Four tools with complete schemas and basic descriptions, but significant gaps in parameter documentation and error handling. All tools are READ_ONLY and follow verb_noun naming (get_*, export_*). Descriptions are present but minimal (10-50 chars), below the 194-char baseline for A+ tools. Parameter descriptions exist but lack detail on formats, constraints, and dependencies. No output schema documentation. No error recovery guidance. No tool annotations. Composition is reasonable, tools are single-purpose and chainable via fileKey/nodeId.

• Expand tool descriptions to 100-200 chars. Example: 'get_figma_file: Retrieve metadata and structure of a Figma file by its key (found in the file URL after /file/). Returns document name, pages, components, and version history. Use this to explore file structure before querying specific nodes.'
• Add minimum/maximum constraints to numeric parameters. For export_figma_image.scale, add 'minimum: 1, maximum: 4' to the schema and update description: 'Scale factor for raster exports (PNG/JPG). Must be 1 - 4; SVG ignores this parameter.'
• Document output schemas for each tool. Example for get_figma_file: 'Returns {name: string, pages: [{id, name, children}], components: {id, name}, version: string}. Strip audit fields and return only fields needed for agent reasoning.'
• Add error recovery guidance in descriptions. Example: 'If the API returns 401, verify FIGMA_API_KEY is set. If 404, check that fileKey is correct (format: alphanumeric, ~20 chars). If 429, wait 60s before retrying.'
• Add tool annotations (readOnlyHint: true) to all four tools to signal to clients that these are safe, non-destructive operations.