/api/v1/mcp is an MCP server over Streamable HTTP, on the same core as the
REST API. Chat apps (Claude, ChatGPT, Perplexity, Gemini), coding agents, app
builders (Lovable, v0, Bolt, Replit) and Figma take the URL and send you to a
consent screen, where you pick what the agent may do: Suggest, Read or Edit.
What it gets is a key bound to you: never more than you can do, and gone with
your access. The Agents page has the setup for each.
/.well-known/oauth-protected-resource. Keys an admin makes still work for
anything headless (n8n, Make, scripts): Authorization: Bearer ab_....
Tools
Brand rules are also resources:
artbucket://brand/rules, and
artbucket://brands/{slug}/rules/{context}.
The skill
npx skills add pwnera/artbucket installs a skill that teaches an agent the
workflow: brand rules first, check_use before publishing, provenance on
anything generated. In Claude Code, the plugin brings the MCP server and the
skill together: