Three tools with basic schemas and descriptions, but significant gaps in parameter documentation and output schema clarity. Tool names follow verb_noun convention (add, scrape_web, search_docs), which is good. However, descriptions are minimal (10-60 chars), parameter descriptions are sparse, and output schemas are not documented. No error handling guidance, no pagination support for list-like tools, and no tool annotations. The 'add' tool is trivial; 'scrape_web' and 'search_docs' lack critical parameter details (e.g., timeout, result format, error cases).

• Expand tool descriptions to 100-200 chars. For scrape_web: 'Fetch and convert a web page to markdown. Use when you need to read content from a URL. Returns markdown text. Fails on 404, timeouts, or blocked sites, try a different URL or ask the user.'
• Document output schemas explicitly. E.g., scrape_web returns {content: string, url: string, title: string}; search_docs returns {results: [{title, url, snippet, relevance_score}], total_count, next_cursor}.
• Add pagination to search_docs: accept limit (1-50, default 5) and offset or next_cursor. Return total_count so LLM knows if more results exist.
• Add parameter constraints: scrape_web url should validate URL format; search_docs query should require min 1 char, max 500 chars.
• Include error recovery hints in descriptions: 'If URL is invalid or blocked, try a simpler URL or ask the user for clarification.'