Skip to main content

Documentation Index

Fetch the complete documentation index at: https://magica.com/docs/llms.txt

Use this file to discover all available pages before exploring further.

Setup takes under a minute. Grab your API key, paste it into your client’s config, and start asking your assistant to run workflows.

1. Get your API key

Open app.magica.com, then Settings → API Keys → Manage to create or copy a key. This is the same Bearer token used for the REST API.

Step-by-step API key walkthrough

See the full Authentication page for click-by-click screenshots, rate limits, key rotation, and revocation.
Treat the key like a password — store it in your password manager or a .env file, never in committed code.

2. Connect your client

Pick your assistant below. Replace YOUR_API_KEY with the key from step 1.
1

Add the MCP server via CLI

Run this in your terminal:
claude mcp add --transport http magica https://api.magica.com/api/mcp \
  --header "Authorization: Bearer YOUR_API_KEY"
2

Verify the server is registered

claude mcp list
You should see magica in the output.
Terminal output of claude mcp list with magica visible
3

Start using tools

Open a new Claude Code session and ask it to list your workflows or start a run.
If you’re using Claude Code or Codex, prefer the CLI install — it’s faster and avoids hand-editing JSON.

3. Start using tools

Your AI assistant now has access to the Magica MCP tools. Try one of these prompts:
  • “List my workflows.”
  • “Build a workflow that takes a prompt and generates an image with Flux 2 Pro.”
  • “Run the Nano Banana Pro model on this image and add a dog to the scene.”
See Available Tools for the full list.