Server has 7 tools with complete input schemas and descriptions present, but quality is inconsistent. Tool names follow verb_noun convention (good), but descriptions are generic and lack actionable context. Parameters have types and descriptions, but lack constraints (enums, ranges, formats). No output schemas documented. Error handling is minimal, generic 'Error: {message}' responses provide no recovery guidance. Missing parameter validation, idempotency hints, and dependency documentation. Baseline: avg tool description ~150 chars (acceptable), but descriptions lack WHEN/WHY context. No tool annotations (readOnlyHint, destructiveHint). ancestry_login exposes credentials via environment variables (correct pattern), but no audit trail or permission gates documented.

• Add output schemas to all tools. Document the structure of search results (e.g., array of {name, profileUrl, birthYear, deathYear}), person details (e.g., {name, birthDate, deathDate, locations, events}), and tree view (e.g., {generations, ancestors, descendants}).
• Expand descriptions with WHEN/WHY context and dependency hints. E.g., ancestry_get_person_details: 'Fetch full profile details (birth/death dates, locations, family connections). Call after ancestry_search_person to get the profileUrl. Returns structured person object with timeline events.'
• Add parameter constraints: birthYear/deathYear as YYYY format with range 1500 - 2100; generations as number 1 - 10; profileUrl as URL pattern; personData.events as non-empty array.
• Implement error classification and recovery guidance. E.g., 'Login failed: credentials invalid. Check ANCESTRY_USERNAME and ANCESTRY_PASSWORD env vars.' or 'Profile not found. Try ancestry_search_person to find the correct URL.'
• Add tool annotations: ancestry_login with destructiveHint=true; all read tools with readOnlyHint=true. This signals agent safety constraints.