ChooChoo for Warp.
Terminal-native agent. Confirm-first guardrails.
Warp's Agent Mode runs commands inside your terminal. ChooChoo writes WARP.md so the agent knows what's safe to run unattended, what needs a confirmation prompt, and what should never run without explicit human OK.
ChooChoo writes WARP.md for Agent Mode.
Warp Agent Mode reads WARP.md for project context and command-safety rules. ChooChoo compiles the file from your real repo and infrastructure boundaries so the agent's surface is explicit and reviewable.
- WARP.md with safe / confirm-first / never-run command surfaces
- Quick commands and project context for fast onboarding
- Pulled from your real CI scripts and infra policies, not written by hand
# WARP.md
# Generated by ChooChoo. Do not edit by hand.
## payments-api
Warp's Agent Mode reads WARP.md for project context.
This file describes what the repo is, how to run it,
and what NOT to touch from the terminal.
## Quick commands
mise run dev Start dev server
mise run test pytest -v
mise run lint ruff check
## Safe surfaces
✓ src/**, tests/**, docs/**
✓ Read-only DB queries against staging
✓ pytest, ruff, mise tasks
## Confirm-first surfaces
⚠ Any DB migration (live or staging)
⚠ Branch deletes, force-pushes
⚠ Anything under infra/, .github/
## Never run without explicit human OK
✗ rm -rf, dropdb, drop table
✗ kubectl apply against prod
✗ Anything that modifies billing config Why use ChooChoo with Warp?
Terminal agents need firm guardrails. ChooChoo writes them down where Warp will read them.
Consistent Output
Generated code follows your linting rules from the start, reducing cleanup work before code review.
Fewer Iterations
The AI understands your standards upfront, producing acceptable code faster and with less back-and-forth.
Team Alignment
Everyone's AI assistant generates code that matches your team's style guide, automatically.
Zero Configuration
Get started instantly with sensible defaults that work out of the box for most projects.
Type-Safe Code
Strict linting rules catch potential bugs and enforce type safety before they reach review.
AI-Ready Rules
Purpose-built rules that AI models understand and follow consistently, session after session.
Ready to run Warp Agent Mode safely?
Compile WARP.md so the agent knows your safe surfaces before it runs the first command.