Getting started with Spec Kitty
Spec Kitty is an open-source CLI workflow for spec-driven development with AI coding agents. It helps teams turn product intent into implementation with a repeatable path: spec → plan → tasks → implement → review → merge.
The fastest way to evaluate it is to install the CLI, initialize a real project, and open your preferred coding agent in that directory.
Why teams use it
- Requirements and design decisions drift over long agent sessions.
- Parallel work is hard to coordinate across branches.
- Review and acceptance criteria become inconsistent from one feature to the next.
Spec Kitty addresses this with repository-native artifacts, work package workflows, and git worktree isolation.
Who it's for
- Engineering teams using tools like Claude Code, Cursor, Codex, Gemini CLI, Copilot, and OpenCode.
- Tech leads who want predictable, auditable AI-assisted delivery.
- Projects where traceability from requirements to code matters.
Start here when you want to evaluate the CLI in a real repo before deciding how far to take the workflow.
Try it now
If you just want the shortest possible path, this is it:
pipx install spec-kitty-cli
spec-kitty init my-project --ai claude
cd my-projectAfter spec-kitty init, your repo has the project structure, command files, and workflow scaffolding needed to run the spec-driven path inside your tool of choice.
Then start your agent
Launch your agent in the initialized project directory. Spec Kitty installs the project-local command files so the slash commands are ready.
claudegeminicursor-agentopencodeexport CODEX_HOME="$(pwd)/.codex"
codexOnce the agent is open, run /spec-kitty.specify on a real backlog item, then move through plan, tasks, implement, review, and merge.
Supported agents referenceStart with the CLI, then go deeper where it matters
Use the docs when you need install details and command syntax. Use a demo or training when you want the workflow to stick across product, engineering, and reviewers.