The Agent Skills Directory

[PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection through its ingestion of external design data. 1. Ingestion points: Figma design data is retrieved from API responses and screenshots, which are controlled by external users/designers. 2. Boundary markers: The skill does not define explicit delimiters or instructions to ignore potential commands embedded within the design metadata or visual text when processing the data via multimodal models or generating artifacts. 3. Capability inventory: The skill has the capability to write files to the local file system (Write), interact with the user (AskUserQuestion), and execute shell commands (bash). 4. Sanitization: There is no documented logic for sanitizing or validating text extracted from Figma designs before it is used in downstream tasks or artifact generation.\n- [EXTERNAL_DOWNLOADS]: The setup guide recommends installing external code, specifically the figma-context-mcp package via npx and official plugins from a remote registry, which introduces a dependency on external software supply chains.\n- [COMMAND_EXECUTION]: The skill describes a fallback mechanism that involves the agent executing bash scripts to perform manual REST API calls to Figma's infrastructure.

figma-design