timeline-cobaltstrike
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data from Cobalt Strike log files, which may contain adversarial instructions embedded in operator commands or outputs.
- Ingestion points: The skill reads various log files (beacon, web, event, and keystroke logs) from the
input/c2logs/cobaltstrike/directory as specified inSKILL.md. - Boundary markers: The instructions do not define delimiters (like XML tags or triple quotes) or provide guidance to the agent to ignore instructions embedded within the logs.
- Capability inventory: The skill performs file read operations and writes normalized data to
output/cs_entries.json. While no direct shell execution is specified in the skill body, the agent processing these files may have broader capabilities. - Sanitization: There is no evidence of sanitization or escaping of the parsed log content before it is processed or written to the output file.
Audit Metadata