Six tools with basic schemas and descriptions, but significant gaps in parameter documentation and output schema clarity. All tools are READ_ONLY, which is good for safety, but descriptions lack actionable context for LLM selection. Parameter descriptions are present but minimal (10-30 chars). No output schemas documented. Tool names follow verb_noun pattern correctly (get_*, create_*, analyze_*). Missing error handling guidance and recovery paths.

• Document output schemas for all tools. Example: get_services should return {services: [{name: string, id: string, tags: string[], status: 'passing'|'warning'|'critical'}], total: number}. This enables LLMs to chain calls.
• Expand parameter descriptions to 50-100 chars with context. E.g., 'If true, return only failing health checks (status=critical or warning). Use this to focus on problems; omit to see all checks.' This guides LLM selection.
• Add tool descriptions that explain when to use each tool. E.g., get_services for discovery, analyze_service for deep inspection of one service, get_service_metrics for performance data. Disambiguate overlapping tools.
• Add error handling guidance to descriptions. E.g., 'If service not found, call get_services() to list available services. If Consul is unreachable, check CONSUL_HTTP_ADDR environment variable.'
• Remove or make optional the default service_name='consul'. If a default is needed, document why 'consul' is special and what happens if it does not exist. Better: require service_name as a non-default parameter.