design-critique

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE] (SAFE): The skill consists entirely of markdown documentation and instructional guidelines. No scripts, binaries, or automated configuration files are included.
  • [Indirect Prompt Injection] (LOW): As a design critique tool, the skill is intended to process external data (via WebFetch and browser tools). While this represents a surface for indirect prompt injection if an attacker-controlled website contains malicious instructions, this is a general risk associated with the skill's primary function. No specific vulnerabilities or lack of safeguards were found beyond standard operational risks.
  • Ingestion points: Tools WebFetch and mcp__claude-in-chrome__* are used to ingest external web content.
  • Boundary markers: None explicitly defined in the prompt instructions.
  • Capability inventory: File reading (Read), web fetching (WebFetch), and browser interaction (mcp__claude-in-chrome).
  • Sanitization: Standard LLM guardrails are assumed; no skill-specific sanitization is implemented.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:29 PM