Five tools with basic verb-noun naming and present descriptions, but significant gaps in parameter documentation and output schema clarity. All tools have descriptions (10-50 chars baseline), but most lack detailed parameter descriptions and explicit output schemas. No error handling guidance, no input validation hints, and no security annotations. Credentials are properly injected via environment variables, but the set_wolai_credentials tool exposes secrets as parameters, a critical anti-pattern. Tool composition is reasonable (single responsibility), but parameter naming lacks type suffixes (e.g., 'block_id' is good, but 'root_id' could be clearer). Output structures are inferred from code but not formally documented in tool definitions.

• Remove app_id and app_secret from set_wolai_credentials parameters. Instead, document that credentials must be set via environment variables (WOLAI_APP_ID, WOLAI_APP_SECRET) in the MCP config. If runtime updates are needed, use a secure vault or configuration service, not tool parameters.
• Document output schemas for all tools. For get_page_content, explicitly state: 'Returns {id, title, type, content, children: [...]}' so LLMs know what fields to expect and can chain calls correctly.
• Expand parameter descriptions to include format, constraints, and discovery hints. E.g., 'block_id: The Wolai page/block ID (UUID format, obtainable via get_root_info or from the Wolai web UI).'
• Add error handling guidance. E.g., 'If block_id is invalid, returns 404 with message. Try get_root_info to verify the root page ID first.' This guides LLM recovery.
• Document pagination and result limits. State: 'Recursively fetches all child blocks. For pages with >1000 children, consider filtering by type or using get_root_info first to preview structure.'