Three tools with complete schemas and descriptions, but descriptions are generic and lack LLM-optimization guidance. All tools are READ_ONLY with proper input schemas including types and constraints (e.g., days: 1-365). However, descriptions lack WHEN/WHY context and recovery guidance. Output schemas are not documented. Error handling returns text but lacks actionable recovery steps. Tool names follow verb_noun pattern correctly.

• Add formal output schemas to all tools. Document the structure of mission objects (name, agency, status, launch_date, rocket, destination, description, live_stream_url), ISS position objects (latitude, longitude, altitude_km, velocity_kmh, orbital_period_minutes, timestamp, next_passes array), and launch objects returned.
• Expand tool descriptions to 150-250 chars with WHEN/WHY context. Example: 'Get detailed information about a specific space mission by ID. Use this to retrieve full mission status, launch date, rocket type, and destination. Call this after searching for missions or when the user asks about a specific mission by name or ID.'
• Add recovery guidance to error responses. Instead of '❌ Error fetching mission details: {error}', return structured error with suggestion: {"error": "Mission not found", "suggestion": "Try getUpcomingLaunches() to see available missions, or verify the mission ID format."}
• Add parameter descriptions for include_passes: 'Whether to include upcoming ISS passes (visible observation windows from ground stations). Set to true only if the user asks about when/where to see the ISS.'
• Declare tool annotations in schema. Add 'readOnlyHint: true' to all three tools' schema definitions to signal to LLMs that these are safe to call without confirmation.