figma-assets-manager
Pass
Audited by Gen Agent Trust Hub on Jul 29, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists entirely of instructional Markdown and static configuration files (JSON and YAML). It does not include any executable scripts, shell commands, or external code dependencies.
- [DATA_EXFILTRATION]: The skill documentation explicitly implements privacy-preserving instructions, directing the agent to block external optimization or CDN uploads for private assets by default. This proactive boundary significantly reduces the risk of accidental data exposure.
- [PROMPT_INJECTION]: The skill interacts with external design data, creating an indirect prompt injection surface. Ingestion points: External Figma design files (SKILL.md). Boundary markers: Explicit instructions in SKILL.md and evals/evals.json to verify privacy modes and maintain data locality. Capability inventory: Asset inventorying, SVG/raster export, and manifest generation for handoff. Sanitization: The skill instructions do not specify sanitization for Figma node metadata before manifest creation, but the focus on local preparation and privacy checks serves as a primary control.
Audit Metadata