figma-slop-check

Pass

Audited by Gen Agent Trust Hub on Oct 8, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes untrusted data from Figma files, including layer names and text node content, which could contain malicious instructions designed to influence the agent's behavior during the audit.
  • Ingestion points: The agent extracts strings from characters and names fields of Figma nodes via the figma-console MCP server (referenced in references/detectors.md and SKILL.md).
  • Boundary markers: No specific delimiters or "ignore instructions" warnings are defined for the processing of text content.
  • Capability inventory: The skill utilizes powerful tools like figma_execute (which runs arbitrary JavaScript in the Figma environment) and performs local file writes to manage the audit ledger.
  • Sanitization: No sanitization or filtering of external text content is performed before the data is analyzed by the agent.
  • [DYNAMIC_EXECUTION]: The skill instructs the agent to dynamically generate and execute JavaScript scripts to collect design data from the Figma environment.
  • Evidence: references/detectors.md provides templates and instructions for the agent to "write a short read-only figma_execute script of your own" to traverse the Figma node tree and extract property records.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 8, 2026, 08:06 AM