ProductDevelopersSolutionsPricingDownload
Integrations
AmpAmpClaude CodeClaude CodeClaude CoworkClaude CoworkCodexCodexGitHub CopilotGitHub CopilotCursorCursorDevinDevinFactory DroidFactory DroidGeminiGeminiKimiKimiKiroKiroOpenClawOpenClawOpenCodeOpenCodePiPiWarpWarpZencoderZencoder
Programs
Open SourceScience & Research
Book a Demo
amp logo
Sourcegraph · Amp

ChooChoo for Amp.

Sourcegraph's coding agent. Your team's standards.

Amp is Sourcegraph's autonomous coding agent. ChooChoo writes the AGENTS.md every Amp session reads at startup so it knows your stack, your standards, and the lines it should never cross. All derived from your codebase, not from a wiki.

Configuration

ChooChoo writes your AGENTS.md for Amp.

Amp follows the cross-vendor AGENTS.md spec for project context. ChooChoo compiles it from your linting config, ADRs, and key files so every Amp run starts with the same accurate picture of the codebase.

  • AGENTS.md with stack, standards, boundaries, and ADR pointers
  • Stays in sync with your codebase as it evolves
  • Same file works for Codex, OpenCode, and other AGENTS.md-aware agents
# AGENTS.md
# Generated by ChooChoo. Do not edit by hand.
# Run: choochoo context generate

## Project: payments-api
Stack: Python 3.12 · FastAPI · PostgreSQL · Redis
Style: Ruff (line-length=100) · strict mypy
Tests: pytest · coverage ≥ 85% enforced on CI

## Standards
- Validate all inputs with Pydantic at API boundaries
- No raw SQL; use SQLAlchemy ORM
- Never log secrets, tokens, or payment data
- Annotate all public functions with types

## Boundaries
READ:   src/** · tests/** · docs/**
WRITE:  src/** · tests/**
NEVER:  .github/** · migrations/** · infra/**

## Architecture Decisions
ADR-001  JWT (RS256) for auth
ADR-004  Event-driven webhooks via Redis Streams
ADR-007  API versioning under /v1/, /v2/ namespaces

Why use ChooChoo with Amp?

Sourcegraph's agent is fast. Give it context that's faster.

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 point Amp at your standards?

Compile AGENTS.md from your real code so Amp acts within your team's guardrails from session zero.