dify-mcp

The most complete MCP server + CLI for Dify

138 tools. 16 namespaces. One registry. Let any AI agent build, test, and ship Dify workflows autonomously — everything a human can do in the UI, now scriptable.

Works with Claude Code · Codex · Gemini CLI · Cursor · Cline · Windsurf · Roo Code · Continue · Aider · Zed — and any other MCP-compatible or shell-capable agent.

What is this?

Dify is a powerful open-source LLM app platform — but its workflow builder is a visual drag-and-drop editor. What if you want an AI agent to programmatically create workflows, wire up nodes, test them, iterate, and publish — without a browser?

dify-mcp is the bridge. It exposes the entire Dify console API as a unified tool registry with two surfaces: a CLI any shell-capable agent can drive, and an MCP server (stdio or Streamable HTTP) any MCP-compatible host can attach. Same 138 tools, same JSON contract, same safety guarantees.

┌──────────────────────────────────────────────────────────┐
│                    dify-mcp                              │
│                                                          │
│   ┌──────────┐    ┌────────────────────┐    ┌─────────┐ │
│   │  CLI     │───▶│   138-tool         │───▶│  Dify   │ │
│   │  difywf  │    │   registry         │    │  API    │ │
│   └──────────┘    │                    │    └─────────┘ │
│   ┌──────────┐    │  app · workflow    │         ▲      │
│   │  MCP     │───▶│  provider · rag    │─────────┘      │
│   │  stdio   │    │  agent · snippet   │                │
│   └──────────┘    │  stats · audio ... │                │
│                   └────────────────────┘                │
└──────────────────────────────────────────────────────────┘

Related MCP server: Dify MCP Server

Works with your favorite agents

dify-mcp is agent-agnostic by design — no SDK lock-in, no proprietary protocol. If your agent can run a shell command, it can use the CLI. If it speaks MCP, it can attach the server. Most popular agents do both:

Don't see your agent? If it supports MCP or can run shell commands, it works. The connect section below has copy-paste configs for each host.

Why you'll star this

• Complete coverage. Not a subset. Not a wrapper around the public API alone. This talks to the internal console API — the same surface the Dify web UI uses. Create apps, author graphs node-by-node, validate, test-run, publish, manage versions, triggers, providers, plugins, RAG pipelines, snippets, agent configs, comments, annotations, audio, stats. If the UI can do it, so can your agent.

• Agent-agnostic by design. No SDK lock-in. The CLI works with any agent that can run a shell command. The MCP server works with any MCP host. Both return structured JSON — { ok, data } or { ok: false, error: { code, message, retryable } } — so agents never scrape human-readable text.

• Cookie auth, handled. Dify's console uses cookie + CSRF double-submit, not Bearer tokens. dify-mcp captures, stores, and auto-refreshes the session — including the server-side refresh-token rotation that silently invalidates naive clients.

• Safe by default. Destructive operations require explicit confirm=true / --yes. Graphs are validated offline before sync. Every mutation is audit-logged. --dry-run shows diffs without saving.

• Zero build step. Runs directly on Node 23.6+ native TypeScript. No compiler, no bundler, no transpiler. Clone, install, go.

Live verified

Every tool category has been tested against cloud.dify.ai with real credentials:

• ✅ Full authoring loop: create app → sync draft → run draft → publish → list versions

• ✅ All 16 namespaces exercised: apps, workflows, providers, plugins, triggers, snippets, RAG, agents, stats, comments, annotations, audio, files, runs, workspace

• ✅ MCP transport: tools/call over stdio and Streamable HTTP with live cookie auth

• ✅ 42 unit tests · typecheck clean · MCP smoke (138 tools)

Quickstart

Prerequisites: Node >= 23.6 (native TypeScript type stripping — no build step).

git clone https://github.com/alexjiaguo/dify-mcp.git
cd dify-mcp
npm install
npm link        # puts `difywf` on your PATH (optional)

Authenticate

The Dify console uses cookie + CSRF auth. The easiest path:

# 1. Export cookies from your browser (cookie-editor extension → Export → JSON)
# 2. Save as cookies.json, then:
difywf auth import-cookies --base-url https://cloud.dify.ai --file cookies.json

# Or self-hosted with email/password (no browser needed):
difywf auth login-console --base-url https://your-dify --email you@x --password '***'

difywf auth status   # confirm: shows base URL + cookie names (values masked)

Build a workflow

difywf agent guide                          # self-onboarding playbook for agents
difywf app list                             # see your apps
difywf app create --mode workflow --name "my-agent-workflow"

difywf wf node defaults <app-id> llm        # get the schema for an LLM node
difywf wf validate --graph graph.json       # offline: structure, refs, cycles
difywf wf draft sync <app-id> --graph graph.json --dry-run   # preview diff
difywf wf draft sync <app-id> --graph graph.json             # save draft
difywf wf test <app-id> --input query="hello"                # test-run the draft
difywf wf publish <app-id> --yes                             # ship it

Connect your agent (MCP)

Same binary, same 138 tools. Copy-paste the config for your host:

claude mcp add dify -- difywf mcp serve

[mcp_servers.dify]
command = "difywf"
args = ["mcp", "serve"]

{
  "mcpServers": {
    "dify": {
      "command": "difywf",
      "args": ["mcp", "serve"]
    }
  }
}

{
  "mcpServers": {
    "dify": {
      "command": "difywf",
      "args": ["mcp", "serve"]
    }
  }
}

Add an MCP server in Windsurf settings (Cmd+, -> MCP Servers) with command difywf and args ["mcp", "serve"].

{
  "context_servers": {
    "dify": {
      "command": "difywf",
      "args": ["mcp", "serve"]
    }
  }
}

Aider doesn't support MCP, but it can run shell commands. Just use the CLI directly:

/run difywf app list
/run difywf wf draft sync <app-id> --graph graph.json

For remote or containerized hosts that can't spawn a local process, run the MCP server over the stateless Streamable HTTP transport:

difywf mcp serve --http --port 8080
# or via env: DIFYWF_MCP_TRANSPORT=http DIFYWF_MCP_PORT=8080 difywf mcp serve

Point any Streamable-HTTP-capable client at http://<host>:8080/mcp. Each POST is a self-contained JSON-RPC message (initialize / tools/list / tools/call); no session is required. GET and DELETE are rejected with 405.

No difywf on PATH? Use the absolute path: node /path/to/dify-mcp/bin/difywf.js mcp serve.

Tools

138 tools across 16 namespaces. Run difywf --help for the full live list, or difywf agent guide for the agent-oriented playbook.

Safety

Error/exit codes: USAGE_ERROR(2), AUTH_REQUIRED(3), CONFIRM_REQUIRED(4), VALIDATION_FAILED(5), RBAC_DENIED(6), NOT_FOUND(7), DSL_VERSION_MISMATCH(8), RATE_LIMITED(9), SERVER_ERROR(10), NETWORK_ERROR(11). Check error.retryable before retrying.

Develop

npm run typecheck   # tsc --noEmit
npm test            # 42 unit tests
npm run smoke:mcp   # MCP stdio smoke (138 tools, JSON-RPC handshake)
npm run smoke:mcp:http   # MCP Streamable HTTP smoke (stateless POST /mcp)

No build step. Source runs directly via Node's type stripping.

License

Apache-2.0 — see LICENSE.

If this saves you time, a ⭐ is the best thank-you.

Built for agents, by agents.