Three tools with basic schemas and descriptions, but significant gaps in parameter documentation and output schema clarity. Tool names follow verb_noun convention (scrape_website, extract_text, batch_scrape), which is good. However, descriptions lack depth on WHEN to use each tool vs. alternatives, and parameter descriptions are minimal. No output schemas documented. Error handling is present but generic ('Error: {message}'). The batch_scrape tool lacks a description for the 'mode' parameter entirely.

• Add explicit output schema documentation for each tool. Example for scrape_website: 'Returns object with fields: title (string), links (array of {url, text}), headlines (array of strings), images (array of {src, alt}), tables (array of arrays), metadata (object), raw_html (string). Actual fields depend on mode parameter.'
• Enhance parameter descriptions with constraints and examples. For 'mode' enum, add: 'Extraction mode: full (all data), metadata (title/description/og tags), links (href and text), headlines (h1-h6), images (src/alt), tables (parsed rows), text (clean text only), custom (CSS selector required).'
• Add WHEN/WHY guidance to tool descriptions. Example: 'Use scrape_website for flexible extraction with multiple modes. Use extract_text when you only need clean readable text. Use batch_scrape to fetch metadata from 2-10 URLs in parallel.'
• Implement structured error responses with recovery hints. Instead of 'Error: Request timeout', return: '{error: 'timeout', message: 'Request took >15s', retry: true, suggestion: 'Try a simpler mode (metadata) or check if URL is accessible'}'
• Add URL format validation and document it. Require 'url' to match pattern '^https?://' and document in description: 'Must be a valid HTTP or HTTPS URL (e.g., https://example.com). Relative URLs not supported.'