Five tools with reasonable naming (verb_noun pattern) and documented schemas, but critical gaps in parameter descriptions and output documentation. Tool descriptions are present but generic (avg ~100 chars, below 194-char baseline). Most parameters lack descriptions entirely, e.g., nlp2cmd_resolve_uri has 'system_map' and 'doql_path' with no guidance on structure or purpose. No output schemas documented for any tool. Error handling is absent, no guidance on recovery, retryability, or actionable error messages. Security concerns: nlp2cmd_resolve_uri and nlp2cmd_uri_plan are marked WRITE but lack confirmation/dry-run patterns. Tool composition is reasonable (each does one thing), but missing chaining IDs in responses and pagination support.

• Add output schema documentation for each tool. Example: nlp2cmd_analyze returns {intent: string, target_kind: string, confidence: number (0-1), execution_plan_ir?: object}. Document all fields LLMs need to chain calls.
• Add descriptions to all parameters. For 'system_map', explain: 'SystemMapIR object from env2llm (dict with keys: system_name, uri_patterns, examples). Required if platform-specific resolution needed.' For 'doql_path', explain: 'Path to DOQL query file for custom URI resolution logic.'
• Add dry-run parameter to WRITE tools (nlp2cmd_resolve_uri, nlp2cmd_uri_plan): 'dry_run: boolean (default false). If true, return planned actions without executing them.' Enables safe preview.
• Enhance tool descriptions with WHEN/WHY. Example: 'Use nlp2cmd_analyze for intent classification only. Use nlp2cmd_plan for full NL→DSL pipeline (requires NLP2CMD_INTEGRATION=1). Use nlp2cmd_resolve_uri to map NL to system URIs (e.g., open app, take screenshot).'
• Add error handling guidance in descriptions. Example: 'If nlp2uri is not installed, returns error code MISSING_DEPENDENCY with message: Install via pip install nlp2uri[envmap]. If system_map is invalid, returns INVALID_CONFIG with details on missing keys.'