architect-interrogator
Pass
Audited by Gen Agent Trust Hub on Mar 5, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill operates entirely as a set of instructional prompts for the AI agent to follow a specific 'interrogator' persona. It does not include any scripts, binary files, or network-enabled tools.
- [SAFE]: No external dependencies, remote code execution patterns, or data exfiltration vectors were detected. The mention of 'SKILL_MEMORY.md' refers to a standard local file-based memory pattern used by agents to maintain state across interactions.
- [PROMPT_INJECTION]: The 'Hard Refusals' and instructions are specific to the skill's persona and do not attempt to bypass core AI safety guardrails or ignore system-level instructions.
Audit Metadata