Claude MCP client

Does your MCP server work in Claude?

Anthropic's AI assistant, on desktop and web. Claude renders MCP apps, and handles tool results its own way. Our catalog records the details below. Test your server against it with MCPJam.

Test against Claude View Claude capabilities

Terminal

npx @mcpjam/inspector@latest

What Claude supports

MCP apps

Claude renders MCP apps, placing tool-result media inline.

Tool-result images

The model sees images returned via direct content and embedded resources, but not via linked resources.

Display

Runs inline by default; supports inline and fullscreen on web.

MCP extensions

Advertises io.modelcontextprotocol/ui.

Emulation model

MCPJam emulates Claude using anthropic/claude-haiku-4.5, so tests reflect how this host actually drives your tools.

How we know

Assumed: Inferred defaults, not yet verified against the client. Last reviewed 2026-07-08.

Test your server against Claude

Test an MCP app in Claude

Walkthrough for test an MCP app's UI in Claude.

Test MCP tools in Claude

Walkthrough for test MCP tools and their results in Claude.

Connect a server to Claude

Walkthrough for connect an MCP server to Claude and verify it.

Claude capability questions

Does Claude support sampling?

Live support matrix for MCP sampling, per client.

Does Claude support elicitation?

Live support matrix for MCP elicitation, per client.

Does Claude support roots?

Live support matrix for MCP roots, per client.

Compare MCP host capabilities side by side

The full Can I Use matrix on caniuse.dev.

How to test your MCP server against Claude

Run npx @mcpjam/inspector@latest, connect the server config you already use, and select Claude. MCPJam drives your server the way this host does, including its model (anthropic/claude-haiku-4.5), so you see the tool calls it actually makes instead of guessing from a bug report.

From there, Trace shows every step from prompt to response, the OAuth Debugger shows where authorization breaks, and Evals score whether the model picks the right tool across realistic requests. All of it runs locally, free and open source.

Frequently asked questions

Yes. Our catalog records Claude as rendering MCP apps, with tool-result media placed inline. (Assumed, last reviewed 2026-07-08.)

The model sees images returned via direct content and embedded resources, but not via linked resources.

Run npx @mcpjam/inspector@latest, connect your server, and select Claude. MCPJam emulates the host with anthropic/claude-haiku-4.5 so you can watch the real tool calls, trace them, and score them with evals.

It comes from MCPJam's host-compat catalog (the same data behind caniuse.dev), recorded as "assumed" and last reviewed 2026-07-08. Inferred defaults, not yet verified against the client.