paniolo-scan — audit

Diagnose the repo's AI-agent harness with the deterministic paniolo scan CLI (via @paniolo/cli). The scanner is read-only — it never writes files, and neither does this skill. Fixing findings is a separate skill: paniolo-scan-remediate. Platform binaries ship with a SHA256SUMS.txt manifest inside the package; paniolo bootstrap verifies the binary it links against it and refuses on a mismatch.

When to use

Use this skill when the user wants to measure how well a repository is set up for coding agents — for example: "scan my harness", "check my AI agent setup", "audit CLAUDE.md / AGENTS.md", "score my agent rules and skills", or the /paniolo-scan trigger. Works in any harness: Claude Code, Cursor, Copilot, Codex, Gemini, and Antigravity.

Flow

1. Scan (deterministic, no writes)