that generates console, Markdown, and HTML reports with prioritized recommendations.