Your AI coding agent scores 10/100. Here's what it's missing.

After testing 2,431 checks across 8 AI coding platforms on real repos, I found a clear pattern: most projects use barely 10% of what's available.

I built Nerviq — a zero-dependency CLI that audits your AI coding agent setup and scores it 0-100.

npx @nerviq/cli audit

Most projects score 10-20 out of 100. After running setup, they jump to 60-80+.

The Top 10 Things You're Probably Missing

1. Instructions file (Critical)

Every AI coding platform has one: CLAUDE.md, AGENTS.md, .cursorrules, GEMINI.md. Without it, the agent doesn't know your build commands, code style, or project rules.

2. Architecture diagrams (73% token savings)

A Mermaid diagram gives your agent the project structure in a fraction of the tokens that prose requires.

3. Hooks > instructions (100% vs 80% compliance)

Written instructions are advisory (~80% compliance). Hooks are deterministic (100%). Auto-lint after every edit. Every time.

4. Verification commands

This is the single highest-leverage thing you can do. — Anthropic Best Practices

Agents perform dramatically better when they can verify their own work.

5. Deny rules

Your AI agent loads .env files automatically. Without deny rules, it can read secrets. This is the most common critical finding.

6. Custom commands

Stop typing the same prompts. Create /test, /deploy, /review as reusable commands.

7. Cross-platform drift

If you use Claude + Cursor + Copilot, their configs probably conflict. nerviq harmony-audit catches this automatically.

8. Permission profiles

Most setups either allow everything or block everything. There are better profiles: read-only, suggest-only, safe-write, power-user.

9. MCP configuration

MCP servers extend what your agent can do, but >10 servers or >80 tools degrades performance.

10. Stack-specific checks

Python, Go, Rust, Java, Ruby, PHP, .NET, Flutter, Swift, Kotlin — each has specific agent configuration needs.

New: Auto-fix

nerviq fix                    # Show what's fixable
nerviq fix claudeMd           # Fix a specific check
nerviq fix --all-critical     # Fix all critical issues

Every fix shows the score impact: Score: 4 → 16 (+12).

Safety first

• nerviq audit — reads only, writes nothing (default)
• nerviq suggest-only — markdown report, no file writes
• --dry-run — preview changes
• --config-only — only touches config files, never source code
• Every write creates rollback artifacts

Try it

npx @nerviq/cli audit              # Quick scan: score + top 3
npx @nerviq/cli audit --full       # All checks + confidence levels
npx @nerviq/cli suggest-only       # Share results with your team
npx @nerviq/cli harmony-audit      # Cross-platform alignment

Zero dependencies. No API keys. Runs locally. AGPL-3.0.

GitHub: github.com/nerviq/nerviq
npm: npmjs.com/package/@nerviq/cli
Website: nerviq.net