Openstatus MCP Health Checker - Test MCP servers like a real AI client, not just a ping
Most monitors just send an HTTP ping. But a 200 OK is useless if the JSON-RPC handshake fails. Our tool is different because it performs a true protocol-level check, acting exactly like a real AI client.
Key features:
Full Handshake: Executes the spec-defined initialize, ping, and tools/list sequence.
Deep Visibility: Inspect exact JSON-RPC payloads and negotiated versions.
Smart Auth: Parses RFC 9728 headers on 401s to surface exact token requirements.


Replies
openstatus
@tibozaurus Many congrats on the launch. I have a quick question: when the tools/list returns empty, does the checker differentiate between "tools not implemented" VS "tools filtered by auth/scope?"
RiteKit Company Logo API
@tibozaurus This is a smart solution to a real problem in the MCP ecosystem. The distinction you're making between HTTP-level success and actual protocol-level functionality is exactly what catches people off guard when deploying agents. The JSON-RPC inspection view sounds particularly useful for debugging those mysterious "tool not found" failures.
when you say "like a real ai client" does it go through initialize → list-tools → actually call a tool, or stop at the handshake? in my experience the gap between "server responds to initialize" and "tools actually execute correctly" is where most bugs hide
openstatus
@trekh First, we initialize the tools list. We still need to refine the final tool execution step to ensure it works perfectly. This will also allow us to monitor if a tool is taking longer than usual to respond.
@tibozaurus This is a useful direction for MCP reliability. A simple 200 OK doesn’t mean an agent can actually use the tool, so testing the real handshake and tools/list flow feels much closer to what production agent systems need.
openstatus
@alpertayfurr yeah 200 doesnt mean anything for mcp servers,
@tibozaurus Exactly. For MCP, the real question is whether the agent can discover the tools, understand the schema, call them correctly, and recover when something fails. That’s a very different health check than “is the server alive?”
AISA AI Skills Test
smart approach testing MCP servers as a real client instead of just pinging endpoints. most of the reliability issues i've seen with MCP integrations come from edge cases in the actual tool call flow, not connectivity. curious if you're also testing for things like response format consistency across different server implementations?
openstatus
@ozandag 200 is useless for mcp 😂
timing on this is interesting because MCP server quality is all over the place right now. a lot of servers were built quickly against early drafts of the spec and there's no standard way to know if something is actually compliant until a client breaks on it. is there any plan to expose a compliance score or badge system so developers can signal to potential users that their server passed a real protocol check rather than just an uptime ping
openstatus
@ansari_adin 100 uptime ping does not make much sense for MCP server
minimi
Launched our MCP server 2 days ago - Open status comes at the right time.
Tabstack by Mozilla
oss ftw!