Critical gaps across all dimensions. Examined src/satisfactory_mcp/server.py: tool definitions lack visible input schemas, parameter descriptions, and output documentation. No tool has a description visible in the source code excerpt. Tool names follow verb_noun convention (e.g., 'search_items', 'list_factories') but lack supporting schema definitions. The 57 tools are registered but their parameter constraints, types, and return structures are not documented in the provided code. Without access to the full implementation, schemas cannot be verified. STDIO transport limits remote usability.

• Add 50-200 character descriptions to every tool explaining WHAT it does, WHEN to use it, and any prerequisites. Example: 'search_items: Find items by name or category. Use this to discover available resources before planning production chains.'
• Document input schemas with explicit types and constraints. For 'search_items', specify: query (string, 1-100 chars), category (enum: ore|ingot|component|fluid), limit (int, 1-50, default 20).
• Document output schemas. Specify what fields each tool returns. Example: 'search_items returns [{item_id, name, category, stack_size, description}]'.
• Rename ambiguous tools: 'factory_query' → 'search_factory_state' or 'query_factory_status'; 'bom' → 'calculate_bill_of_materials'; 'stock' → 'list_inventory'.
• Add error handling guidance to destructive tools (commission_plan, plan_factory, forget_factory). Include recovery hints: 'If plan creation fails, call list_plans to verify state before retrying.'