Three tools with basic schemas and descriptions, but significant gaps in parameter documentation and output schema clarity. scan_site and scan_page have reasonable descriptions (60-80 chars), but lack detail on output structure, error cases, and parameter constraints. list_rules has minimal description (47 chars). No parameter descriptions visible for 'depth' beyond a brief note. Output schemas not documented, LLMs cannot plan downstream operations or extract specific fields. No error handling guidance. Tool names follow verb_noun pattern (good), but descriptions lack WHEN/WHY context and actionable recovery paths.

• Document output schema for scan_site and scan_page. Specify: returns {pages: [{url, score, issues: [{rule, severity, message, fix}]}], summary: {totalIssues, byCategory}}. Include example structure so LLMs can extract fields.
• Add parameter constraints to 'depth': 'depth (integer, 1 - 100, default 10): Maximum pages to crawl. Higher values increase runtime; use 10 - 20 for small sites, 50+ for large sites. Timeout after 5 min regardless of depth.'
• Expand list_rules description to 100 - 150 chars: 'List all SEO rule IDs and descriptions that seomapper checks. Call this first to understand available rules, then use rule IDs to filter scan results or ignore specific checks.'
• Add error handling to all tool descriptions: 'Returns error if URL is invalid, unreachable, or times out after 30s. Partial results returned if some pages fail; check 'errors' array in response.'
• Include per-tool WHEN guidance: scan_site: 'Use to audit entire site structure and cross-page issues (duplicates, missing sitemaps).'; scan_page: 'Use for single-page audits or when you only have one URL.'