Veil Mail MCP has well-structured tool definitions with consistent naming (verb_noun pattern) and complete input schemas. All 10 tools are properly registered with JSON Schema. However, descriptions are generic and lack LLM-optimization guidance. Parameter descriptions are minimal (often single phrases). No output schemas are documented, forcing LLMs to guess response structure. Error handling is absent from tool definitions. Missing tool annotations (readOnlyHint, destructiveHint). The send_email and add_subscriber tools lack confirmation/dry-run patterns despite being irreversible operations.

• Add output schemas to all tools. Document the structure of responses (e.g., send_email returns {id, status, scheduledFor, createdAt}). This enables LLMs to chain tools correctly.
• Enhance descriptions to 80-150 chars. Include WHEN to use the tool and what it returns. Example: 'Send an email via Veil Mail. Returns email ID and delivery status. Use after validating recipient with validate_email.'
• Add tool annotations: mark send_email and add_subscriber with destructiveHint=true; mark all read-only tools with readOnlyHint=true.
• Implement dry-run or confirmation for send_email. Add optional 'dryRun' parameter that returns preview without sending, or require explicit 'confirm' parameter.
• Add pagination to list_emails: include 'offset' and 'limit' parameters (default limit=20, max=100), and return 'total' count in response.