Protocol-level handshake testing
Runs the same connection sequence an MCP client uses on first contact, including initialize, notifications/initialized, and tools/list.
OpenStatus MCP Server Health Check
OpenStatus MCP Server Health Check is a browser-based tool for validating an MCP endpoint’s JSON-RPC handshake, tool exposure, and authentication behavior. It helps operators and developers quickly spot protocol errors, missing headers, and unreachable servers without installing anything.
Browser-based MCP protocol health check
MCP Server Health Check is a browser-based tool from OpenStatus for testing whether a Model Context Protocol endpoint responds the way an AI client expects. Instead of checking only for an HTTP 200 response, it performs the JSON-RPC handshake used by MCP clients and verifies the server’s initialize, ping, and tools/list behavior.
The tool is aimed at people who run or integrate MCP servers and need to confirm that an endpoint is reachable, speaks the protocol correctly, and exposes tools as expected. It can also surface authentication issues, protocol mismatches, and common deployment mistakes such as returning HTML from the wrong URL or requiring headers that were not supplied.
What the MCP health check does
Header support for authenticated servers
Accepts a server URL plus optional request headers, including Authorization for protected endpoints.
Step-by-step inspection
Shows the exact JSON-RPC request and response for each step, along with per-step latency, negotiated protocol version, server info, and session id.
Readable outcome states
Returns clear verdicts for healthy servers, partial MCP implementations, auth failures, and unreachable endpoints.
Browser-based request flow
Sends Content-Type: application/json and Accept: application/json, text/event-stream on every request, matching the page’s documented request behavior.
OAuth-aware error handling
Surfaces auth challenges by parsing WWW-Authenticate bearer metadata when available, which helps users identify the authorization server involved.
Common ways to use it
Validate a fresh deployment
Check whether a newly deployed MCP endpoint is speaking the expected protocol before pointing clients at it.
Diagnose auth-protected servers
Debug authentication errors by adding the required bearer token or examining a 401 challenge response.
Verify tool availability
Confirm that an endpoint advertised to agents actually exposes tools and not just a reachable URL.
Troubleshoot misconfigured servers
Investigate failures caused by the wrong endpoint, HTML responses, missing methods, or session problems.
Preflight manual checks
Use the health check as a quick browser-based sanity check before moving to deeper monitoring or client testing.
Pros and Cons
Pros
- Checks the MCP handshake rather than just HTTP status, which is more useful for AI clients.
- Shows request and response details for each step, making failures easier to diagnose.
- Supports optional headers for protected servers.
- Recognizes common failure modes such as wrong content type, missing methods, authentication problems, and unreachable endpoints.
- Works in the browser with no install and no signup.
Cons
- It is a point-in-time check, not continuous monitoring; the page points users to a separate monitoring guide for ongoing checks.
- The tool depends on the endpoint being reachable from the browser session and on any required headers being provided correctly.
- The page is focused on MCP handshake validation, so it does not replace broader infrastructure or application monitoring.
FAQ
What does the MCP Server Health Check verify?
It verifies that an MCP endpoint speaks the JSON-RPC handshake expected by clients such as Claude Desktop, Claude Code, Cursor, and Continue. The tool runs initialize, notifications/initialized, and tools/list so you can see whether the server is reachable, authenticates correctly, and advertises tools.
How do I test my MCP server?
Paste a Streamable HTTP MCP server URL into the form, optionally add an Authorization header if the server requires one, and run the check. The page then shows each step of the handshake, including the request and response details and the verdict.
What results can the health check return?
The page reports Healthy, Partial, Authentication Required, or Unreachable. Partial means the server speaks MCP but exposes no tools; Authentication Required means the server returned 401; Unreachable covers DNS, connection, or parse failures.
Does the check run locally or require installation?
It is a browser-based tool with no install and no signup. The page says the whole flow takes about 30 seconds and that the only data persisted is the report you choose to share.
Is this tool free to use?
OpenStatus says paid plans are available monthly or annually, and the pricing page also lists a free Hobby plan for uptime monitoring and a status page. The MCP health check page itself is presented as free online access.
Quick Facts
- Category
- Developer Tool
- Product
- OpenStatus MCP Server Health Check
- Platform
- Browser-based web tool
- Primary use
- Validate MCP JSON-RPC handshake and tool exposure
- Related OpenStatus products
- Uptime monitoring and hosted status pages
- Website
- openstatus.dev
OpenStatus MCP Server Health Checkの代替品
FixMyCWV
FixMyCWV is a web-based Core Web Vitals audit tool that produces developer-focused recommendations for improving LCP, INP, and CLS. It works across many site stacks and claims to handle sites with bot protection.
Ably Chat
Ably Chat is a chat API platform for building custom realtime chat applications. It supports room-based messaging, typing indicators, presence, reactions, and message updates, with usage-based pricing options for different deployment stages.
Ghost
Ghostは、チャット、コード生成、コマンドラインでのタスク実行に対応したターミナル向けAIアシスタントです。無料モデルを同梱し、Linux、macOS、Windowsで使えます。オープンソースです。
AakarDev AI
AakarDev AI helps teams manage AI provider access, project-level setups, logs, and analytics from one dashboard. It supports BYOK workflows and lists providers including OpenAI, Google Gemini, Anthropic, Groq, Mistral AI, and Perplexity AI.
DeepMotion
DeepMotion is a web-based AI motion capture and 3D animation platform with Animate 3D for video-to-animation and SayMotion for text-to-animation. It helps creators and teams generate motion in a browser and export results in common production formats.
Arduino VENTUNO Q
Arduino VENTUNO Q is an edge AI computer for AI and robotics applications. It combines AI inference and deterministic control on a single board and is designed to work with Arduino App Lab.