For developers who build with AI coding agents
Stop babysitting your coding agent.
Spec Kitty is the decision layer for autonomous software delivery: your AI agents write the code, and your people make the calls. Each person curates what they know in Brain Kitty; Team Kitty combines everyone's and hands it to your coding agents through Build Kitty, the free Spec Kitty CLI.
- KnowBrain KittyContact us
- DecideTeam KittyEarly access
- BuildBuild Kitty (Spec Kitty CLI)Free, MIT
Start with the free CLI: MIT, inside Claude Code, Codex, Cursor, Copilot and 20+ others, on your own keys.
What changes in your day
An agent that knows what your team decided
Your agent guesses at conventions nobody wrote down. With Team Kitty, it loads your team's Brain Kitty graphs over MCP: the standards and decisions your colleagues curated, cited.
A second reader on every change
Your agent asks first and writes a spec before it builds. Then a separate reviewer agent reviews the work against the spec and your rules, with checks it can't skip.
Isn't this GitHub's Spec Kit?
It began as a fork and credits it. It is now a different product: parallel worktrees, separate review, and your team's Brain Kitty knowledge served by Team Kitty over MCP.
Run one real ticket through it.
Pick a ticket you'd normally babysit and let Spec Kitty run it.