Runchat Docs

Runchat hosts a remote Model Context Protocol server, so any MCP-compatible AI client can work with your canvas directly — listing and creating workflows, building and wiring nodes, editing code, and running executions. The server lives at:

https://runchat.com/api/mcp

Want Claude to walk you through it? Paste this into any Claude chat:

Read https://docs.runchat.com/concepts/integrations/mcp.md and walk me
through connecting Runchat. Ask me which Claude plan I'm on first, since
the steps differ, then go one step at a time and check what I'm seeing
before moving on.

Claude can’t click the buttons for you, but it will follow along and help when the screen doesn’t match what you expected.

Add the connector

Setup takes about a minute. You do it once — Claude connects to Runchat from Anthropic’s servers, not from your machine, so the connector follows your account into Claude Desktop, claude.ai, and mobile. There’s nothing to install and nothing to repeat per device. The steps below are the same in the desktop app and the browser. Pick the section that matches your plan — this matters, because on Team and Enterprise plans only an Owner can add a custom connector.

Free, Pro & Max
Team & Enterprise
Claude Code (CLI)
Other MCP clients

Open Connectors

In the left sidebar, click Customize, then click Connectors.

The Claude sidebar with the Customize entry highlighted

Start a custom connector

Click the Add dropdown at the top right, then choose Add custom connector.

The Connectors settings page with the Add dropdown open, showing Browse connectors and Add custom connector

Enter the name and URL

Name it Runchat, and paste https://runchat.com/api/mcp into the URL field.Leave Advanced settings alone — the OAuth Client ID and Secret are optional and Runchat doesn’t need them. Click Add.

The Add custom connector dialog filled in with the name Runchat and the URL https://runchat.com/api/mcp

Connect and approve

Runchat now appears in the connectors list with a Connect button. Click it.A browser window opens. Sign in to Runchat if you aren’t already, then approve access when the consent screen appears.

Check it worked

Back in the connectors list, Runchat should show type Web with a Custom badge and a checkmark instead of the Connect button.Ask Claude “list my runchats” to confirm the tools are live.

Free accounts can have one custom connector at a time. If you already have one, you’ll need to remove it first.

On Team and Enterprise plans, custom connectors are added at the organization level. An Owner adds Runchat once, and then every member connects their own Runchat account to it.

If you’re a member rather than an Owner, you won’t see an Add custom connector option at all — that’s expected, not a bug. Skip to For everyone else below, and ask an Owner to do the first part.

For Owners (once per organization)

Open organization connectors

Go to Organization settings → Connectors.

Add a custom web connector

Click Add, hover Custom, then select Web.

Enter the name and URL

Name it Runchat and paste https://runchat.com/api/mcp as the URL. Leave Advanced settings empty — Runchat doesn’t need OAuth credentials. Click Add.

Adding the connector doesn’t sign anyone in. Each member authorizes their own Runchat account separately, and only ever sees their own workflows.

For everyone else

Open Connectors

In the left sidebar, click Customize, then click Connectors.

Connect

Find Runchat in the list and click Connect. A browser window opens — sign in and approve access when the Runchat consent screen appears.

Check it worked

Runchat should now show a checkmark instead of the Connect button. Ask Claude “list my runchats” to confirm.

Run:

claude mcp add --transport http runchat https://runchat.com/api/mcp

The next time you use a tool, Claude Code opens the browser sign-in and consent flow automatically.

This configures Claude Code only. It writes to Claude Code’s own config and has no effect on the Claude desktop app — if you want Runchat in both, set them up separately.

Add a remote / HTTP (Streamable HTTP) server pointing at https://runchat.com/api/mcp. Clients that support OAuth will discover the sign-in flow automatically. For clients without OAuth, authenticate with an API key (see below).

Authenticating with an API key

If your client doesn’t support OAuth, or you’re scripting server-to-server, you can authenticate with a Runchat API key instead of signing in:

Sign into Runchat, click your account button, then Get Runchat API Key.
Create a key and copy it.
Configure your MCP client to send it as a bearer token: Authorization: Bearer <your_api_key>.

Troubleshooting

I don't see an 'Add custom connector' option

You’re almost certainly on a Team or Enterprise plan, where only Owners can add custom connectors. Ask an Owner to add Runchat at the organization level — see the Team & Enterprise tab above — and it will then appear in your list with a Connect button.

The Add button is greyed out, or adding fails

Free accounts are limited to one custom connector. Remove the existing one and try again, or upgrade.

I approved access but Claude says it has no tools

Reopen Customize → Connectors and check Runchat shows a checkmark rather than Connect or Reconnect. If it says Reconnect, the authorization expired — click it and approve again. Then start a new chat: connectors added mid-conversation aren’t picked up by the conversation already in progress.

I signed into the wrong Runchat account

Remove the connector, sign out of Runchat in your browser, then add the connector again and sign in with the account you want. The consent screen uses whichever Runchat session your browser already has.

Do I have to repeat this on my other devices?

No. Remote connectors are attached to your Claude account, not your machine — once added, Runchat is available in the desktop app, claude.ai, and mobile. Claude Code is the one exception; see its tab above.

Claude can't see a specific workflow

The agent can only reach workflows you own or that are shared with your team. Share the workflow’s URL directly in the chat, or use Copy ID from the runchat menu and paste that.

What the agent can do

Once connected, the agent has the same canvas abilities as the in-app assistant:

Find & create runchat workflows — list_runchats, create_runchat
Read the canvas and inspect node parameters — get_canvas, read_nodes
Build prompt, code, input, image, note, and sub-workflow nodes — create_prompt_node, create_code_node, create_input_node, create_image_node, create_note, place_tool, create_artifact_node, update_node
Connect & organize nodes into workflows — connect_nodes, organize_nodes, delete_nodes, delete_edges
Edit code in code nodes with find-and-replace or full rewrites — read_files, edit_file, create_files, delete_files, read_status
Discover models and their parameters — list_models, get_model_params
Find & run published tools — search the tool library (and your own runchats), inspect how a tool is built, and run one directly without adding it to a canvas — search_tools, inspect_tool, execute_tool
Load skills for specialized environments (Rhino, Blender, Revit, HTML) — use_skill
Drive Rhino, Grasshopper & Blender whenever Runchat is open inside the CAD app on the same account — read/build the Grasshopper canvas, run Rhino commands, run Python in Blender, and capture the viewport — grasshopper_api, run_rhino_command, run_blender_command, take_screenshot
Run nodes and read the results — run_nodes
Publish Create reusable tools and apps from your workflows — publish_runchat

The agent can only read, edit, and run workflows you own or that are shared with your team — the same access you have in the app. You can share a URL to a runchat workflow that you want the agent to edit or run, or use Copy ID from the runchat menu and share that instead. The agent can also search your runchats if required.

Using the MCP with Rhino and Grasshopper

The MCP reaches Rhino through your signed-in Runchat plugin session. The plugin just needs to be open — it doesn’t matter which workflow (if any) it’s showing:

Launch Runchat in Rhino and sign in with the same account the MCP is connected to
Ask the agent to run Rhino commands, build a Grasshopper definition, or capture the viewport

The agent still saves scripts and results to a workflow on the canvas. To have it work in a specific one, share that workflow’s URL (or use Copy ID from the Runchat menu); otherwise it can create a new workflow itself.

Example

You: Can you make a runchat that converts a hero image to a collection of marketing assets?

Agent: [creates a new runchat, searches for models, adds nodes to the canvas, runs nodes to test, shares a link]

Running nodes consumes credits. The agent will confirm before executing any run_nodes calls. Share the editor link it returns to watch progress or take over in the app.

Programmatic access

Prefer a terminal to an MCP client? The @runchat/cli command line is a port of this server — the same tools as scriptable runchat … commands. Or call the Canvas API over plain HTTP.

Get Started

Plugins

Clipchat

Core Concepts

Node Reference

Examples

MCP Server

Add the connector

For Owners (once per organization)

For everyone else

Authenticating with an API key

Troubleshooting

What the agent can do

Using the MCP with Rhino and Grasshopper

Example

Programmatic access

​Add the connector

​For Owners (once per organization)

​For everyone else

​Authenticating with an API key

​Troubleshooting

​What the agent can do

​Using the MCP with Rhino and Grasshopper

​Example

​Programmatic access

Add the connector

For Owners (once per organization)

For everyone else

Authenticating with an API key

Troubleshooting

What the agent can do

Using the MCP with Rhino and Grasshopper

Example

Programmatic access