All 5 tools have zero-parameter schemas (empty input objects) and descriptions that, while present, are extremely long (200-300+ chars) and marketing-focused rather than agent-optimized. No parameter descriptions exist because there are no parameters. Tools are read-only discovery endpoints with no error handling guidance, no output schema documentation, and no composition support. This is a catalog/info server, not an agent-actionable toolkit.

• Rewrite descriptions to 50-150 chars, agent-optimized: 'get_services' → 'Fetch turva.dev service catalog: pricing, duration, deliverables for audits, implementation, advisory.' Remove marketing copy.
• Document output schemas for each tool. E.g., get_services returns {services: [{id, name, price, duration, summary, deliverable}], bundled_implementation: [...]}. This enables LLM planning.
• Add error handling guidance: 'If this tool fails, the service catalog is unavailable. Suggest the user visit turva.dev directly or retry in 30 seconds.'
• Consider adding a search_services(keyword: string) tool to enable agents to filter by service type (e.g., 'find audit services'). Current get_services returns everything.
• Add a verify_agent_readiness(domain: string) tool to allow agents to check readiness for arbitrary domains, not just turva.dev. This increases utility and composition.