Free tools Windows power users keep installed
One-click scans. No signup required.
The message “Couldn’t reach the MCP server” does not identify the cause. First find the earliest failed step: public network access, the documented MCP endpoint, OAuth discovery or authorization, or the authenticated MCP request. A protected endpoint that returns HTTP 401 may be reachable but rejecting a request without valid credentials; that response alone does not prove the service is down.
What the error does—and does not—tell you
The exact message reported in one WordPress MCP connection issue was: “Couldn’t reach the MCP server. You can check the server URL and verify the server is running. If this persists, share this reference with support.” That report concerned a self-hosted WordPress site connected to Claude, using WordPress MCP plugin version 0.2.5 with MCP, Create Tools, and Update Tools enabled. The reporter configured a JWT and listed https://shop.mydomain.co.uk/wp-json/wp/v2/wpmcp/streamable as the endpoint; opening it directly returned HTTP 401 with a JSON unauthorized response. The issue remains open, and no maintainer-confirmed cause or fix is posted, so the reporter’s theory that the client did not pass the token is not established. Read the issue.
Interpret the result you observe rather than treating every failure as “server down.” A 401 differs from DNS failure, a timeout, 404, 403, or an upstream server error. Its meaning depends on the integration’s authentication requirements and documented behavior.
Identify your WordPress MCP integration first
WordPress MCP setups are not interchangeable. The WordPress MCP Adapter project describes its role as bridging the Abilities API to the Model Context Protocol so clients can discover and invoke abilities provided by WordPress plugins, themes, and core. That description does not establish that every WordPress MCP plugin uses the adapter, the same routes, or the same authentication flow. See the WordPress MCP Adapter project.
Quick wins for a faster PC:
Scan for outdated or missing drivers - takes under a minuteDriver Scan →Clear out junk files and repair common Windows errorsFree Scan →Before changing settings, record the details needed to reproduce the failure:
- The MCP plugin or adapter name and version, and the MCP client and version.
- The exact configured URL, copied from the integration’s own documentation or settings.
- Whether the client connects from the same machine or network as WordPress, or remotely.
- The authentication method and where that integration expects credentials to be supplied.
- The full error text and the point in the connection flow when it appears.
Debug the connection in order
1. Request the exact documented endpoint
Use the endpoint specified for your particular integration; do not infer it from another plugin or copy a route from a different WordPress MCP setup. Request that exact URL independently and note the HTTP status, response body, and headers. Compare the result with the integration’s documented unauthenticated and authenticated behavior. A 401 can indicate that the request reached a protected route but did not satisfy its authentication requirement; it is not by itself proof that the endpoint is unavailable.
2. Verify public reachability from outside the WordPress host
If the client is remote, check that the hostname resolves publicly and that the relevant HTTPS URL can be reached from a network outside the WordPress server’s local environment. A URL that opens in an administrator’s browser does not necessarily reproduce the route, network, or request made by a remote MCP client. A local-only hostname cannot be assumed reachable from an external service.
3. Check OAuth discovery only when your integration uses OAuth
For an OAuth-based connection, identify whether the failure occurs during metadata discovery, dynamic client registration, browser consent, token exchange, or the first authenticated MCP call. Meow Apps’ guide for its AI Engine setup recommends testing both path-suffixed and host-root .well-known discovery URLs. Those routes and that sequence are specific to its documented setup, not universal WordPress MCP paths; use the URLs documented by your own integration. Consult the AI Engine troubleshooting guide.
4. Compare edge behavior with WordPress logs
Watch the relevant PHP and web-server logs while reproducing the same connection attempt. If a request receives 403 or 404 and does not appear in application logs, investigate the host, CDN, WAF, routing, or request-filtering path. If it reaches WordPress, use the integration’s available logs to determine whether registration, consent, token exchange, or the authenticated MCP request failed. These are diagnostic possibilities, not evidence that a particular firewall or host caused the error.
5. Change one relevant setting, then repeat the same check
Once the first failing stage is clear, change a setting related to that stage and repeat the same request. Keeping the request and observations consistent gives you a useful before-and-after comparison and avoids attributing the result to an unrelated change.
How to interpret common results
| Observed result | What it establishes | Next check |
|---|---|---|
| DNS failure or timeout | The test did not establish successful access to the endpoint; the specific cause is not identified by that result alone. | Test public DNS and HTTPS reachability from outside the WordPress host, then check host or network routing. |
| HTTP 401 | The request received an authentication-related response. It does not, on its own, establish whether the endpoint is healthy for a correctly authenticated MCP request. | Check the integration’s documented authentication flow and whether credentials are supplied as expected. |
| HTTP 403 or 404 with no application log entry | The request may have been rejected or routed before reaching WordPress; the status alone does not identify which edge component handled it. | Compare host, CDN, WAF, and web-server behavior while reproducing the request. |
| Discovery succeeds, but consent, token exchange, or the first MCP call fails | The connection progressed beyond discovery; the later failing stage needs its own diagnosis. | Correlate the client attempt with application and server logs for that stage. |
Choosing a WordPress connector or custom connector
There is no evidence-based universal rule to choose a WordPress-branded connector over a custom connector, or vice versa. Compare the options for your specific setup by checking whether the client is local or remote, which endpoint and discovery paths the integration documents, how and where it expects authentication, whether requests reach the WordPress application, and what diagnostics it exposes. The GitHub report raises this connector-choice question but does not resolve it; the right choice depends on the actual plugin and client documentation.
Quick Recap
Best Value
- easy to use
- Free app
- Compatible with all devices
- It gives the best comparison between ten different hosts
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
Recommended Free Tools
