Usewarden — AI skill for Claude Code

usewarden

**Stop your AI coding agent before it touches something it shouldn't.**

One local guardrail for Claude Code, Cursor, Gemini CLI, Copilot CLI, Codex and OpenCode. It blocks writes outside the repo you're in, `.env` reads, `rm -rf` above the project root, force pushes to `main`, and `curl | sh` — deterministically, in under a millisecond, with **no API key and no tokens**.

Free. Local. No account.

Two people opened issues asking this before anything else, so it belongs here rather than in a FAQ ([#9](https://github.com/djayamah/usewarden/issues/9), [#14](https://github.com/djayamah/usewarden/issues/14)).

• It is free. MIT licence. No paid tier, no account, no sign-up, no trial. npm install -g usewarden is the whole transaction.
• It needs no API key. The blocking is deterministic pattern and scope matching: zero tokens, zero API calls, on every event. The one thing that can ever cost money is the optional Layer 2 drift judge, and only if you hand it your own key for a paid API. Give it nothing and it says so and the blocking carries on.
• Your code stays on your machine. Layer 1 never leaves it. Telemetry is off by default and this version ships no endpoint at all, so there is nowhere for a payload to go. The single exception is that optional judge, which you switch on yourself, to a provider you pick, with your own key — and its input is redacted and length-capped before it goes.

Longer answers in the [FAQ](#faq).


*Real output from a real `claude --dangerously-skip-permissions` session. The agent read the reason, stopped, and explained itself instead of routing around the block.*

Is this your week?

• "It deleted files I never asked it to touch." rm -rf ~/ and its cousins are now a recurring, publ