dhdna-profiler
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements strong ethical boundaries, explicitly instructing the agent to decline profiling for high-stakes decisions like hiring or credit.
- [SAFE]: Privacy controls are integrated into the instructions, requiring user consent before analyzing conversation history and ensuring all data remains local to the session.
- [SAFE]: External links point to legitimate research repositories and the author's official domains, representing expected vendor resources.
- [INDIRECT_PROMPT_INJECTION]: The skill analyzes user-provided text, creating an ingestion surface for untrusted data. Risk is mitigated by strict guidelines in SKILL.md requiring authorization before analyzing session history and limiting capabilities to 'Read' and 'Write' tools without network or shell access. The primary function is cognitive inference rather than direct execution of input content.
Audit Metadata