Single tool 'deckextract' has a basic description (56 chars) and minimal parameter documentation. Input schema visible with 'url' and 'format' parameters, but 'format' lacks description entirely and no enum constraints are defined. No output schema documented. Tool name lacks action verb prefix (should be 'extract_deck' or 'extract_presentation'). No error handling guidance, no pagination support, no idempotency markers. Descriptions are below the 194-char baseline for production tools.

• Rename tool to 'extract_deck' or 'extract_presentation' to start with action verb and clarify intent.
• Expand tool description to 150-250 chars: explain that it extracts structured data (text, images, metadata) from DocSend/Papermark presentation URLs, when to use it (e.g., 'to analyze investor decks'), and what formats are supported.
• Add description to 'format' parameter: 'Output format for extracted data. Supported values: pdf, json, markdown, html. Default: json.' Include enum constraint in schema.
• Document output schema with example: { "title": string, "slides": [{ "number": int, "text": string, "images": [url] }], "metadata": { "pages": int, "extracted_at": ISO8601 } }
• Add error handling: return structured errors like { "error": "invalid_url", "message": "URL must be a valid DocSend or Papermark link", "recovery": "Check the URL format and try again" } instead of raw HTTP errors.