job-change-decision
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No security issues detected. The skill instructions prioritize user privacy and prevent autonomous actions like submitting job applications or resignations.
- [DATA_EXFILTRATION]: The helper script scripts/_common.py includes regular expressions to detect and avoid processing sensitive PII (emails, phone numbers, and social security numbers) to prevent accidental data exposure.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests user descriptions of career situations. Boundary markers are present in the instructions, requiring the agent to separate facts from hypotheses. Capability inventory shows the bundled Python scripts lack network, shell, or filesystem write access. Sanitization is provided via PII detection patterns in the shared utility script.
Audit Metadata