Skip to content
Canopy is in pre-release. These docs describe the product at its public launch — commands, tool names, and integration examples reflect what you'll see once binaries ship. Join the waitlist →
Canopy logo

Canopy Documentation

Local code intelligence for AI coding agents. One routed tool. No cloud. No API keys.
Get started in 5 minutes

Start here

Section titled “Start here”

Install Canopy

Download the binary for your platform and verify with canopy --version.

Install guide →

Run setup

canopy setup does everything: provisions the embedding runtime, wires every detected editor, and indexes your repo — one Enter per step.

Index your first repo →

Ask your first questions

canopy answer "what depends on \src/auth.ts`?”` — phrasing picks the recipe; answers come with file:line citations.

Question shapes →

Understand Canopy

Why one routed tool beats a tool catalog. How the index answers by structure, not regex.

Concepts →

Documentation sections

Section titled “Documentation sections”
Getting Started Install, run setup, ask — about five minutes end to end.
Concepts What Canopy is (and isn't), how the index works, one tool and its question shapes.
How-To Guides Task-oriented guides: editor setup, CI integration, cached indexes, debugging.
Reference Complete CLI and canopy_answer reference, configuration, environment variables, error codes.
Troubleshooting Symptom-based diagnostic guides: activation failures, slow indexing, MCP connection issues.
Changelog Release history — the v2 single-tool redesign and the v1.x series before it.

Editor integrations

Section titled “Editor integrations”

canopy setup wires every editor it detects automatically. Manual recipes for each:

Claude Code Anthropic's agentic CLI
Cursor AI-first editor
Zed High-performance editor
VS Code + Continue Open-source AI extension

Need something else?

Section titled “Need something else?”