Circuitry

Provides access to Circuitry for AI agents to visualize code, understand diagrams, create flowcharts, and generate data visualizations.

typescript

0

GitHub Stars

typescript

Language

6 months ago

First Indexed

3 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "circuitry-dev-circuitry-mcp-server": {
      "command": "npx",
      "args": [
        "-y",
        "@circuitry/mcp-server"
      ],
      "env": {
        "CIRCUITRY_ACCESS_KEY": "YOUR_ACCESS_KEY",
        "CIRCUITRY_ESERVER_URL": "http://localhost:3030"
      }
    }
  }
}

You set up the MCP server to give AI coding agents access to Circuitry, enabling bidirectional code-node syncing, understanding user-drawn diagrams, and generating flowcharts and data visualizations from code. This server acts as the bridge between Circuitry and your AI client, letting you automate code visualization and diagram analysis with ease.

How to use

Install and run the MCP server, then configure your AI client to connect to it. Once connected, you can visualize code files as Circuitry code nodes, understand user-drawn flows, generate new flowcharts, and create data visualizations from code analysis. Use a supported MCP client (Claude Code, Cursor, VS Code, Gemini CLI, etc.). When you give commands like “show me the auth files as code nodes,” the agent creates code nodes from those files and syncs changes back to your source. When you draw a flow, the agent analyzes it in Circuitry and maps it to nodes. When you ask for a new flowchart, the agent creates a visual diagram through Circuitry and returns a completed chart with nodes and connections.

How to install

Prerequisites you must have before installing the MCP server are listed here.

Circuitry Server from circuitry.dev/download and Node.js 18+.
An MCP-compatible AI client (for example Claude Code, Cursor, VS Code, Gemini CLI, etc.).

Configuration and usage notes

To enable the MCP server, you will set up an access key in Circuitry Server and run the MCP setup to store credentials locally. The final configuration points your client to the local EServer and uses the access key for authentication.

Troubleshooting and tips

If you run into connection issues, verify that the Circuitry Server is running and that the EServer URL is reachable. Restart the Circuitry Server if needed and re-run the MCP setup to refresh credentials. If you see an invalid access key, generate a new one in Circuitry Server Preferences and re-run the setup.

Available tools

circuitry.status

Check the connection status between MCP and Circuitry Server.

circuitry.connect

Request a connection to Circuitry, prompting permission if needed.

workflow.getActive

Retrieve information about the currently visible workflow in Circuitry.

workflow.getStructure

Obtain a simplified structure of the current workflow for analysis.

workflow.resolveFlow

Resolve a user reference like "this flow" to the corresponding node IDs.

workflow.getNodeSummary

Get concise details about a specific node in the workflow.

nodes.list

List all nodes within the current workflow.

nodes.get

Retrieve a node by its ID.

nodes.update

Update a node’s configuration.

nodes.delete

Delete a node from the workflow.

code.create

Create a code node from a file path with bidirectional sync.

code.createBatch

Create multiple code nodes from a list of files.

code.setCode

Update the code content and sync it back to source when applicable.

sheet.create

Create a spreadsheet node with data.

sheet.setData

Replace data in an existing spreadsheet node.

agent.chat

Send a message to Circuitry's chat agent for processing.

agent.createFlowchart

Ask the agent to generate a new flowchart in Circuitry.

agent.poll

Poll for an asynchronous agent response.