Strong schema definitions with comprehensive parameter descriptions and proper JSON Schema typing. All 12 tools have clear, action-oriented names (verb_noun pattern) and detailed descriptions (100-300 chars). Input schemas are well-structured with type constraints, enums, patterns, and min/max bounds. However, output schemas are not documented, LLMs cannot predict response structure for downstream chaining. Error handling guidance is absent. No tool annotations (readOnlyHint, destructiveHint, idempotentHint) despite all tools being read-only. Parameter descriptions are excellent (detailed constraints, examples of valid values), but lack explicit guidance on when to use each tool vs. similar ones (e.g., yahoo_search_free_agents vs. yahoo_get_waivers distinction).

• Document output schemas for all tools. Include example responses showing field names, types, and nesting. Ensure responses include IDs (league_id, team_id, player_id) needed for downstream tool calls.
• Add readOnlyHint annotation to all 12 tools in the tool registration to signal to LLMs that these are safe, non-destructive calls.
• Enhance tool descriptions with recovery hints: 'If league_id is invalid, call yahoo_list_my_leagues to discover valid leagues. If player not found, try yahoo_search_free_agents with a partial name.'
• Clarify the distinction between yahoo_search_free_agents (FA status) and yahoo_get_waivers (W status) in both tool descriptions. Add a note: 'Use yahoo_search_free_agents for unrestricted free agents; use yahoo_get_waivers for players pending waiver clearance.'
• Document pagination behavior explicitly. If count is a hard limit with no cursor/offset, state 'Returns up to N results; no pagination available.' If pagination is supported, add offset/cursor parameters and document the response structure (total_count, next_cursor).

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

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