
Routine
An MCP server that allows users to create custom tools on the fly by stitching together actions from multiple MCP tools into reusable routines.
README
Routine
🌅🍳☕️📰🚗💼 Routine is an MCP server that lets you create custom tools on the fly. It stitches actions executed by multiple MCP tools into a reusable routine.
Installation
Copy the following into your MCP client's JSON file
{
"mcpServers": {
"myRoutines": {
"command": "npx",
"args": ["-y", "mcp-routine"],
"env": {
"ROUTINE_FILENAME": "/path/to/your-tool-file.json"
}
}
}
}
Example
**User**: "add an Onboarding page to the Getting started page
**AI**: I'll help you add an "Onboarding" page to the Getting Started page. First, I'll need to find the Getting Started page ID.
Let me search for it:
Great, I found the Getting Started page. Now I'll create a new "Onboarding" page under it:
I've successfully created a new page titled "Onboarding" under the Getting Started page. The page has been created with a basic title, and you can now access it at the URL provided in the response. Would you like me to add any specific content to this new Onboarding page?
**User**:Yes, add some placeholder information for a standard onboarding experience
**AI**: I'll add some structured placeholder content to the Onboarding page to outline a standard onboarding experience.
**User**: now convert all these steps into a route to insert a page into a parent page
**AI**: I'll create a routine that captures the steps we just performed to insert a page into a parent page with structured content. This will make it reusable for future similar tasks.
I've created a routine called insert_structured_page that captures the workflow we used. The routine has three main steps:
1. Search for the parent page using its title to get the page ID
2. Create a new page under the found parent page
3. Add structured content to the new page with sections and bullet points
The routine is parameterized so you can customize:
- parent_page_title: The title of the parent page to search for
- new_page_title: The title for the new page
- welcome_message: Introduction text for the page
- section1_title: Title for the first section (can add more sections as needed)
- bullet_point1, bullet_point2, bullet_point3: Bullet points under sections
- help_message: The help/support message at the bottom
To use this routine, you would just need to provide these parameters with your desired content, and it will create a structured page with consistent formatting.
Development
Build
npm run build
Execute
npx -y --prefix /path/to/local/mcp-routine routine
Publish
npm publish --access public
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.