Quickstart

From the dashboard click New map, give it a name (e.g. acme-platform), then pick how you want to populate it. There's no wrong choice - all paths land on the same editable canvas.

You are an expert software architect. Generate a `canopy.json` architecture map for Canopy (https://canopy.8starlabs.com).

First, read Canopy's machine-readable docs so you use the exact schema and only valid icon slugs:
- Map schema (every field + accepted values): https://canopy.8starlabs.com/docs/maps/architecture-as-code.md
- Icon / brand slugs: https://canopy.8starlabs.com/docs/maps/icons.md
- Full docs in one file (optional): https://canopy.8starlabs.com/llms-full.txt

Then, from the architecture I describe at the bottom, output ONLY a valid canopy.json - a single JSON object, no prose and no markdown code fence - shaped like:
{
  "name": "<short map name>",
  "nodes": [
    { "id": "web", "label": "Web app", "brand": "nextdotjs", "group": "Frontend" },
    { "id": "api", "label": "API", "brand": "nodedotjs", "group": "Backend" }
  ],
  "edges": [ { "source": "web", "target": "api" } ]
}

Rules:
- Every node needs a unique `id` and a `label`. Use a real `brand` slug from the icon library above; if nothing fits, omit `brand` (Canopy renders a generic box).
- Add `edges` for real dependencies / data flow (`source` and `target` reference node ids).
- Optionally add `group` (a logical layer) and, if I give you costs, spend metadata per the schema.
- Be accurate to what I describe - do not invent services.

My system:
<paste your architecture here: a description, your README / docs, your infra list, or a repo file tree>

• Import from GitHub - install the Canopy GitHub App, pick a repo, and Canopy detects services automatically from your package.json, lockfile, and infra files.
• Browse integrations - search 200+ services and hand-pick the ones that make up your stack.
• Browse templates - start from a curated starter or community template in the templates gallery. Canopy copies it straight into your workspace, ready to edit.
• Ask Canopy AI Coming soon - describe your system in plain English and Canopy generates the map. Pro
• Start blank - drop nodes onto the canvas one at a time.

Canopy will prompt you to install the Canopy GitHub App on your account or organisation. Once installed, select the repo and Canopy parses it and places a node for each detected service.

No GitHub? Skip straight to step 3 and add nodes manually or from integrations.

Every node has an edge handle on its right side. Drag it to another node to draw a directed arrow - web → api, api → db, and so on. Hit Auto-layout in the toolbar to re-flow the graph at any time, or drag nodes into position manually.

Labels are optional - double-click an edge to add one (gRPC, webhook, async).

Click any node to open the Inspector panel on the right. Fill in the monthly cost, billing model (usage-based, flat, per-seat), owner, SLA, and status. Costs roll up per group - see Spend tracking for how roll-ups and budgets work.

Open the map's Export menu and choose CLAUDE.md or AGENTS.md. Drop the file in your repo root so Claude Code, Cursor, or any agent that reads project files sees your full architecture - services, spend, owners, and topology - before it touches a line of code. See all export formats.