Server has 10 tools with complete input schemas and descriptions present for all tools and parameters. Naming follows verb_noun convention (fetch_url, search_web, browser_session_*). However, descriptions are generic and lack LLM-optimization guidance (e.g., 'Fetch and parse a URL' doesn't explain WHEN to use vs screenshot_url). Parameter descriptions are minimal (e.g., 'The URL to fetch' lacks format/constraint details). Output schemas are not documented, LLMs cannot infer what fields to expect from responses. Error handling is absent from tool definitions. No tool annotations (readOnlyHint/destructiveHint) despite clear risk profiles. browser_session_* tools form a coherent chain but lack dependency hints in descriptions.

• Document output schemas for all tools. Example: fetch_url returns {text: string, url: string, title?: string, error?: string}. This enables LLMs to chain tools correctly.
• Expand tool descriptions to 50-200 chars and answer: what does it do, when should I use it instead of similar tools, what does it return? Example: 'Fetch and parse a URL into markdown text. Use when you need article content; use screenshot_url for visual inspection. Returns extracted text, title, and original URL.'
• Add tool annotations: mark fetch_url, screenshot_url, search_web, search_gutenberg, get_gutenberg_book, browser_session_navigate, browser_session_screenshot, browser_session_get_content as readOnlyHint=true; mark browser_session_close as destructiveHint=true.
• Add dependency hints to browser_session_* tools. Example for browser_session_navigate: 'Navigate to a URL in an existing browser session. Requires a session_id from browser_session_create. Returns success status and page load time.'
• Enhance parameter descriptions with format and constraint details. Example: 'url (string, required): Full URL starting with http:// or https://. Max 2048 chars. Timeout: 30s.' Example: 'session_id (string, required): Browser session ID returned by browser_session_create. Format: alphanumeric, 16-32 chars.'