Loading ToolBench
Unified OpenClaw Figma skill with direct inspect/read flows and a heavier write path when needed
Descriptions are present but generic (10-50 chars), below the 50-200 char LLM-optimized baseline. No parameter descriptions exist. No output schemas documented. No error handling guidance. Tool naming is reasonable (verb_noun pattern), but the lack of schema and parameter detail makes tools unusable by agents without trial-and-error. This server is not production-ready.
Create a new Figma file
Export an image from a Figma file
Get design context and metadata from a Figma file
Get a screenshot/image export from a Figma file or node
Get a thumbnail image from a Figma file
Render an entire Figma document as an image
Render a specific node from a Figma file as an image
7 of 9 tools have empty input schemas ({}). Agents cannot determine what parameters to pass.
No parameter descriptions for any tool. Even use_figma (the only tool with a schema) lacks descriptions for fileKey, description, and code parameters. LLMs cannot infer what values to pass.
Tool descriptions are 10-50 characters, well below the 50-200 char LLM-optimized baseline. Descriptions lack WHAT the tool does, WHEN to use it, and what it returns. Example: 'Get a screenshot/image export from a Figma file or node' (56 chars) does not explain required parameters, output format, or when to prefer get_screenshot vs export_image vs get_thumbnail.
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 35 | <=2025-11-25 | v2 |
Execute Figma Plugin API code against a file
Get current authenticated user information
No output schemas documented. Agents do not know what fields to expect from any tool. This breaks tool chaining, if get_design_context returns a file_id, downstream tools cannot reference it.
No error handling guidance. Agents do not know what to do if a tool fails. No recovery hints, no categorization of retryable vs fatal errors, no actionable error messages.