figma
Pass
Audited by Gen Agent Trust Hub on Mar 22, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches design specifications and assets from Figma's official API (api.figma.com), which is a well-known service.
- [COMMAND_EXECUTION]: Provides bash snippets using
curlandjqto interact with remote APIs and process structured data. - [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection by ingesting untrusted content from Figma files.
- Ingestion points: Data is retrieved from Figma file nodes, styles, and components via the REST API (SKILL.md).
- Boundary markers: The skill lacks explicit instructions to treat design metadata as untrusted or to use delimiters when processing Figma content.
- Capability inventory: The skill has access to
Bashfor command execution,Writefor file system modification, andWebFetchfor network operations (SKILL.md frontmatter). - Sanitization: There is no mention of sanitizing or validating component names, descriptions, or annotations before they are processed by the agent or used in code generation.
Audit Metadata