The Agent Skills Directory

[SAFE]: The skill implements robust privacy guidelines, mandating the masking of Personally Identifiable Information (PII) and client-side redaction of sensitive data before transmission, specifically citing legal precedents and compliance standards like GDPR and CCPA.
[DATA_EXPOSURE]: The skill handles sensitive user session logs, but its core contract and boundaries strictly prohibit the exposure of PII and require anonymized evidence for all reports. It includes specific mitigations for potential litigation risks associated with session recording.
[PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it processes external event streams and session data. However, this is mitigated by instructions to analyze behavioral patterns rather than identity data and to use anonymized session references, which aligns with its primary analytical purpose.
[EXTERNAL_DOWNLOADS]: References to third-party services (Contentsquare, FullStory, Quantum Metric) are within the context of industry-standard analytics tools and do not involve untrusted code execution or suspicious remote downloads.

trace