Critical gaps across naming, descriptions, and schemas. Tool names lack consistent verb prefixes (unity_search_assets vs unity_get_asset vs unity_execute_csharp). Descriptions are present but generic (10-50 chars for most tools, well below the 194-char baseline). Parameter descriptions exist but lack constraints, ranges, and format specifications. No output schemas documented. The unity_execute_csharp tool is a severe security and composition violation, it bypasses all tool boundaries and accepts arbitrary code execution. Most tools lack error handling guidance. Schemas are partially visible but incomplete (missing type info for nested objects like 'target' in unity_scene_view).

• Remove unity_execute_csharp entirely. Replace with specific, constrained tools for common editor tasks (e.g., unity_run_editor_script accepting only whitelisted script names, or unity_invoke_menu_item with enum validation).
• Add enums to all parameters with known valid values. E.g., 'command' in unity_editor_command should be an enum: ['play', 'pause', 'stop', 'step', 'save_scene', 'save_all', 'undo', 'redo', 'refresh_assets', 'compile']. Same for 'asset_type', 'target' (build platforms), 'filter_type'.
• Expand descriptions to 100-200 chars. Include WHEN to use the tool, WHAT it returns, and any prerequisites. E.g., 'Get asset metadata and properties by path. Use this to inspect an asset's serialized fields before modifying. Returns asset type, name, and properties up to the specified depth (1-5).'
• Document output schemas for every tool. Specify which fields are returned, their types, and which are safe to pass to downstream tools. E.g., unity_search_assets should document: returns array of {path: string, type: string, name: string, instance_id: string}.
• Add error recovery guidance. E.g., 'If asset not found, try unity_search_assets() with a partial name.' Categorize errors as retryable (network timeout) vs. user-fixable (invalid path) vs. fatal (permission denied).