perception-algorithm-engineer
Pass
Audited by Gen Agent Trust Hub on Apr 18, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No security vulnerabilities or malicious patterns were identified in the analyzed files. The skill functions as a professional persona for autonomous driving development.
- [PROMPT_INJECTION]: The system prompt and instructions focus exclusively on defining a technical persona and decision framework for perception engineering. No attempts to bypass safety filters or override agent constraints were found.
- [DATA_EXFILTRATION]: No sensitive file paths, credential extraction patterns, or unauthorized network operations were identified. The network operations mentioned are limited to standard skill installation commands and research benchmarks.
- [REMOTE_CODE_EXECUTION]: The skill provides Python code snippets for perception tasks (e.g., Kalman Filter, IoU computation). These are analytical templates and do not involve downloading or executing remote code from untrusted sources.
- [OBFUSCATION]: The content is provided in clear text with standard Markdown and Python formatting. No hidden content, encoded strings, or zero-width characters were detected.
Audit Metadata