One spec, multiple agents

Aigon orchestrates Claude Code, Codex, Cursor CLI (and more) to implement your specs in parallel, judged by an LLM, or through pre-defined review stages.

Schedule your specs to run when you have quota (the dreaded 5hr window), or when you are asleep, avoiding FONUT — Fear Of Not Using Tokens.

Aigon is an open-source, spec-driven harness for multi-agent coding. Each feature is a Markdown spec in your repo. Each agent runs in its own git worktree under tmux. You merge the diff that survives review. No SaaS account, no token resale, no lock-in.

Why use Aigon?

Four reasons one agent isn’t enough.

Catch what one model misses. Different model families catch different classes of bug. When Claude reviews Claude, the reviewer shares the implementer’s blind spots. Aigon Fleet runs Claude Code, Codex, Cursor CLI (and more) in parallel against the same spec, with a different coding agent as the merge reviewer.

Stop hitting quota walls. Agent subscriptions hit 5-hour quota limits. Aigon polls your quotas every 30 minutes, surfaces headroom on the dashboard, and auto-fails over to whichever coding agent still has quota. Three subscriptions, one elastic pool. Schedule specs to run when you have quota or while you sleep — no more FONUT (Fear Of Not Using Tokens).

Use the subscriptions you already own. Aigon never sits in your billing path. Spend the Claude Pro, ChatGPT Plus, and Cursor Pro subscriptions you already have — Aigon orchestrates the agent CLIs, not their underlying model APIs. No tokens, no markup, no metering.

Watch the work happen, locally. Each agent runs in its own git worktree under tmux — tmux attach to any session and watch the work happen live. Specs live in your repo as Markdown; every workflow move is a git diff. No cloud sandbox, no opaque execution.

How you drive it

Three surfaces, one state. Slash commands inside your existing CLIs (Claude Code, Gemini CLI, Codex, Cursor). A local web dashboard for the Kanban view. The aigon CLI for everything else. Every surface reads and writes the same specs and workflow state in your repo. Lifecycle: inbox → backlog → in-progress → review → done.

Sections