Single tool 'get_design_context' has a clear verb-noun name and basic schema with two parameters (fileId, pageId). However, the description is minimal (27 chars), parameter descriptions are generic, and there is no documented output schema. The tool lacks error handling guidance, pagination support for potentially large design contexts, and field-level documentation. No evidence of idempotency guarantees or recovery patterns. Naming is acceptable but descriptions fall well below the 50-200 char LLM-optimized baseline for A-grade tools.

• Expand tool description to 80-150 chars: 'Retrieve design context (layout, components, styles, tokens) from a Lanhu design file. Use this to understand design structure before generating code or recreating pages. Requires a valid Lanhu file ID.'
• Add format and discovery hints to parameters: fileId description should include 'A valid Lanhu design file ID (alphanumeric, typically 16-32 chars). Obtain from the Lanhu app URL or file list.' pageId should clarify 'Optional. If omitted, returns context for all pages. Page IDs are listed in the design file structure.'
• Document the output schema explicitly: 'Returns an object with fields: pages (array of page objects with id, name, components), components (array with name, type, styles), tokens (design tokens for colors, typography, spacing), metadata (file name, version).'
• Add error handling examples: 'If fileId is invalid, returns {error: "File not found", suggestion: "Check the file ID in the Lanhu app."}. If pageId does not exist, returns {error: "Page not found", available_pages: [...]}'.
• Implement result limiting: Add optional 'limit' parameter (default 50) to cap returned components/tokens. Document: 'Large design files may return thousands of items. Use limit to control response size and avoid context window exhaustion.'