All 13 tools have basic descriptions (10-50 chars) and simple input schemas with a 'message' parameter. However, descriptions are generic and lack actionable context for LLM selection. Most tools lack output schema documentation. Parameter descriptions are minimal. No error handling guidance, no enum constraints, and no indication of what each tool returns or when to use it instead of similar tools. Tools like elicit_enum have an 'options' parameter but lack detail on defaults and validation. No tool annotations (readOnlyHint, destructiveHint, idempotentHint) despite all being READ_ONLY. Naming is consistent (elicit_*) but descriptions do not explain the semantic difference between elicit_text, elicit_number, elicit_integer, etc., or when an LLM should choose one over another.

• Add output_schema to all tools documenting the structure of elicitation results (e.g., {"type": "object", "properties": {"value": {"type": "string"}}, "required": ["value"]}).
• Expand tool descriptions to 100-200 chars explaining semantic differences: 'Elicit a plain text string (no format validation). Use for free-form input; use elicit_email for email-specific validation, elicit_uri for URLs.'
• Add tool annotations: set readOnlyHint=true for all tools since they do not modify state.
• Document elicit_enum 'options' parameter more clearly: specify that it is optional, explain default behavior, and clarify the expected array format.
• Add error handling guidance in descriptions: 'Requires client to declare elicitation capability. If unsupported, the tool will return an error; ask the user to upgrade their client.'