Two tools with minimal descriptions and incomplete schemas. 'echo' is a test utility with a trivial 18-char description. 'dust-query' has a 67-char description but lacks output schema documentation. Neither tool has error handling guidance, recovery hints, or parameter constraints. No tool annotations (readOnlyHint, destructiveHint, idempotentHint) present. Parameter descriptions are present but generic. No pagination, no result limits documented. Schemas visible but lack depth (no output types, no constraints on sessionId format). This is a basic proof-of-concept server, not production-grade.

• Expand 'echo' description to 100-150 chars: explain it is a diagnostic/testing tool, when to use it (e.g., connection validation), and what it returns (echoed message + metadata).
• Expand 'dust-query' description to 150-200 chars: explain it sends messages to Dust AI, when to use it vs other tools, what sessionId does (conversation context), and what response structure to expect.
• Add explicit output schemas for both tools. Define return types: echo should return {message: string, timestamp: string}; dust-query should return {response: string, sessionId: string, confidence?: number}.
• Add parameter constraints: sessionId should document format (UUID? alphanumeric?), whether it is required, and what happens if omitted (new session created?).
• Add error handling guidance: document what happens if Dust API is unreachable, if sessionId is invalid, or if message is too long. Provide recovery hints (retry, check API key, truncate message).

Inferred effective spec: <=2025-11-25.

• Relies on Logging (deprecated) - log to stderr or use OpenTelemetry