// free tool
What does this MCP server expose?
Point it at a remote MCP endpoint. See the tools, resources and prompts it advertises — and which of them can delete data, run commands, move money or read credentials.
Questions
Does it call the tools it finds?
No — deliberately. It completes the handshake and lists capabilities, nothing more. Executing a stranger's tool from our infrastructure would be both a security problem and a way to cause real side effects on someone else's system.
Which transports work?
Streamable HTTP, including servers that answer over SSE rather than plain JSON, and those that issue a session id on initialize. stdio servers can't be inspected remotely — there's nothing to connect to over the network.
How is risk decided?
Heuristics over each tool's name and description: command execution, deletion, payments and credential access are high; writes, outbound messages, filesystem and infrastructure changes are medium; everything else is low. It's a prompt to look, not a verdict — a tool flagged high risk may be exactly what the server is for.
It says authentication required. Now what?
Many MCP servers gate everything behind OAuth or an API key, so there's nothing public to inspect. That's correct behaviour on their part — you'd need to connect with credentials from a real client instead.
Why did it fail with 405 or 404?
405 means that URL doesn't accept POST, so it isn't a Streamable HTTP MCP endpoint. 404 usually means the path is wrong — the endpoint is often at /mcp rather than the site root.
// ready_when_you_are
Let's build your agentic advantage
Tell us what should be automated, connected or sold. We'll map the fastest path from idea to a production system.