Four tools with basic schemas and descriptions, but significant gaps in parameter documentation and output schema clarity. Tool names follow verb_noun convention (search, fetch, resume). Descriptions exist but are brief (20-60 chars) and lack context on when to use each tool vs. alternatives. Parameters have types but most lack descriptions. No documented output schemas. Error handling and recovery guidance absent. search_and_fetch combines two concerns (search + fetch) violating single-responsibility principle.

• Split search_and_fetch into separate tools or clearly document in description when to use vs. calling search then fetch sequentially. Each tool should have one clear responsibility.
• Add parameter descriptions to all inputs. For 'query': 'The search query string (1-500 chars)'. For 'k': 'Number of results to return (1-50, default 10)'. For 'engine': document valid values as enum.
• Document output schemas for each tool. E.g., search returns [{title, url, snippet, rank}], fetch returns {url, title, content, extracted_at}. Include field types and whether pagination is supported.
• Expand tool descriptions to 80-150 chars explaining WHAT, WHEN, and WHY. E.g., 'Search the web using Bing. Use this for general queries; use fetch() to extract full content from a specific URL. Requires active browser session.'
• Add error handling guidance. Document what errors can occur (network timeout, CAPTCHA, invalid query), which are retryable, and what the LLM should do (retry, call resume, ask user).