Claude and MCP

The integration works both ways: Claude Code reads and edits the canvas through MCP, and a node on the canvas asks Claude a question through the agent.

Claude reads and edits the canvas

Download the native MCP client from the MCP page. Run ./myetern-mcp --url https://myetern.com/api --login --app-token-file ./mcp.token, approve the code in your browser, then configure the binary and token file paths in your MCP client. Access follows your application permissions. The user token is exchanged for a temporary access token and renewed automatically. Revoke the device from account settings when needed.

Tools: list workspaces, the scene and single nodes, node text, types, adding and editing nodes and edges, deletion, running a flow by webhook and putting a code schema on the canvas. Edits are sent as targeted patches to nodes and edges rather than replacing the entire scene. Subscribers receive updates when the workspace changes.

Resources myetern://<ws>, /bundle, /node/<key> and /node/<key>/text can be read and subscribed to. The subscription goes through the server room over WebSocket, notifications arrive when the workspace changes. Prompts code_schema, extend_screen and review_scene embed canvas data into the message.

MCP does not expose direct SQL or command execution. It can edit the canvas and invoke supported webhook workflows; permissions and agent configuration govern external actions.

The "Ask Claude" node

"Flow → 🤖 Claude": fields agent, prompt, context and model. The context is assembled by edges from neighbouring nodes. "Ask" sends the agent a claude -p command with the prompt and context, which needs claude * allowed in agent.json and a signed-in Claude on the agent machine. The answer lands in the "answer" field and can be passed along edges anywhere. Claude keys never leave the user's machine, the server is a transit under write access and the cap.shell capability.

← Database schema, migrations and the grid with writes Workplace: modes, layouts, profiles →