
- Monthly visits
- 2,593
- 2026-09
- Monthly growth +878.5%
Multimeter - Git-native API requests and tests for VS Code and CI
Multimeter is a Git-native REST client and API test runner for VS Code. It stores requests, tests, suites, mocks, and API documentation as YAML .mmt files in a repository, then runs the same files locally or in CI with testlight. It also supports AI-assisted test generation and response judging through configured models.
- AI Coding
- Test Automation
- Coding Assistants
- VS Code API testing
- Git-native REST client
- YAML API test runner




Capabilities
YAML API requests
Create and send API requests from editable YAML .mmt files in the VS Code editor. Define request details such as URL, method, headers, and body, and inspect the response.
Reusable API test flows
Define multi-step API tests and suites as YAML .mmt files, with calls, assertions, loops, branches, and parallel groups. Run the same files in VS Code and CI using testlight.
HTTP and WebSocket mocks
Run local HTTP and WebSocket mocks defined in .mmt files. Configure routes, status codes, dynamic responses, reflect mode, and template variables for development and integration tests.
AI test generation and judging
Generate or refine tests from a description or OpenAPI spec, and score non-deterministic responses for relevance, similarity, faithfulness, and factuality. Use Cursor, Copilot, or Claude through MCP and configure a model for judge steps.
Generated API documentation
Generate API documentation from .mmt API files and export it as HTML or Markdown. Documentation can include interactive examples with dark and light themes.
Pricing
Multimeter is free and the site states there are no subscriptions. Choose it based on the included API request, test, mock, documentation, and CI workflows; no paid plan is listed.
| Plan | Price | Included features and limits |
|---|---|---|
| Free | Free; no subscription |
|
Use cases
Send and inspect API requests
When developing or troubleshooting an API, define a request in a YAML .mmt file and send it from the VS Code editor. Inspect the returned status and response while keeping the request in Git alongside the project.
Run repeatable API tests in CI
When an API needs checks in development and a pipeline, define calls and assertions in .mmt test flows and suites. Run those same files locally or with testlight in CI, and export supported reports for pipeline gates.
Mock API endpoints for development
When client work or integration tests need predictable endpoints, define HTTP or WebSocket routes and responses in server .mmt files. Start the built-in mock server locally from suites, tests, or the VS Code panel.
Generate and evaluate API tests with AI
When drafting tests from an endpoint description or OpenAPI spec, use an MCP-connected agent to generate .mmt flows, then review and run them. For variable model responses, add a judge step using a model you configure alongside deterministic assertions.
Limitations
AI judge steps require a model provider that the user configures; they are not a substitute for deterministic status, JSON, or assertion checks.
Load testing is identified as beta; the feature reuses .mmt scenarios and provides threads, ramp-up, and load-oriented reports.
Who it is for
API developers who want request files and tests versioned in Git
QA engineers who need repeatable API test flows in VS Code and CI
Developers building integrations that need local HTTP or WebSocket mocks
FAQ
What is Multimeter?
Multimeter is a Git-native REST client and API test runner for VS Code and Cursor. Requests, tests, suites, mocks, and docs are YAML .mmt files in a repository; testlight runs the same files in CI. It is Apache 2.0 licensed and does not require an account.
Is Multimeter free, and do I need an account?
Yes. The product is free under the Apache License 2.0, and an account is not required. The site states there are no subscriptions or feature gates.
Can I import existing API definitions and tests?
Yes. Multimeter can import or convert Postman collections, OpenAPI or Swagger, WSDL, Bruno, .http files, and curl commands. It can also open .http and Bruno files directly.
How do I run the same tests in CI?
Use the testlight CLI with a suite .mmt file, such as npx testlight run path/to/suite.mmt. The site describes use with GitHub Actions, GitLab CI, Jenkins, and Azure Pipelines, with JUnit XML, HTML, or Markdown reports.
How does Multimeter use AI?
MCP-connected agents such as Cursor, Copilot, and Claude can scaffold, validate, format, and run .mmt files. AI can draft tests from a description or OpenAPI spec; judge steps score non-deterministic responses using a model you choose.
Comments