The Agent Skills Directory

[PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it is designed to ingest and analyze untrusted external forensic data.
Ingestion points: Event logs (.evtx), memory dumps (mem.raw), and disk images (disk.img) referenced in SKILL.md.
Boundary markers: Absent; there are no specific instructions or delimiters provided to prevent the agent from interpreting instructions that might be embedded within the forensic artifacts being analyzed.
Capability inventory: Includes powerful forensic tools (dd, volatility3), file system access, and commands for host/network containment.
Sanitization: Absent; the skill relies on the raw output of forensic analysis tools which could contain attacker-controlled strings.
[COMMAND_EXECUTION]: The skill provides numerous command-line examples for forensics and system administration, which are necessary for its primary purpose.
Evidence: Documented use of chainsaw, hayabusa, dd, vol3, log2timeline.py, and psort.py for evidence collection and analysis.

incident-response