Server has 8 well-structured tools with complete input schemas and descriptions. Naming follows verb_noun convention (endpoint_add, endpoint_list, webhook_send). Most parameters are typed and described. However, several parameter descriptions lack actionable detail (e.g., 'Extra headers', 'Extra metadata' are vague). Output schemas are not documented, LLMs cannot predict response structure. Error handling guidance is absent. Tool descriptions are adequate (50-150 chars) but could better explain when to use each tool vs. alternatives. No tool annotations (readOnlyHint, destructiveHint) despite clear risk levels (DESTRUCTIVE for endpoint_delete, WRITE for endpoint_add).

• Add output schema documentation to all tools. Example for endpoint_add: 'Returns {endpoint_id: string, name: string, url: string, status: string, created_at: ISO8601, ...}'. This enables LLMs to chain tools and extract required fields.
• Enhance parameter descriptions with format and constraint details. Replace 'Extra headers' with 'Custom HTTP headers as key-value pairs (e.g., {"Authorization": "Bearer token"}). Merged with endpoint defaults.'
• Add tool annotations: mark endpoint_delete with destructiveHint=true, endpoint_list/endpoint_get/delivery_list with readOnlyHint=true, webhook_send with idempotentHint=true (if idempotency_key is provided).
• Include recovery hints in tool descriptions. E.g., endpoint_get: 'If endpoint not found, call endpoint_list to discover available endpoints or check the endpoint_id spelling.'
• Clarify delivery_list endpoint_id parameter: 'Optional. If provided, returns deliveries for that endpoint only. If omitted, returns deliveries for all endpoints (paginated). Defaults to all endpoints.'