Conductor

Conductor

Interact with Conductor (OSS and Orkes) REST APIs.

12
Stars
4
Forks
9
Releases

Overview

oss-conductor-mcp is a Model Context Protocol (MCP) server for Conductor (OSS and Orkes). It runs an MCP server capable of interacting with a Conductor instance and provides the basic operations that may be needed by an MCP client for workflow creation, execution, and analysis. The package can be installed via pip install conductor-mcp and configured with a JSON config containing CONDUCTOR_SERVER_URL (the /api path is required for most applications), CONDUCTOR_AUTH_KEY, and CONDUCTOR_AUTH_SECRET. The server can be plugged into AI agents such as Claude or Cursor by adding an mcpServers entry that points to conductor-mcp with the appropriate --config path. It supports running as a UV-based service (uv run conductor-mcp --config <path>) and can also rely on environment variables if a config file is not provided. A local_development.py exists to assist development and is used when the --local_dev flag is set. The README also demonstrates example prompts for creating and executing Conductor workflows, illustrating how the MCP server enables AI-driven workflow orchestration with Conductor’s REST APIs (OSS and Orkes).

Details

Owner
conductor-oss
Language
Python
License
Apache License 2.0
Updated
2025-12-07

Features

Conductor REST API integration (OSS and Orkes)

MCP server can interact with Conductor's REST endpoints to manage workflows.

Workflow creation, execution, and analysis support

Provides essential MCP operations for creating, running, and analyzing workflows via MCP clients.

Configurable via JSON or environment variables

Config can be provided as a JSON file with server URL and credentials or via environment variables.

AI agent integration (Claude and Cursor)

Designed to be plugged into AI agents, enabling natural language prompts to drive Conductor workflows.

PyPI package: conductor-mcp

Installable via pip and ready for quick setup.

UV-based runtime

Runs under the uv runtime with commands like 'uv run conductor-mcp --config <path>'.

Local development support

Local development workflow supported via local_development.py and the --local_dev flag.

Audience

AI DevelopersIntegrate into AI assistants to create, run, and analyze Conductor workflows.
Workflow EngineersUse MCP to manage Conductor workflows programmatically through prompts.
Platform IntegratorsPlug MCP server into platforms to extend automation with AI assistants.

Tags

ConductorMCPWorkflowOrkesOSSREST APIAI integrationuvPythonconfigauthentication