warpos defines 19 tools with complete JSON schemas and descriptions. However, most descriptions are terse (10-50 chars), lack actionable context, and omit parameter descriptions entirely. Tool names follow verb_noun convention but lack clarity on resource type (e.g., 'domain_profile_put' vs 'update_domain_profile'). No output schemas documented. Error handling is minimal, no recovery guidance or categorization. Security: API keys injected via env vars (good), but no audit logging or permission gates visible. Composition is reasonable (single responsibility per tool), but parameter descriptions are sparse, forcing LLMs to infer intent.

• Add 50-150 character descriptions to every parameter. Example: 'id: The profile identifier in repo-relative format (e.g., example/developer). Must not contain .. or absolute paths.'
• Document output schemas for all tools. Example for domain_profile_get: {type: object, properties: {id: {type: string}, yaml: {type: string}, updated_at: {type: string, format: date-time}}, required: [id, yaml]}
• Expand tool descriptions to 100-200 chars with WHEN/WHY context. Example: 'List all available domain profiles. Call this first to discover profiles before calling domain_profile_get. Returns profile IDs only; use domain_profile_get to fetch full YAML.'
• Add tool annotations: mark domain_profile_put, mcd_put, task_instance_generate, task_execute as destructiveHint: true; mark all read tools as readOnlyHint: true.
• Implement structured error responses with recovery hints. Example: {error: 'Profile not found: example/developer', suggestion: 'Call domain_profile_list to see available profiles', retryable: false}