claude-design-critic

Pass

Audited by Gen Agent Trust Hub on Jun 12, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is focused on design and copy auditing. No malicious patterns, command execution, or data exfiltration concerns were found.\n- [PROMPT_INJECTION]: This skill has an Indirect Prompt Injection surface because it ingests untrusted data from external websites. Ingestion points: WebFetch tool (Step 1) and Read tool (Step 1). Boundary markers: Absent; instructions do not contain delimiters for fetched content. Capability inventory: The agent can use Bash, Write, and Agent tools (Step 4). Sanitization: Absent. This surface is necessary for the skill's function and no malicious intent is detected.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 12, 2026, 08:32 PM
Security Audit — agent-trust-hub — claude-design-critic