Three tools with comprehensive parameter schemas and detailed descriptions. All tools have explicit JSON Schema definitions with type information and parameter descriptions. Tool names follow verb_noun pattern (loot_evaluate, loot_load_order, loot_plugin_metadata). Descriptions are detailed (200+ chars) and explain WHAT, WHEN, and HOW. However, output schemas are not documented in the visible code, only input schemas are defined. Error handling is minimal (no recovery guidance or error categorization). No tool annotations (readOnlyHint, etc.) despite all being read-only operations.

• Add formal output schema definitions to each tool. Document the structure of load_order_current (array of plugin names?), load_order_suggested, general_messages (array of message objects with severity?), and plugins map (dict of plugin_name → metadata_yaml string?).
• Add readOnlyHint: true to all three tool definitions in the MCP server registration to signal that these tools are safe to retry and have no side effects.
• Enhance error handling: document common failure modes (game not found, plugins.txt missing, invalid game_type) and provide recovery guidance. E.g., 'If game_path is invalid, verify the directory contains the game executable.'
• Simplify parameter descriptions for LLM clarity. Replace 'libloot, suggested load order, general messages (default)' with 'Returns LOOT's recommended plugin load order and any warnings or errors.'
• Add min/max constraints to numeric pagination parameters: plugin_metadata_offset ≥ 0, plugin_metadata_limit ≥ 1 and ≤ 1000 (or appropriate max).