Single tool 'scrape_markdown' has minimal definition quality. Description is present (56 chars) but lacks context on when to use it, prerequisites, or what data structure is returned. Input schema exists with basic type info but no parameter description. No output schema documented. No error handling guidance. Tool name is verb-noun compliant but lacks depth expected for production use.

• Add parameter description to 'url': 'The absolute URL to scrape (http/https only). Must be publicly accessible and not blocked by robots.txt.'
• Expand tool description to 150-200 chars: 'Fetch a URL and convert its HTML content to markdown. Use this to extract readable text from web pages for analysis or summarization. Returns markdown-formatted content. Respects robots.txt and rate limits to 10 req/min per API key.'
• Document output schema: { type: 'object', properties: { content: { type: 'array', items: { type: 'object', properties: { type: { enum: ['text'] }, text: { type: 'string' } } } } } }
• Add error handling guidance in description: 'Returns error if URL is invalid, unreachable, or blocked. Retry with exponential backoff for timeout errors (429, 503).'
• Implement per-request timeout (e.g., 30s) and document it in the description to prevent agent hangs.