ClaudeMap

MCPJam Inspector

An open-source testing and evaluation platform for MCP servers — Postman for the MCP ecosystem.

SDKs & Toolingtestingmcpevaluationdebugging
Visit projectMCPJam/inspector

MCPJam Inspector lets you connect to MCP servers over any transport, call and debug tools, run multi-server setups, and drive LLM evaluations against your server — in an open-source local UI or their hosted app. Complements the official MCP Inspector with testing workflows and evals the official tool doesn't cover.

What it does

  • Connect over stdio, Streamable HTTP, and SSE; manage multi-server workspaces
  • Chat with real models against your servers to test tool-calling end to end
  • LLM evaluation runs for regression-testing server changes
  • Open source locally (npx) or hosted at app.mcpjam.com

When to use it

Best for server authors shipping iteratively: smoke-test each release, then pin eval sets so refactors don't silently change tool behaviour.

Getting started

npx @mcpjam/inspector@latest

# or Docker:
docker run -p 127.0.0.1:6274:6274 mcpjam/mcp-inspector:local

Things to know

A third-party platform, not the official MCP tooling — spec coverage can run ahead of or behind specific clients. Keep the official Inspector in your toolbox for protocol-level checks.

Repository

MCPJam/inspector

Charted

Related on the map

  • E2B

    SDK / Tool

    Open-source secure sandboxes where AI agents run code — the de facto execution layer for agent apps.

    • sandbox
    • code-execution
    • sdk
    • agents
    View

  • FastMCP

    SDK / Tool

    Pythonic framework for building MCP servers — decorators, type hints, and zero boilerplate.

    • framework
    • python
    • mcp
    • type-driven
    View

  • Claude Agent SDK

    SDK / Tool

    Anthropic's official TypeScript SDK for building agentic applications.

    • official
    • sdk
    • typescript
    • anthropic
    View