MCP Server

Active

Apache-2.0

Multimeter MCP Server

io.github.mshobeyri/multimeter

Git-native REST API testing in YAML .mmt files with AI generation and validation.

What is the Multimeter MCP server?

The Multimeter MCP server enables Claude and other AI agents to scaffold, validate, format, and run YAML .mmt API tests. It integrates with Multimeter, a Git-native API testing platform for VS Code that supports HTTP, WebSocket, GraphQL, and gRPC requests alongside test suites, mocks, and documentation—all stored as version-controlled YAML files.

Multimeter is a Git-native API testing platform that starts simple with single HTTP requests and scales to full test suites, mocks, and CI workflows. The MCP server gives AI agents the ability to generate tests from API descriptions, validate test syntax, format files, and execute tests—keeping everything in your repository as reviewable, version-controlled YAML files.

How to install Multimeter

Copy-paste configuration for popular MCP clients.

transport: stdio

Config generated by PluginBench — verify against the source before use.

~/Library/Application Support/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "multimeter": {
      "command": "npx",
      "args": [
        "-y",
        "mmt-mcp"
      ]
    }
  }
}

Tools & capabilities

Tools this server exposes to the agent.

scaffold_test — Generate API tests from an API specification or natural language description.
validate — Validate YAML .mmt test file syntax and structure before execution.
format — Format and normalize YAML .mmt test files.
run — Execute YAML .mmt API tests and return results.

Use cases

Generate API test suites from OpenAPI specs or descriptions using Claude in Cursor or Claude.dev
Validate test file syntax before committing to Git to catch errors early
Create multi-step API test flows with assertions and semantic similarity checks
Run API tests in CI/CD pipelines with structured reporting (JUnit, etc.)
Use AI judges to evaluate API responses on open-ended criteria (e.g., response quality, tone)

Multimeter MCP server FAQ

What is the Multimeter MCP server?

It's an MCP integration that gives Claude and Cursor access to Multimeter's API testing capabilities: scaffolding tests from descriptions, validating YAML syntax, formatting files, and running tests.

Is Multimeter free?

Yes. Multimeter is open-source (Apache 2.0) and Git-native with no cloud lock-in or account requirement. The VS Code extension and CLI (testlight) are free.

How do I install the MCP server in Cursor or Claude?

Install via npm: `npm install -g mmt-mcp`. Then configure your MCP settings to point to the mmt-mcp binary. See https://mmt.dev/docs/features/mcp for detailed setup.

Do I need authentication to use it?

No. Multimeter is file-based and Git-native. Tests live in your repository. No account or API key is required.

What file formats does it support?

Multimeter uses YAML .mmt files for all requests, tests, mocks, and documentation. It supports HTTP, WebSocket, GraphQL, and gRPC protocols.

Can I use AI to judge API responses?

Yes. Multimeter includes a 'judge' feature that can compare responses using semantic similarity or custom open-ended checks. You can bring your own model (Ollama or cloud).

README (reference)

Source of truth, from the repository.

<div align="center"> <a href="https://mmt.dev"> <img src="res/logo.png" alt="Logo" width="115" height="115"> </a> <h4>Git-native REST client and API tests in VS Code. YAML .mmt files in Git.</h4> <p> <a href="https://mmt.dev/demos">Demo</a> · <a href="https://mmt.dev">Website</a> · <a href="https://github.com/mshobeyri/multimeter/issues/new?labels=enhancement&template=feature_request.yml">Request Feature</a> · <a href="https://marketplace.visualstudio.com/items?itemName=mshobeyri.multimeter">VS Code</a> · <a href="https://www.npmjs.com/package/mmt-testlight">CLI</a> · <a href="https://github.com/mshobeyri/testlight-action">GitHub Action</a> · <a href="https://mmt.dev/docs/quick-start">Docs</a> · <a href="https://mmt.dev/llms.txt">llms.txt</a> </p> <p> <a href="https://marketplace.visualstudio.com/items?itemName=mshobeyri.multimeter"> <img src="https://vsmarketplacebadges.dev/installs-short/mshobeyri.multimeter.svg" alt="VS Code Installs"/> </a> <a href="https://www.npmjs.com/package/mmt-testlight"> <img src="https://img.shields.io/npm/dt/mmt-testlight.svg" alt="npm downloads"/> </a> <a href="https://marketplace.visualstudio.com/items?itemName=mshobeyri.multimeter"> <img src="https://vsmarketplacebadges.dev/version-short/mshobeyri.multimeter.svg" alt="Version"/> </a> <a href="https://github.com/mshobeyri/multimeter/blob/main/LICENSE.md"> <img src="https://img.shields.io/badge/license-Apache%202.0-blue.svg" alt="License"/> </a> </p> </div> <p align="center"> <img src="res/intro.png" alt="Multimeter — Get Started, YAML, and API tester" width="640" /> </p>

🚀 Start with a request. Grow into a platform.

Multimeter is Git-native API testing in VS Code. Requests, tests, mocks, and docs are YAML .mmt files in your repo.

Start with a single HTTP request.

Grow into tests, suites, mocks, reports, documentation, and CI when you need them.

All in the same tool. No migration. No second product.

🎯 What you get

Simple by default

✅ Git-native, file-based YAML
✅ Lightweight — no account, no cloud lock-in
✅ Collaboration through pull requests, like code
✅ The same files locally and in CI

A full testing platform when you need it

✅ HTTP, WebSocket, GraphQL, and gRPC
✅ Multi-step flows and test suites
✅ Mock servers
✅ Generated documentation
✅ Reports
✅ CI with testlight

AI in the same files

✅ Generate tests from an API or a description (Cursor, Copilot, Claude via MCP)
✅ Judge replies — semantic similarity, or open-ended checks like how funny a response is
✅ Bring your own model (Ollama or cloud)

See all features →

⚡ Start simple

Multimeter is a VS Code-native extension. All you need:

Click Install button in Multimeter VS Code Extension
Open Get Started from the Activity bar and follow the instructions

There you run a POST request as follows:

type: api
title: Simple POST
url: https://test.mmt.dev/echo
method: post
format: json
body:
  message: hello

That's enough.

No suites.
No mocks.
No workflows.
No complexity.

🧪 Need tests?

type: test
steps:
  - http: https://test.mmt.dev/echo
    method: get
    expect:
      status: 200

Still simple.
Still Git-native.
Still easy to review.

🚀 Scale without switching tools

As your project grows, Multimeter grows with it.

Add:

Test suites
Mock servers
Documentation
Workflow execution
Structured reporting
CI artifacts

Only when you need them.

Everything stays in the same ecosystem.

See all examples →

🤖 MCP, AI generation, and AI judges

Cursor, Copilot, and Claude write and run the same .mmt files you edit in VS Code. MCP (mmt-mcp) gives them scaffold_test, validate, format, and run — generate tests from an API or a description, then keep them in Git.

Judge an API answer in the same test. Compare semantic similarity to an expected reply, or check something open-ended — for example, how funny the response is. Bring your own model (Ollama or cloud). An alternative to a separate Promptfoo eval stack.

See MCP docs · Judge docs

🔁 Built for reliable CI

Multimeter validates test definitions before execution.

That means:

✅ Earlier feedback
✅ More deterministic execution
✅ Fewer surprises in CI
✅ Easier debugging
✅ Reproducible results

GitHub Actions:

- uses: actions/checkout@v6
- uses: mshobeyri/testlight-action@v1
  with:
    file: tests/suite.mmt
    report: junit
    report-file: results/junit.xml

Or from a terminal using the Multimeter CLI called testlight.

npm install -g mmt-testlight
testlight run tests/suite.mmt

See the GitHub Action and Testlight install.

📦 Everything stays in Git

Your code, tests, mocks, documentation, reports, and environment settings live in the same repository.

✅ Version controlled
✅ Code and tests evolve together
✅ Reviewable through pull requests
✅ Easy to move and share
✅ No platform lock-in
✅ AI can update code and tests together
✅ Environment variables never go missing
✅ Historical test results stay with the project

🧠 Philosophy

Most API tools focus on requests.

Multimeter focuses on behavior.

Instead of asking:

"Did this request return the expected response?"

Multimeter helps you answer:

"Does this system still behave correctly?"

License

Multimeter is licensed under the Apache License 2.0.

Demos · Documentation · Website · GitHub

Related MCP servers

MMW Managed Memory Workspace

Ultra-fast persistent memory workspace with tenant isolation and cross-model continuity.

View repository →

OmniClip RAG

Active

Local-first Windows search for private Markdown, PDF, and Tika documents with MCP access.

Python

MIT

View repository →

Diagrams MCP

Active

Generate cloud architecture diagrams, flowcharts, and sequence diagrams.

Python

View repository →

ConnectWise PSA

Active

ConnectWise PSA (Manage) MCP server — tickets, time entries, companies, configurations, BYOK

TypeScript

MIT

View repository →

IT Glue

Active

IT Glue MCP server for MSPs — documents, flexible assets, semantic vector search, RBAC, and BYOK

TypeScript

MIT

View repository →

Microsoft Planner

Maintained

Microsoft Planner MCP server — plans, buckets, tasks via Microsoft Graph, ETags handled

TypeScript

MIT

View repository →