AutoMaya has 5 tools with basic naming and descriptions, but critical gaps in schema documentation and parameter descriptions. Tool names follow verb_noun convention (maya_get_status, maya_execute_python), which is good. However, only maya_execute_python shows an explicit input schema in the provided code. The other 4 tools lack visible parameter schemas, making it impossible to verify LLM-safe input constraints. Descriptions are present but brief (10-50 chars), below the 50-200 char baseline for LLM optimization. No output schemas documented. Error handling and recovery guidance absent. The maya_execute_python tool is particularly risky: it accepts arbitrary Python code with no validation, sanitization, or confirmation step despite being marked IRREVERSIBLE.

• Add explicit input schemas to all 5 tools with typed parameters and descriptions. Example: maya_get_status should document what fields it returns (status, scene_name, etc.) and their types.
• Expand tool descriptions to 50-200 characters following LLM-optimized format: 'WHAT it does, WHEN to use it, WHAT it returns.' E.g., 'Get current Maya scene status including node count, file path, and unsaved changes. Call before executing commands to verify scene state.'
• Add confirmation-request pattern to maya_execute_python: require a dry-run preview or explicit confirmation before executing arbitrary code. Return a summary of what will execute and ask user to approve.
• Document output schemas for all tools. Specify return type (object/array), field names, types, and which fields are needed for chaining to other tools (e.g., if maya_scene_summary returns node IDs, maya_execute_python should accept them).
• Implement error handling with recovery guidance: categorize errors as retryable vs fatal, include the invalid value and constraint violated, and suggest next steps (e.g., 'Maya not running. Start Maya and retry.' or 'Invalid Python syntax. Check code and retry.').