{"path":"/guide/getting-started/connect-agent/troubleshooting","content":{"title":"Troubleshooting your Everything connection","summary":"Visitor-friendly help with MCP setup, missing tools, sign-in prompts and public search results.","breadcrumbs":"[All topics](/guide) / [About Everything](/topics/about-everything) / [Getting started](/guide/getting-started) / [Connect your agent](/guide/getting-started/connect-agent)","body":"If your AI client has trouble connecting to Everything, start with the public reading checks below. No Everything account or password is required.\n\n## I entered the URL in a browser and saw an error\n\nThe MCP endpoint `https://everything.emlynoregan.com/mcp/` is a protocol endpoint, not a web page. A browser GET is not a valid test of every MCP operation; the client normally negotiates protocol messages over HTTP. Use [First tests](/guide/getting-started/connect-agent/first-steps) or visit the human-facing [Everything homepage](/guide).\n\nUse only the address shown in the [connection guides](/guide/getting-started/connect-agent).\n\n## The MCP server won't appear\n\n1. Confirm you used `https://everything.emlynoregan.com/mcp/`, including the `/mcp/` path.\n2. Ensure the client supports **remote Streamable HTTP** MCP servers rather than only local `stdio`.\n3. Inspect the client's server list, permissions and connection-error message. Refresh/restart the client if necessary.\n4. Verify the configuration format: Cursor uses `mcpServers`, VS Code uses `servers`, Codex uses `mcp_servers` in TOML.\n5. In ChatGPT, custom connection availability depends on plan/workspace permissions; in Claude, the connector may need to be enabled for the chat.\n\n## The server appears, but my agent ignores it\n\nAsk explicitly for Everything MCP's `get_page`, name a known path and ask the model to report the actual returned title. Some agents prefer their own knowledge or web search without a specific tool request. Verify tool invocation rather than accepting a plausible summary.\n\n## The client asks me to sign in\n\nPublished pages can be read anonymously. Select **No sign in** or **No authentication** if available. If your client requires a login and cannot offer anonymous access, that client may not support this public connection mode. Do not enter an unrelated account password.\n\n## Can I edit articles?\n\nNo. Visitor access is read-only, even when a client shows names of tools that are not available to the public.\n\n## Requests are slow or time out\n\nInitial requests may involve cold application or storage dependencies. Avoid rapid repeated searches. Retry a **read** deliberately, or read a known page by route instead. Client/network/protocol overhead is distinct from storage time.\n\n\n## A page exists but does not appear in search\n\nDirect `get_page` by known route is the right verification. Indexes and related-result metadata can lag publication. Incomplete or partial search coverage must not be interpreted as proof of absence. Try the known page URL directly instead of relying on search results.\n\n## Where to go next\n\n- [Connection guides](/guide/getting-started/connect-agent) — choose your AI client and set up the server\n- [Check the connection](/guide/getting-started/connect-agent/first-steps) — verify a genuine page read\n- [Access and privacy](/guide/getting-started/connect-agent/access-and-privacy) — understand the public read-only connection\n\n[Browse Everything](/guide)","navigation":"[Connect your agent](/guide/getting-started/connect-agent) · [First tests](/guide/getting-started/connect-agent/first-steps)","related":[{"title":"Connection guides","url":"/guide/getting-started/connect-agent"},{"title":"Check the connection","url":"/guide/getting-started/connect-agent/first-steps"},{"title":"Access and privacy","url":"/guide/getting-started/connect-agent/access-and-privacy"}]}}