FusionAL

Free open-source tool

Your AI harness has a score.

agentstack-init audit checks your Claude Code or Hermes config for broken MCP wiring, missing CLAUDE.md, and unreachable servers — then gives you a score out of 100.

Book free 30-min reviewView on GitHub
terminal
$ pip install agentstack-init
$ agentstack-init audit

✗  CLAUDE.md not found at project root
   Fix: Run agentstack-init init --harness claude_code
✗  No MCP config at ~/.claude/claude_mcp_settings.json
   Fix: Run agentstack-init init --harness claude_code

Score: 60/100
Full report: .fusional/audit-report.json

→ Book a free 30-min review: agentstack.fyi/audit

Three things most harnesses get wrong.

The audit catches the config mistakes that make your AI harness unreliable before you waste a week debugging them.

01

CLAUDE.md present

Your harness needs a project-level CLAUDE.md to give the model context.

02

MCP config wired

Are your MCP servers registered in the right config file for your harness?

03

Servers reachable

A config that points at a dead port is worse than no config.

Get a free 30-min harness review.

Run the audit yourself. Then book a call and we'll walk through the report together, fix what's broken, and wire in anything that's missing.

No pitch. No deck. We look at your config and fix it.

Fix it yourself, or fix it with me.

The CLI is free. Paid tiers are for teams that want the setup done right the first time.

Solo session

$750

one-time

We fix your harness together in a 90-min working session. One harness, one project, everything wired by the time we hang up.

  • Audit run before the call
  • Live fix session (90 min)
  • CLAUDE.md + MCP config written for your project
  • Score rechecked at the end
Book a session

Team setup

from $4,000

one-time

Full harness rollout for engineering teams. We scope the engagement on a call, then build it out: memory, MCP wiring, CLAUDE.md templates, and handoff docs.

  • Everything in Solo session
  • Multi-developer harness config
  • Shared memory and recall setup
  • Written handoff + team walkthrough
Scope a project

Retainer

$99 / mo

solo · team from $750/mo

Ongoing harness health. Monthly audit, config updates as your tooling changes, and a standing async channel for harness questions.

  • Monthly audit + score report
  • Config updates when tools change
  • Async Q&A (48h response)
Start a retainer

Free tier: pip install agentstack-init and run agentstack-init audit locally. The CLI is open source and always will be.