Skip to content
MCP server

tinyprobe

API tests in YAML that coding agents write and a 9 MB native binary runs in CI

Runs locallyStatus and history
Pricing
Open source
Platforms
CLI, Linux, macOS
Runs
Locally
Version
0.25.0
Listing updated
2 days ago

About tinyprobe

tinyprobe is an open-source MCP server for developer tools. It runs locally on your machine and is started by your AI client.

Tinyprobe has no importers. Instead, a built-in guide (tinyprobe guide) tells your coding agent how to convert a Postman collection, an OpenAPI document or backend routes into YAML test files in your repository. The agent validates each file, tries requests and runs workflows until they pass, then writes MIGRATION.md listing what it could not convert. On the project's Postman benchmark of 20 requests with 4 planted unconvertible cases, Claude Code with Sonnet and with Haiku covered all 18 convertible requests and reported all 4 untranslatable ones.

Files are validated as a whole before any request is sent, with file:line:column errors and did-you-mean hints. Endpoints are declared in API files and workflows run them, with values passed between steps through save, so each workflow starts from zero. A failing check shows the actual value.

It ships as one Native AOT binary of about 9 MB that starts in about 20 ms, plus a GitHub Action, Docker image, JUnit reports, a VS Code extension and a small app. It is free and open source under GPL-3.0, with no account or cloud.

Install

Run in a terminal

claude mcp add tinyprobe -- docker run -i --rm ghcr.io/kikutano/tinyprobe:0.25.0

Add to claude_desktop_config.json

{
  "mcpServers": {
    "tinyprobe": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "ghcr.io/kikutano/tinyprobe:0.25.0"
      ]
    }
  }
}
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "tinyprobe": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "ghcr.io/kikutano/tinyprobe:0.25.0"
      ]
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "tinyprobe": {
      "type": "stdio",
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "ghcr.io/kikutano/tinyprobe:0.25.0"
      ]
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "tinyprobe": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "ghcr.io/kikutano/tinyprobe:0.25.0"
      ]
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "tinyprobe": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "ghcr.io/kikutano/tinyprobe:0.25.0"
      ]
    }
  }
}

What it does

Agent-written tests
tinyprobe init adds a section to AGENTS.md pointing agents to the built-in guide.
Validation before requests
Lists every problem with file:line:column and a did-you-mean suggestion.
Explicit workflows
Endpoints are declared once; workflows run them and save values step to step.
CI-ready binary
One about 9 MB native binary, GitHub Action, Docker image and JUnit output.
Same files everywhere
Terminal, VS Code, a desktop app and MCP all use the same YAML.
Migration report
MIGRATION.md records what was converted and what had no equivalent.

Good for

  • Convert a Postman collection into tests kept in the repository
  • Generate API tests from an OpenAPI document and run them against a local backend
  • Gate pull requests on API checks with annotated failing steps
  • Find where a backend differs from its documented responses

Pricing

Open source

Free, GPL-3.0

tinyprobe FAQ

Does it need an account or cloud?

No. Everything runs locally from one binary.

What licence is it under?

GPL-3.0, free and open source.

How does it migrate from Postman?

Your own coding agent follows the built-in guide to convert the collection and checks each file with inspect, call and run.

How do I install it?

A curl script on Linux and macOS, a PowerShell script on Windows, or the ghcr.io/kikutano/tinyprobe Docker image.

Can it run in CI?

Yes, with a GitHub Action that annotates failing steps and JUnit reports for other CI servers.

Is tinyprobe free?

Yes. tinyprobe is open source, so you can use and self-host it for free.

How do I add tinyprobe to Claude Code?

Run this command in your terminal: claude mcp add tinyprobe -- docker run -i --rm ghcr.io/kikutano/tinyprobe:0.25.0

What are the best alternatives to tinyprobe?

The closest alternatives to tinyprobe are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.

Alternatives to tinyprobe

  1. REA

    MCP server

    CLI and MCP server that lets coding agents reverse engineer apps and binaries

    Runs locally
    • Open source
    • 464K GitHub stars
    • Developer tools
  2. Codebase Memory

    MCP server

    Local MCP server that builds a code knowledge graph for AI coding agents

    Runs locally
    • Open source
    • 48.7K GitHub stars
    • Knowledge and memory
  3. GitHub

    MCP server

    Official MCP server connecting AI assistants to GitHub repos, issues and workflows

    Not checked yet
    • Remote
    • Open source
    • 34.5K GitHub stars
    • Developer tools
  4. MCP toolkit giving coding agents symbol-level code retrieval and editing

    Runs locally
    • Open source
    • 31.5K GitHub stars
    • Developer tools
  5. XERJ

    MCP server

    Local search engine for AI agents with cited answers over your code and docs

    Runs locally
    • Open source
    • 19.2K GitHub stars
    • Knowledge and memory
  6. Ouroboros

    MCP server

    Spec-first layer that pins requirements and verifies results of AI coding agent work

    Runs locally
    • Open source
    • 6.4K GitHub stars
    • Developer tools