The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes external data sources which could be used for indirect prompt injection. Ingestion points: The scripts/agent.py script reads and analyzes data from external log files (dlp_alerts.csv and access_logs.json). Boundary markers: The skill does not implement specific boundary markers or instructions to ignore potential commands embedded within the log data. Capability inventory: The skill's automated script performs data analysis and generates JSON-formatted reports; it does not possess network exfiltration, arbitrary command execution, or sensitive system file access capabilities. Sanitization: Log data is parsed using standard Python csv and json libraries, but no additional content sanitization is performed on the log values before they are presented to the agent.

investigating-insider-threat-indicators