Single tool 'run_python' has a reasonable description (127 chars, within 10-1024 range) but critical gaps in schema completeness, parameter documentation, and error handling. Input schema exists but lacks formal type/description structure in JSON Schema format. No output schema documented. No error handling guidance, no input validation, no security considerations for arbitrary code execution. Tool name 'run_python' is generic and does not follow verb_noun pattern (should be 'execute_python' or 'run_code'). Missing parameter constraints, format specifications, and recovery guidance.

• Add formal JSON Schema output definition documenting return type (string), structure (e.g. {"output": string, "error": string, "execution_time_ms": number}), and example.
• Enhance parameter schema: add 'type': 'string' and 'description' field to 'code' parameter in JSON Schema. Document format constraints (e.g. 'Valid Python 3.11+ syntax') and examples.
• Implement error handling with categorization: return structured errors distinguishing syntax errors (user-fixable), timeouts (retryable), and runtime exceptions (fatal). Include recovery hints.
• Add input validation: enforce code length limits (e.g. max 10KB), execution timeout (e.g. 30s), and blacklist dangerous imports (e.g. os.system, subprocess). Document these constraints in description.
• Add confirmation/dry-run pattern for destructive operations: detect file deletion, network calls, or system commands and require explicit confirmation before execution.