Workflow MCP

Overview

Workflow MCP is a Model Context Protocol (MCP) server: Session-bound Docker workflow server through Codex. State is discarded with the container. In practice that means any MCP-compatible AI assistant can call Workflow MCP's tools directly — the model decides when to use them in a conversation or agent run.

It runs locally: the Docker image docker.io/juliusolsson/workflow-mcp@sha256:4b6889f9e78d0208e6d1d343ba23d0c5d092b1a1513a73c9c85deb2d00ab5855 speaks MCP over stdio on your machine. Configuration happens through environment variables — you'll need OPENAI_API_KEY. The current release is v1.1.0, published under the MIT license, with source at Juliusolsson05/workflow-mcp on GitHub.

Workflow MCP is listed under Cloud Service on mcp.site and works with any MCP client — Claude Desktop, Claude Code, Cursor, VS Code, Windsurf, Zed and the rest of the ecosystem — using the install snippets below. If you maintain Workflow MCP for Juliusolsson05, claim this listing to verify ownership, earn the Verified badge, and keep the details current.

Installation

// claude_desktop_config.json
{
  "mcpServers": {
    "workflow_mcp": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "docker.io/juliusolsson/workflow-mcp@sha256:4b6889f9e78d0208e6d1d343ba23d0c5d092b1a1513a73c9c85deb2d00ab5855"
      ],
      "env": {
        "OPENAI_API_KEY": "<openai-api-key>"
      }
    }
  }
}

Requires OPENAI_API_KEY