Three tools with clear, action-oriented names (forge_runbook, analyze_gaps, get_template) and solid descriptions (100 - 150 chars each). All tools have input schemas with type declarations. However, parameter descriptions are minimal or absent: 'notes' param lacks guidance on expected format/length; output schemas are not formally documented in the tool definitions; no error handling guidance or recovery paths. Schemas are present but sparse, no constraints, enums, or validation hints. Tool composition is sound (each does one thing), but descriptions could be more LLM-optimized per the 50 - 200 char baseline.

• Add parameter description to 'notes': 'Rough incident/ops notes as plain text or markdown (max 10,000 chars). Include symptoms, diagnosis, resolution steps, and any known gaps marked with TODO.'
• Document output schemas in tool descriptions. E.g., 'Returns a dict with keys: title (str), completeness (0 - 100 int), gaps (list of {section, severity, reason}).'
• Add error handling guidance: 'If notes are empty or malformed, returns a partial result with completeness=0 and gaps listing all required sections. Retry with more detailed notes.'
• Distinguish forge_runbook vs analyze_gaps in descriptions: 'forge_runbook generates markdown; analyze_gaps returns only the gap analysis. Call analyze_gaps first to assess completeness before forging.'
• Add constraints to 'notes' parameter: min 50 chars, max 10,000 chars, plain text or markdown. Helps LLMs validate input before calling.