fec-implement-from-design

Pass

Audited by Gen Agent Trust Hub on Jun 24, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted data from external design sources such as Figma, Sketch, and MasterGo to generate and modify code, creating a surface for indirect prompt injection.
  • Ingestion points: Design data obtained via MCP tools (e.g., figma, mastergo) or user-provided screenshots and annotations as described in SKILL.md.
  • Boundary markers: No explicit delimiters or instructions to ignore embedded commands within the design data are provided to the agent.
  • Capability inventory: The agent is instructed to search the codebase, modify framework-specific components, write implementation reports, and execute local commands (SKILL.md).
  • Sanitization: No validation or sanitization of the input design data or visual descriptions is mentioned.
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute local developer tools for verification.
  • Evidence: "If there is lint, execute lint; If there are tests, execute them" in Step 8 of SKILL.md.
  • Context: These are standard development operations for the intended use-case and do not involve suspicious parameters, privilege escalation, or unexpected network activity.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 24, 2026, 04:37 PM