security-awareness
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOWNO_CODE
Full Analysis
- [SAFE] (SAFE): No malicious patterns, exfiltration attempts, or unauthorized access commands were detected. The skill consists entirely of instructional content for agent behavior.
- [NO_CODE] (INFO): This skill contains no executable scripts (Python, JavaScript, Bash) or dependency manifests, which eliminates risks associated with remote code execution or unauthorized system access.
- [PROMPT_INJECTION] (INFO): The instructions establish a 'senior cybersecurity analyst' persona to encourage defensive reasoning. This is a legitimate use of role-framing to improve agent safety and does not attempt to bypass core system constraints.
- [INDIRECT_PROMPT_INJECTION] (LOW): While the skill is intended to guide the agent when processing untrusted data (emails, URLs), it does not provide ingestion or execution capabilities itself. It establishes defensive heuristics (domain verification, content scanning) to mitigate risks at the decision-making level. 1. Ingestion points: Mentions emails and URLs from external sources. 2. Boundary markers: Absent. 3. Capability inventory: No code or tools provided. 4. Sanitization: Provides logic for manual verification before action.
Audit Metadata