inspector MCP — Use Cases, Install & Live Demo

Name: inspector MCP Server
Author: MCPJam

Why use it

Key features

Full JSON-RPC tracing across host/server
Inline LLM chat while inspecting tool calls and context
Evaluation runs across multiple models
OAuth flow debugging with conformance checks
Workspace sharing, CLI, SDK, and CI/CD integration

Live Demo

What it looks like in practice

inspector.replay ▶ ready

0/0

Install

Pick your client

~/Library/Application Support/Claude/claude_desktop_config.json · Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "inspector": {
      "command": "npx",
      "args": [
        "-y",
        "inspector"
      ],
      "_inferred": true
    }
  }
}

Open Claude Desktop → Settings → Developer → Edit Config. Restart after saving.

~/.cursor/mcp.json · .cursor/mcp.json

{
  "mcpServers": {
    "inspector": {
      "command": "npx",
      "args": [
        "-y",
        "inspector"
      ],
      "_inferred": true
    }
  }
}

Cursor uses the same mcpServers schema as Claude Desktop. Project config wins over global.

VS Code → Cline → MCP Servers → Edit

{
  "mcpServers": {
    "inspector": {
      "command": "npx",
      "args": [
        "-y",
        "inspector"
      ],
      "_inferred": true
    }
  }
}

Click the MCP Servers icon in the Cline sidebar, then "Edit Configuration".

~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "inspector": {
      "command": "npx",
      "args": [
        "-y",
        "inspector"
      ],
      "_inferred": true
    }
  }
}

Same shape as Claude Desktop. Restart Windsurf to pick up changes.

~/.continue/config.json

{
  "mcpServers": [
    {
      "name": "inspector",
      "command": "npx",
      "args": [
        "-y",
        "inspector"
      ]
    }
  ]
}

Continue uses an array of server objects rather than a map.

~/.config/zed/settings.json

{
  "context_servers": {
    "inspector": {
      "command": {
        "path": "npx",
        "args": [
          "-y",
          "inspector"
        ]
      }
    }
  }
}

Add to context_servers. Zed hot-reloads on save.

claude mcp add inspector -- npx -y inspector

One-liner. Verify with claude mcp list. Remove with claude mcp remove.

Use Cases

Real-world ways to use inspector

How to debug why your MCP tool isn't being called

👤 MCP server developers ⏱ ~15 min intermediate

When to use: You wrote a tool, Claude doesn't use it, and you don't know why.

Prerequisites

Node — nodejs.org or nvm

Flow

Launch Inspector

npx @mcpjam/inspector@latest and open the printed URL.✓ Copied

→ UI opens in browser
Connect your server

In the UI, add your server (stdio or HTTP) and connect.✓ Copied

→ Tools list populates
Simulate a prompt

In the Chat tab, ask a question that should trigger your tool. Watch the JSON-RPC trace.✓ Copied

→ Either tool fires (good) or you see why the model rejected it (bad description, missing input schema)

Outcome: Concrete reason (wrong tool description, schema issue, model choice) with a visible fix.

Pitfalls

Testing with a weaker model that never uses tools — Test across multiple models in the Evals tab — Haiku, Sonnet, GPT-4o — behavior varies

How to run eval tests on PRs for your MCP server

👤 Teams maintaining production MCPs ⏱ ~30 min advanced

When to use: You want to catch 'oops that prompt no longer picks my tool' in CI.

Flow

Author evals

In Inspector, create eval cases: (prompt, expected_tool_called, expected_args).✓ Copied

→ Eval suite saved
Wire into CI

Use the CLI to run evals on every PR; fail the build on regression.✓ Copied

→ .github/workflows/mcp-eval.yml runs evals

Outcome: PRs that break tool selection don't merge.

Pitfalls

Evals are flaky when temperature > 0 — Use temperature 0 in eval runs

Combine with: github

How to diagnose a broken OAuth flow on your MCP server

👤 MCP server devs shipping OAuth-protected tools ⏱ ~20 min advanced

When to use: Tokens aren't refreshing, or redirect returns 400.

Flow

Enable OAuth debug in Inspector

Configure your server's OAuth endpoints and click 'Test flow'.✓ Copied

→ Step-by-step trace of authorize, token, refresh
Identify the break

Inspector highlights nonconforming steps (wrong content-type, missing PKCE, bad scope).✓ Copied

→ Specific failure node

Outcome: Conforming OAuth implementation.

Combinations

Pair with other MCPs for X10 leverage

inspector + github

Run Inspector evals in a GitHub Action on PRs for your MCP server

Set up a GitHub Action using @mcpjam/inspector CLI to run evals on every PR touching src/tools/.✓ Copied

Tools

What this MCP exposes

Tool	Inputs	When to call	Cost
(inspector app) connect-server	transport: stdio\|http, command or url	First step	free
(inspector app) execute-tool	tool, args	Manual testing	depends on tool
(inspector app) chat-and-inspect	model, prompt	End-to-end simulation	1 LLM call
(inspector app) run-evals	suite_id	CI regression testing	N LLM calls

Cost & Limits

What this costs to run

API quota: Your LLM/eval keys determine the budget
Tokens per call: Evals can burn tokens — limit concurrency + size
Monetary: Inspector itself is free. Hosted app at app.mcpjam.com has tiers.
Tip: Run evals on Haiku for cheap smoke tests; reserve Sonnet/Opus for nightly runs.

Security

Permissions, secrets, blast radius

Minimum scopes: Whatever scopes your MCP server needs LLM provider keys for evals

Credential storage: Local env for CLI; workspace secrets for the hosted app

Data egress: Tool calls stay between your MCP and LLM; hosted app stores eval results in your workspace

Never grant: Don't upload production OAuth client secrets to the hosted app without review

Inspector can call write tools during eval runs — use a sandbox server, not production.

Troubleshooting

Common errors and fixes

Can't connect to my stdio server

Make sure the command Inspector runs matches what Claude Desktop runs — working dir and env vars included.

Chat tab says 'model not configured'

Add your Anthropic/OpenAI key in Settings.

Evals pass locally, fail in CI

Env mismatch — pin the same model + temperature in CI as locally.

OAuth test reports 'state mismatch'

Your server isn't preserving the state param; compare to the spec in Inspector's conformance report.

Alternatives

inspector vs others

Alternative	When to use it instead	Tradeoff
@modelcontextprotocol/inspector (official)	You want the first-party reference inspector	Smaller feature surface; no evals/CI
mcp-client-for-ollama	You want a terminal-only Ollama-backed client	Not a debugger, just a client

More

Resources

📖 Read the official README on GitHub

🐙 Browse open issues

🔍 Browse all 400+ MCP servers and Skills