figma-design-system-rules
Pass
Audited by Gen Agent Trust Hub on Aug 26, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests untrusted data from an external source to generate instructions that control future agent actions.
- Ingestion points: Data is extracted from Figma files, including variables, component names, properties, and documentation pages (SKILL.md).
- Boundary markers: The generated output templates in SKILL.md do not include delimiters or specific instructions to the agent to disregard embedded directives found within the Figma source data.
- Capability inventory: The skill utilizes Figma MCP for read operations and performs file writes to sensitive configuration paths such as
.cursor/rules/,AGENTS.md, andCLAUDE.md(SKILL.md). - Sanitization: The skill lacks explicit sanitization or validation logic for strings retrieved from Figma before they are interpolated into the generated markdown governance files.
Audit Metadata