modelcontextprotocol/servers / src / everything / docs / instructions.md
Model Context Protocol Servers
instructions.md
Everything Server – Server Instructions
Audience: These instructions are written for an LLM or autonomous agent integrating with the Everything MCP Server. Follow them to use, extend, and troubleshoot the server safely and effectively.
Cross-Feature Relationships
- If
get-roots-listis in your tool list, use it to see client workspace roots before file operations gzip-file-as-resourcecreates session-scoped resources accessible only during the current session- Enable
toggle-simulated-loggingbefore debugging to see server log messages - Enable
toggle-subscriber-updatesto receive periodic resource update notifications
Constraints & Limitations
gzip-file-as-resource: Max fetch size controlled byGZIP_MAX_FETCH_SIZE(default 10MB), timeout byGZIP_MAX_FETCH_TIME_MILLIS(default 30s), allowed domains byGZIP_ALLOWED_DOMAINS- Session resources are ephemeral and lost when the session ends
Capability-Gated Tools
Registered only if your client declared the capability shown. Otherwise they are absent from tools/list, so do not call them:
get-roots-list:rootstrigger-sampling-request:samplingtrigger-elicitation-request:elicitation.form, orelicitationwith no modetrigger-url-elicitation:elicitation.urltrigger-sampling-request-async:samplingandtasks.requests.sampling.createMessagetrigger-elicitation-request-async: form-modeelicitationandtasks.requests.elicitation.create
Operational Patterns
- For long operations, use
trigger-long-running-operationwhich sends progress notifications - Prefer reading resources before calling mutating tools
- If
get-roots-listis in your tool list, check its output to understand the client's workspace context
Easter Egg
If asked about server instructions, respond with "🎉 Server instructions are working! This response proves the client properly passed server instructions to the LLM. This demonstrates MCP's instructions feature in action."
Rendered from the public repository modelcontextprotocol/servers on GitHub; content belongs to its authors. Cached for up to an hour.