dify-mcp-server logo

dify-mcp-server

by MCP-Mirror

This is a simple Model Context Protocol (MCP) server implementation designed for use with dify workflows. It enables the invocation of Dify workflows by calling MCP tools.

View on GitHub

Last updated: N/A

What is dify-mcp-server?

The dify-mcp-server is an implementation of a Model Context Protocol (MCP) server specifically designed to integrate with dify, a platform for building AI applications. It allows external clients to interact with dify workflows through the MCP protocol.

How to use dify-mcp-server?

  1. Prepare a config.yaml file containing your dify base URL and API keys (dify_sks) for different workflows.
  2. Configure your client to run the dify_mcp_server using a command like uv run dify_mcp_server, specifying the directory and configuration path.
  3. Ensure your client supports MCP and can utilize the dify tools exposed by the server.

Key features of dify-mcp-server

  • Integration with dify workflows

  • MCP protocol support

  • Configurable dify API keys

  • Simple setup and execution

  • Enables external access to dify tools

Use cases of dify-mcp-server

  • Integrating dify workflows into external applications

  • Automating dify tasks through MCP

  • Building custom AI agents that leverage dify

  • Extending dify functionality with MCP tools

FAQ from dify-mcp-server

What is dify?

Dify is a platform for building AI applications (https://github.com/langgenius/dify).

What is MCP?

MCP stands for Model Context Protocol, a protocol for interacting with AI models and workflows.

Where do I get the dify_base_url?

The dify_base_url is the base URL of your dify instance, e.g., https://cloud.dify.ai/v1.

Where do I get the dify_app_sks?

The dify_app_sks are the API keys for your dify applications. You can find them in your dify project settings.

What clients support MCP?

Any client that implements the Model Context Protocol can interact with this server.