gotoolkits/DifyWorkflow
Free用于查询和执行 Dify 工作流的工具。
FreeFree tier
About gotoolkits/DifyWorkflow
mcp-difyworkflow-server is an MCP (Model Context Protocol) server tool that enables querying and executing Dify workflows. It supports multiple custom Dify workflows, allowing users to define workflow names and corresponding API keys in configuration. The server exposes two tools: list_workflows (returns authorized workflows) and execute_workflow (runs a specified workflow with an input message). It is written in Go, licensed under Apache-2.0, and can be built from source or deployed as a pre-compiled binary. Designed for integration with MCP-compatible clients, it allows natural language prompts to trigger Dify workflow execution.
Key Features
Query authorized Dify workflows via list_workflows tool
Execute specified Dify workflows with custom input via execute_workflow tool
Support multiple custom Dify workflows configured by name and API key
Built on the MCP (Model Context Protocol) standard
Configurable base URL and environment variables for Dify API
Written in Go with easy build using go build or Makefile
Open source under Apache-2.0 license
Pros & Cons
Pros
- Open source and free to use
- Supports multiple Dify workflows with simple configuration
- Standard MCP protocol enables wide compatibility with MCP clients
- Lightweight and easy to build from source (Go)
- Clear separation of concerns: workflows are defined and managed in Dify
Cons
- Requires self-hosting and building the server binary
- Depends on a running Dify instance with API keys
- Currently only supports workflows with input variable named 'message'
- Limited to two tools (list_workflows and execute_workflow)
- Documentation is primarily in English and Chinese, may not cover all edge cases
Best For
Integrating Dify workflow automation into MCP-based AI assistantsTriggering document translation, image generation, or other Dify workflows via natural language promptsBuilding custom MCP clients that need on-demand access to multiple Dify workflows
FAQ
How do I configure workflow names and API keys?
Set the DIFY_WORKFLOW_NAME and DIFY_API_KEYS environment variables, ensuring they correspond one-to-one in order. Workflow names are user-defined in prompts, and API keys must be generated in the Dify platform for each workflow.
What tools does this MCP server provide?
It provides two tools: list_workflows (lists authorized workflows) and execute_workflow (executes a specified workflow with a required input variable named 'message').
Can I use multiple Dify workflows at once?
Yes, you can configure multiple workflows by providing comma-separated names in DIFY_WORKFLOW_NAME and corresponding API keys in DIFY_API_KEYS, matching them in order.
Is this tool free to use?
Yes, the project is open source under the Apache-2.0 license and free to use. You only need to build or download the binary and configure it with your own Dify instance.