eval-harness

Tonone

Founder + Tonone = whole company.

31 specialists. Engineering executes. Product decides. Operations runs. One session, two commands, zero meetings. 214 skills across every discipline. MIT licensed.

The idea

A solo founder used to have one choice: stay small, or hire. Now there's a third path.

Tonone is an open-source AI team you install into Claude Code. Not a generalist assistant — specialists. Each agent owns one domain deeply: infrastructure, security, user research, product strategy, growth. They share context, hand off cleanly, and produce work you can ship.

The engineering team (15 agents) builds and ships. The product team (12 agents) decides what to build and why. The operations team (4 agents) keeps the company running. Together, one founder can run what used to take a company.

Install

Prerequisites: Claude Code v1.0+

From your terminal:

claude plugin marketplace add tonone-ai/tonone
claude plugin install tonone@tonone-ai

Or inside an active Claude Code session:

/plugin marketplace add tonone-ai/tonone
/plugin install tonone@tonone-ai

Codex CLI

Prerequisites: Codex CLI installed

git clone https://github.com/tonone-ai/tonone
cd tonone
codex

Codex reads AGENTS.md automatically. Invoke agents and skills by describing what you want:

> Read agents/forge.md and act as Forge — audit this infrastructure
> Read agents/apex.md — plan this project with S/M/L options
> Follow the workflow in skills/warden-audit/SKILL.md

Skills are markdown workflow documents in skills/<name>/SKILL.md. Read them and follow the steps — no slash commands needed.

Usage

> /apex-plan Build a real-time analytics platform for our IoT fleet
> /helm-brief Define the next product sprint
> /forge-infra Set up cloud infrastructure for a new SaaS product
> /spine-api Design a REST API for user management
> /warden-audit Run a full security audit on this codebase
> /echo-interview Run a user research session
> /crest-roadmap Build a product roadmap
> /mint-runway How long is our runway and how do we extend it?
> /folk-hire Build a hiring pipeline for a senior engineer
> /brace-sla Define our support SLA tiers

Every specialist ships in three modes:

Mode	What It Means	Example Skills
Build	Create from scratch — production-ready, not tutorials	/forge-infra, /spine-api, /prism-ui, /touch-app
Review	Audit and fix existing systems	/warden-audit, /relay-audit, /prism-audit, /vigil-check
Recon	Survey a domain for system takeover	/forge-recon, /spine-recon, /flux-recon, /apex-takeover

The Leads

Apex leads the engineering team. Tell it what you're building:

You: "Build user authentication for our SaaS"

Apex: I see 3 ways to approach this:

  S — Quick & focused (Spine + Warden, ~30K tokens, ~$0.05)
      Basic JWT auth with security review.

  M — Solid implementation (Spine + Warden + Flux + Relay, ~120K tokens, ~$0.20)
      Auth + session management + user schema + CI tests.

  L — Full build-out (+ Vigil + Atlas, ~250K tokens, ~$0.45)
      Everything in M + monitoring + documentation.

  My recommendation: M. Which level?

Helm is the head of product. It orchestrates research, strategy, design, and marketing — then hands off a structured brief to Apex when it's time to build.

System Takeover

Inherited a codebase? Apex runs parallel reconnaissance across all specialists:

> /apex-takeover

Phase 1 — Recon (parallel):
  Atlas maps the architecture
  Forge inventories infrastructure
  Relay assesses the pipeline
  Warden scans for security issues
  Vigil checks observability

Phase 2 — Deep dive (targeted):
  Spine reviews backend quality
  Flux assesses database health
  Prism audits frontend

Phase 3 — Takeover report:
  System map, risk assessment, quick wins, roadmap

The Team

Engineering — 15 agents