Harness ScoreHow well-harnessed is your repository?

One deterministic command measures your Cursor harness — AGENTS.md, rules, hooks, tests, CI — and tells you exactly what to fix. No AI in the scan. No network. Same score every time.

Scan your repo

Install in Cursor

Read the guide

The problem

Your AI agent is only as reliable as the harness around it.

Two repos can run the same model and get wildly different results. One has guides that steer the agent, sensors that verify its work, and guardrails that stop damage — the other has none. Harness Score measures that harness in seconds and tells you what to fix next.

How it works

Scan

Run npx harness-score in your repo. The CLI reads your filesystem — 36 checks, zero LLM calls, zero network.

Level

Get a maturity level L0–L4, a 108-point breakdown across six dimensions, and the exact gap blocking the next level.

Fix

Every failed check links to a remediation recipe in the guide — or let the Cursor plugin's /harness-audit skill apply the fixes.

Example output

A diagnosis, not a vibe check

Deterministic: same commit, same score — on your laptop or in CI. Gate merges with --min-level 3 so maturity only ratchets up.

Run the scanner →

  harness-score v0.3.0  ~/my-app

  Maturity: L2 · Guided   Score: 70/108 (65%)

  Context & Guides     ████████████████░░░░  80%
  Skills & Commands    █████████████░░░░░░░  65%
  Hooks & Guardrails   ░░░░░░░░░░░░░░░░░░░░   0%
  Sensors & Feedback   ████████████████░░░░  80%
  CI Feedback          ██████████████░░░░░░  71%
  Hygiene & Safety     ███████████████░░░░░  74%

  To reach L3: sensors ≥ 60%; ci ≥ 50%

Get started