research-engineer
Pass
Audited by Gen Agent Trust Hub on Feb 19, 2026
Risk Level: SAFE
Full Analysis
- PROMPT_INJECTION (SAFE): The skill uses authoritative language to define a persona ('Senior Research Engineer', 'not an assistant') and mandates specific behavior ('Zero-Hallucination', 'Objective Neutrality'). These instructions guide the AI's tone and methodology but do not attempt to bypass system safety protocols or extract internal model configurations.
- DATA_EXFILTRATION (SAFE): There are no patterns indicating attempts to access sensitive files, credentials, or perform network requests to external domains.
- REMOTE_CODE_EXECUTION (SAFE): No remote code patterns, download commands (curl/wget), or package installations were detected. The skill contains code examples in C++ and Python within markdown blocks for documentation purposes only.
- COMMAND_EXECUTION (SAFE): No shell commands, subprocess calls, or privilege escalation attempts (sudo/chmod) are present in the instruction set.
- OBFUSCATION (SAFE): The text is clear and readable, with no evidence of Base64, zero-width characters, or other hidden encoding techniques.
Audit Metadata