Two tools with complete input schemas and descriptions, but both descriptions are in Chinese and lack LLM-optimized English clarity. Schemas are well-formed with proper types and nested object definitions. However, descriptions are verbose (200+ chars) and lack actionable guidance on when/why to use each tool. No output schema documentation visible. No error handling guidance. Parameter descriptions exist but are minimal. Tool names are reasonably clear (analyze_*, compare_*) but lack verb-noun English convention. Missing security considerations, idempotency hints, and recovery guidance.

• Translate all descriptions to English and rewrite in LLM-optimized prose: 'Analyze an astronomy observation site by coordinates. Returns elevation, light pollution (Bortle scale), and suitability score (1-10). Use this to evaluate a single location before comparing multiple sites.' Keep under 150 chars.
• Document output schema explicitly in tool definitions. For analyze_astronomy_site, declare: {overall_score: number, elevation_info: {elevation_m: number, ...}, light_pollution_info: {bortle_scale: number, ...}, best_for: [string], limitations: [string], recommendations: [string]}
• Add error handling guidance to descriptions: 'Returns error if latitude/longitude out of range (-90..90, -180..180) or detail_level not in [basic, detailed]. If elevation/pollution services unavailable, returns partial results with available data.'
• Convert detail_level to formal enum constraint in schema instead of documenting in description text. Add description: 'Analysis depth: basic returns score + Bortle scale; detailed includes elevation profile, light pollution map, and seasonal recommendations.'
• Add toolAnnotations (readOnlyHint=true) to both tools since they only read external data and do not modify state. This signals to agents these calls are safe to retry.

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

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