housekeep-design

Pass

Audited by Gen Agent Trust Hub on Aug 25, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill performs legitimate design system maintenance tasks such as token normalization and component refactoring. It uses standard utilities like rg (ripgrep) for searching and eslint for applying guardrails.
  • [SAFE]: No malicious patterns, such as data exfiltration, unauthorized credential access, or remote code execution from untrusted sources, were detected.
  • [DATA_EXPOSURE]: The skill creates a local state file at .cursor/housekeep-design-state.md and uses .playwright-mcp/ for visual verification screenshots. These are local audit trails and do not involve external exfiltration.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests design plans from .cursor/plans/. While this represents an attack surface where a malicious plan could influence code modification, it is a standard workflow for refactoring skills. It includes a human-review checkpoint ('STOP and ask the human') for changes that affect visual branding.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 25, 2026, 05:56 AM
Security Audit — agent-trust-hub — housekeep-design