Skip to main content
Glama
deenesjakoruzh
zhaohongyuziranerran

Workflow Connector MCP Server

by zhaohongyuziranerran
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Hybrid

Workflow Connector MCP Server

AI Agent Workflow Connector MCP Server - Connect SaaS services, create automated workflows, manage webhooks, and provide workflow templates.

Features

  • Notification Sending: DingTalk/WeChat Work/Slack/Email

  • Webhook Management: Create/Receive/Manage webhook endpoints

  • Workflow Orchestration: Triggers + Action chains, supporting conditional logic/delays/HTTP calls/data transformation

  • Pre-built Templates: 8 out-of-the-box workflow templates

10 Tools

Tool

Function

send_notification

Send notifications to DingTalk/WeChat Work/Slack/Email

create_webhook

Create a webhook endpoint

list_webhooks

List all webhooks

delete_webhook

Delete a webhook

create_workflow

Create an automated workflow

list_workflows

List all workflows

run_workflow

Manually execute a workflow

delete_workflow

Delete a workflow

get_workflow_template

Get template details

list_templates

List all templates

8 Pre-built Templates

Template ID

Name

Category

Trigger Method

monitor-alert

Monitoring Alert Notification

devops

HTTP

content-publish

Content Publishing Automation

content

Manual

price-monitor

E-commerce Price Monitoring

ecommerce

Scheduled

daily-report

Automated Daily Report Generation

productivity

Scheduled

lead-nurture

Automated Lead Follow-up

sales

Webhook

data-sync

Data Synchronization Workflow

integration

Scheduled

ai-review

AI Content Moderation

content

Webhook

stock-alert

Stock Volatility Monitoring

finance

Scheduled

Local Installation

pip install -r requirements.txt
python server.py

REST Deployment

python -m workflow_connector_mcp.rest_server

Defaults to listening on 0.0.0.0:8000, supports MCP Streamable HTTP transport.

Usage Examples

Send DingTalk Notification

send_notification(
    channel="dingtalk",
    title="测试通知",
    content="这是一条测试消息",
    webhook_url="https://oapi.dingtalk.com/robot/send?access_token=xxx"
)

Create Workflow

create_workflow(
    name="监控告警",
    description="服务异常时发送通知",
    trigger_type="manual",
    steps='[{"name":"发送告警","action":"notify","config":{"channel":"dingtalk","title":"告警","content":"服务异常"}}]'
)

Execute Workflow

run_workflow(workflow_id="wf_xxx", input_data='{"service":"api","status":"down"}')

Tech Stack

  • FastMCP 2.x

  • httpx (Async HTTP)

  • In-memory storage (Replaceable with Redis/DB)

Online Access (Recommended)

No local installation required, configure directly in your MCP client:

{
  "mcpServers": {
    "workflow-connector-mcp": {
      "url": "http://www.mzse.com/workflow-mcp/"
    }
  }
}

REST API

curl http://www.mzse.com/workflow-mcp/

Deployment Status

Project

Address

Domain

http://www.mzse.com/workflow-mcp/

GitHub

https://github.com/zhaohongyuziranerran/workflow-connector-mcp

License

MIT

This server cannot be deployed

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Resources

Related MCP Connectors