Four tools with basic schemas and descriptions, but significant gaps in parameter documentation and output structure. Tool names follow verb_noun convention (health_check, list_models, get_current_model, chat_completion). Descriptions exist but are generic and lack LLM-optimization guidance. The chat_completion tool has the most detailed description and schema, but parameters lack constraint documentation (e.g., temperature range 0.0-1.0 stated in description but not in schema; max_tokens default 1024 with no bounds). No output schemas documented for any tool. Error handling and recovery guidance absent. No tool annotations (readOnlyHint, destructiveHint, idempotentHint). Parameter descriptions are minimal or missing for several fields.

• Add output schemas for all tools. Document the structure of responses (e.g., health_check returns {status: string, message: string}, list_models returns {models: [{name: string, size: string, ...}], total: number}).
• Expand tool descriptions to 100-200 characters. Include WHEN to use (e.g., 'Call this before chat_completion to verify LM Studio is running'), prerequisites, and what the tool returns.
• Add min/max constraints to numeric parameters in schema. For chat_completion: temperature type number with minimum 0.0, maximum 1.0; max_tokens with minimum 1, maximum 32768.
• Add error recovery guidance to descriptions. E.g., 'If health_check fails, ensure LM Studio is running on localhost:1234. Retry after starting the service.'
• Declare tool annotations in schema. Add readOnlyHint: true to all four tools to signal they are safe to retry and do not modify state.

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

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