AILIENANT has 17 tools with significant definition gaps. Most tools lack input schemas entirely (read_file, write_file, run_command, task_create, task_get, task_list, task_stop, security_audit, validate_ast, sandbox_bash, check_type_integrity, validate_wbs_dependencies, estimate_plan_budget, ask_user_question, toggle_plan_mode have no visible schema). Only 3 tools (read_file, write_file, todo_write, ask_user_question, tool_search) show partial schemas in the provided code. Descriptions are present but generic (10-50 chars for most). No parameter descriptions visible for most tools. No output schemas documented. Error handling and recovery guidance absent. Tool names follow verb_noun convention (positive), but lack of schemas and descriptions severely limits LLM usability.

• Add complete JSON Schema input definitions for all 12 tools missing schemas. Include type, description, and constraints (enum, min/max, pattern) for every parameter.
• Expand tool descriptions to 100-200 characters. Include: WHAT it does, WHEN to use it (vs similar tools), and WHAT it returns. E.g., 'Execute a shell command in the sandboxed workspace. Use for build, test, or file operations. Returns stdout, stderr, and exit code.'
• Document output schemas for all tools. Specify field names, types, and whether results are paginated. E.g., task_create should return {task_id: string, status: string, created_at: ISO8601}.
• Add error handling guidance to destructive tools. E.g., write_file: 'If write fails due to permissions, error will include required permissions. If path does not exist, parent directories will be created.'
• Consolidate run_command and sandbox_bash into a single tool with a 'sandbox' parameter (boolean), or clearly differentiate them in descriptions (e.g., 'run_command uses host environment; sandbox_bash uses isolated container').