partnerloop

partnerloop connects to CreationLoop's hosted MCP endpoint and exposes its public design-partner surface: the canonical company overview, the design-partner program offer, the FAQ, and a way to submit a design-partner application. One binary serves a command-line interface and a stdio MCP server; both simply forward each call upstream, so the text you get back is CreationLoop's own canonical copy, never a local paraphrase.

No API key or account is needed. The endpoint can be overridden with CREATIONLOOP_MCP_URL; when it is not set, calls go to https://creationloop.ai/api/mcp.

Install

npx @reneturcios/partnerloop overview

Or run the container image:

docker run -i ghcr.io/cloudcorprecords/partnerloop

The container defaults to the MCP server (see below). Pass CLI arguments to use it as a CLI instead.

CLI

partnerloop overview
partnerloop offer
partnerloop faq
partnerloop apply --email founder@example.com --website https://example.com \
  --name Ada --company Example --message "Scaling content."

partnerloop --version

The three read commands print the upstream text exactly as the hosted server returns it. apply prints the upstream confirmation. Exit codes: 0 on success, 1 when a call fails (a validation problem, an upstream error, or a rate limit), 2 for usage errors.

Applications are checked locally before anything is sent: email (a valid address, 3–320 characters) and website (1–500 characters) are required; name and company may be up to 200 characters each and message up to 2000. Unknown fields are rejected. A failed check exits 1 with the problem named, and no request is made.

MCP server

One bin serves both faces. The mcp subcommand starts a stdio MCP server:

npx -y @reneturcios/partnerloop mcp

MCP client configuration:

{
  "mcpServers": {
    "partnerloop": {
      "command": "npx",
      "args": ["-y", "@reneturcios/partnerloop", "mcp"]
    }
  }
}

To point the server at a different CreationLoop-compatible host, add CREATIONLOOP_MCP_URL to the client's env for this server.

Tools

Upstream errors — including the endpoint's rate limits (30 requests/minute overall, 5 submissions/minute for applications) — come back as error results carrying the upstream message.

Development

npm install
npm run build   # tsc -> dist/
npm test        # node test runner over test/*.test.ts, no network (fake fetch fixtures)
npm run dev     # run the CLI from source via tsx

License

MIT — see LICENSE.