skills/liangmiqwq/skills/cc-design/Gen Agent Trust Hub

cc-design

Pass

Audited by Gen Agent Trust Hub on May 10, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses Bash and Node.js scripts to perform legitimate design tasks such as video rendering via ffmpeg and PDF/PPTX export via Playwright. These scripts are well-documented, include input validation (e.g., in add-sfx-to-video.sh), and are essential for the skill's visual design functionality.
  • [EXTERNAL_DOWNLOADS]: The skill performs version checks and fetches design system specifications from established repositories. It references the author's own GitHub (ZeroZ-lab/cc-design) for updates and the getdesign.md catalog for design tokens. These are legitimate resources for a high-fidelity design tool.
  • [DATA_EXFILTRATION]: No unauthorized data transmission was detected. The skill implements a 'Context Preservation' mechanism that saves design tokens (colors, font names) to a local file (.claude/design-context.json) to prevent loss during context compression. This data remains within the local project directory and is used solely for session recovery.
  • [REMOTE_CODE_EXECUTION]: The skill contains an update check script (hooks-lib/update-check.sh) that fetches a version string from GitHub. It does not execute this string or any remote scripts directly; it identifies if a newer version exists and instructs the user to update manually via the package manager.
Audit Metadata
Risk Level
SAFE
Analyzed
May 10, 2026, 12:26 AM
Security Audit — agent-trust-hub — cc-design