mcp-geo has 4 tools with basic descriptions and minimal parameter documentation. All tools accept a single 'url' parameter with minimal type info. Descriptions are present but generic (10-60 chars), below the 50-200 char LLM-optimized baseline. No output schemas documented. No error handling guidance. No parameter validation rules or constraints. Tools lack verb-noun naming clarity (e.g., 'check_llms_txt' vs 'validate_llms_txt'). No idempotency hints, no permission gates, no audit trail. Composition is reasonable (single responsibility per tool), but parameter descriptions are trivial.

• Document output schemas for each tool. Example for check_llms_txt: {type: object, properties: {'/llms.txt': {type: object, properties: {exists: boolean, size: number, status: number}}}}. This enables LLMs to extract and chain results.
• Expand parameter descriptions to 50-150 chars with format and constraint hints. Example: 'url (string, required): Full URL starting with http:// or https://. Must be accessible and return HTTP 200. Used to fetch and validate GEO metadata.'
• Add error recovery guidance. When fetch fails, return: {error: 'URL not accessible (HTTP 404). Verify the domain is correct and the page exists. Try a simpler URL like https://example.com'}.
• Standardize verb naming: rename 'check_llms_txt' → 'validate_llms_txt', 'audit_robots_for_ai' → 'check_robots_txt_for_ai', 'validate_schema' → 'extract_schema_markup'. Use consistent verb_noun pattern.
• Add tool descriptions explaining when to use each. Example for check_llms_txt: 'Checks if a domain publishes an llms.txt file per llmstxt.org spec. Use this to verify AI crawler access policies. Returns file existence, size, and HTTP status.'