frontend-design
Pass
Audited by Gen Agent Trust Hub on Feb 15, 2026
Risk Level: LOW
Full Analysis
- Prompt Injection (SAFE): No malicious instructions or safety filter bypasses were detected. The use of 'IMPORTANT' and 'CRITICAL' markers is consistent with benign stylistic guidance for the AI.
- Data Exposure & Exfiltration (SAFE): The skill does not access sensitive local files or perform unauthorized network requests.
- External Downloads (SAFE): Contains a reference to an official Anthropic cookbook on GitHub, which is a verified trusted source. No runtime script downloads or executions occur.
- Obfuscation (SAFE): No encoded strings, zero-width characters, or hidden payloads were found within the documentation or skill files.
- Indirect Prompt Injection (SAFE): While the skill processes user design requirements, it does so to generate UI code for display purposes and lacks capabilities for file modification, command execution, or network exfiltration.
- Dynamic Execution (SAFE): No patterns of runtime compilation, unsafe deserialization, or code injection were identified.
Audit Metadata