Studio MCP
Studio MCP transforms any CLI command into an AI tool by using a simple template syntax to define input schemas and arguments. It enables AI agents like Claude and Cursor to execute local scripts and system commands through a dynamically generated Model Context Protocol server.
README
Studio MCP
Make any CLI into an AI tool with studio MCP server.
Bright Studio Apt ā Just enough, nothing more ā $0/mo OBO.
Text 404 for more details!
What's Included?
studio is the simplest possible way to add CLI tools that your AI Agent can use right now. Built on the Model Context Protocol, studio spawns a single tool mcp server with your command.
The tool turns everything after studio command into an MCP tool that can be called by Cursor, Claude, etc.
studio is great for patching CLIs into Claude, debugging MCPs or providing custom scripts to your LLM without having to deal with MCP.
It uses a very simple Mustache-like template syntax to provide inputs and descriptions telling the LLM how to use your command.
npx -y @studio-mcp/studio command "{ required_argument # Description of argument }" "[optional_args... # any array of arguments]"
studio turns this into an input schema for the MCP tool so that tool calls know what to send:
{
"type": "object",
"properties": {
"required_argument": {
"type": "string",
"description": "Description of argument"
},
"optional_args": {
"type": "array",
"items": { "type": "string" },
"description": "any array of arguments"
}
},
"required": ["required_argument"]
}
You can run almost any command, but you might need to put the full path for scripts or commands installed via other package managers. Just run which cmd to find it's full path.
Since you're just renting the place, please be a good tenant and don't rm -rf anything.
Move-In
These install instructions are like my lease agreement: full of gotchas. Have your lawyer read it over. (You do have a lawyer right?)
You can install to your system with npm or use npx directly.
npm install -g @studio-mcp/studio
Unpack (it's an apartment metaphor)
Most MCPs don't run in your shell environment š You'll probably need to add the full path somewhere.
We'll use the MacOS say command as an example command. If you're not on a Mac, use echo (it's worse than useless, but it's easy to understand).
Claude Desktop
Go to the Claude Desktop settings and click Developer > Edit Config.
It should open your Claude Desktop MCP configuration. (e.g. ~/Library/Application Support/Claude/claude_desktop_config.json on Mac):
{
"mcpServers": {
"say": {
"command": "npx",
"args": [
"-y",
"@studio-mcp/studio",
"say",
"-v",
"siri",
"{speech # A concise message to say outloud}"
]
}
}
}
Cursor
Add to your ~/.cursor/mcp.json (in your home or project directory) or go to Tools section of the Cursor UI.
Note: A bug in Cursor breaks args with spaces.
{
"mcpServers": {
"say": {
"command": "npx",
"args": [
"-y",
"@studio-mcp/studio",
"say",
"-v",
"siri",
"{speech#say_outloud}"
]
}
}
}
VSCode
It's a lot of the same here. Don't get confused betwee studio and stdio (that's how I got the name).
{
"mcp": {
"servers": {
"echo": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@studio-mcp/studio",
"echo",
"{text#What do you want to say?}"
]
}
}
}
}
Template Syntax
Studio uses templates to keep your studio tidy.
studio say -v "{voice# Choose your Mac say voice}" "[args...#Any additional args]"
This creates a Studio server with two arguments: voice and args.
Everything after the # will be used as the description for the LLM to understand.
Studio templates are a modified mustache format with descriptions: {name # description} but they also add shell like optional [--flag] booleans, [optional # an optional string] and [args... # array with 0 or more strings] for additional args:
{name}: Required string argument[name]: Optional string argument[name...]: Optional array argument (spreads as multiple command line args)[--flag]: Optional boolean namedflagthat prints--flagonly when true.{name...}: Required array (1 or more arguments required).
Inside a tag, there is a name and description:
name: The argument name that will be shown in the MCP tool schema. Only letter numbers and underscores (dashes and underscores are interchangeable, case-insensitive).description: A description of what the argument should contain. Reads everything after the#to the end of the template tag.
Note: Double curly braces {{name}} are still supported for backward compatibility, but single braces {name} are preferred and used throughout the docs.
What about {cool_template_feature: string /[A-Z]+/ # Fancy tags}?
This is a simple studio, not one of those fancy 1 bedroom flats.
Maybe the landlord will get around to it at some point (but your rent will go up).
Utilities Included
To build and test locally:
make
make test
studio echo "{text # what you want said to you}"
Did something break?
The landlord definitely takes care of the place...
- more than none tests
- files! lots of 'em!
- maybe even some test coverage
- you still need Proof of Renters Insurance
Uncovered portions are tenant's responsibility. (no one appreciates how hard it is for us landlords)
Home Is Where You Make It
This is your studio too. Bugs, bedbugs, features, ideas? Swing by the repo during open-house:
š https://github.com/studio-mcp/studio
Lease Terms: MIT
Move in under standard terms, no fine print. Full text here: MIT License.
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.
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.
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.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.
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.
E2B
Using MCP to run code via e2b.