Integrations

Why connect an assistant here

Without a connection, an assistant only knows what you paste into the chat — a snippet of code, a copied error message. Connected, it can look things up itself: inspect a workflow, pull today's logs, run a test, all without you leaving your editor or chat window to switch into the builder. You stay the one deciding what it does — see Permissions & safety.

Claude Desktop & Claude Code

Add the server to your MCP config file (Claude Desktop: claude_desktop_config.json; Claude Code: .claude/mcp.json in your project), then restart the app:

{
  "mcpServers": {
    "fimaflow": {
      "url": "https://mcp.fimaflow.com/p/<project-id>",
      "headers": { "Authorization": "Bearer <token>" }
    }
  }
}

Ask Claude to list your tools to confirm the connection: "What Fimaflow tools do you have access to?"

Cursor

Open Settings → MCP and add a new server with the same URL and token. Cursor lists the connected tools under the agent panel — enable the ones you want available in a given chat.

ChatGPT

Add Fimaflow as a custom connector from ChatGPT's connector settings, using the same URL and token. Custom connectors are available on ChatGPT plans that support them — check your plan if you don't see the option.

JetBrains AI

Across the JetBrains IDE family, add the server from Settings → Tools → AI Assistant → MCP. The same URL/token pair works everywhere — one connection per project you want to expose, not one per IDE.

If the assistant can't see your tools

• • Restart the assistant after adding or editing the config — most don't pick up MCP changes live.
• • Check the token hasn't been revoked from the project's MCP settings.
• • Confirm the URL includes the right project id — a token only works for the project it was generated for.