Skip to main content

MCP Inspector

The MCP Inspector is an interactive developer tool for testing and debugging MCP servers.

Adapted from MCP Inspector.

MCP Inspector — Source: MCP docs

Source: modelcontextprotocol/docs (mcp-inspector.png).

Getting started​

npx @modelcontextprotocol/inspector <command>
npx @modelcontextprotocol/inspector <command> <arg1> <arg2>

npm / PyPI packages​

npx -y @modelcontextprotocol/inspector npx @modelcontextprotocol/server-filesystem /Users/username/Desktop
npx @modelcontextprotocol/inspector uvx mcp-server-git --repository ~/code/mcp/servers.git

Local servers​

npx @modelcontextprotocol/inspector node path/to/server/index.js args...
npx @modelcontextprotocol/inspector \
uv \
--directory path/to/server \
run \
package-name \
args...

Always read the server README for the authoritative launch command.

Feature overview​

AreaWhat you can do
Server connectionChoose transport; customise args and env for local servers
ResourcesList, inspect MIME/metadata, read content, test subscriptions
PromptsList templates, fill arguments, preview messages
ToolsInspect schemas, call with custom inputs, view results
NotificationsView server logs and protocol notifications

Development workflow​

  1. Launch Inspector with your server.
  2. Verify connectivity and capability negotiation.
  3. Change server code → rebuild → reconnect → retest.
  4. Probe edge cases: invalid inputs, missing prompt args, concurrent calls, error payloads.

Negative cases​

  • Wrong working directory → resources/files not found; set --directory / cwd deliberately.
  • Missing env vars → tool fails only in Inspector if host injects different env; align both.
  • Testing only happy paths → ship brittle servers; add invalid schema tests.

Discussion

Comments​

Share feedback or questions about this page. No account required.

Loading comments…