LinuxLens has 20 tools with consistent naming (verb_noun pattern) and reasonable descriptions (avg ~100 chars). All tools have input schemas with types. However, parameter descriptions are sparse or missing entirely, most params lack guidance on valid ranges, formats, or constraints. Output schemas are undocumented. Error handling is absent from descriptions. No tool annotations (readOnlyHint/destructiveHint). The simulation tools (14-20) are well-scoped but lack depth in parameter validation guidance. Observation tools (1-13) are clearer but still need richer descriptions and output documentation.

• Document output schemas for all tools. Example: get_memory_info returns {total_kb: int, available_kb: int, used_kb: int, buffers_kb: int, cached_kb: int}. This enables LLM planning and composition.
• Add parameter constraints to descriptions. Example: 'pid (integer, 1 - 2147483647): Process ID to query. Use get_process_tree() to discover valid PIDs.' This prevents invalid inputs and guides multi-step planning.
• Add tool annotations via FastMCP. Mark observation tools with readOnlyHint=true, delete_trace with destructiveHint=true, and idempotent tools with idempotentHint=true. This helps agents reason about side effects.
• Enrich error descriptions with recovery guidance. Example: 'If trace_id not found, call list_traces() to see active traces.' This enables self-correction without extra agent reasoning.
• Document preset configurations. For simulate_*_presets tools, add a description listing available presets (e.g., 'light', 'medium', 'heavy') and their characteristics. Or add an optional 'preset_name' parameter with enum values.