detecting-fileless-attacks-on-endpoints

Pass

Audited by Gen Agent Trust Hub on Apr 7, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill documentation (SKILL.md) provides PowerShell instructions to modify system registry keys under 'HKLM:\SOFTWARE\Policies\Microsoft\Windows\PowerShell' to enable Script Block and Module Logging. These operations require administrative privileges and modify the host's security telemetry configuration.
  • [EXTERNAL_DOWNLOADS]: The 'scripts/agent.py' script relies on the 'python-evtx' library for parsing Windows Event Log files, which must be installed from an external repository.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted external data.
  • Ingestion points: The 'scripts/agent.py' and 'scripts/process.py' scripts ingest data from EVTX and CSV log files provided via CLI arguments.
  • Boundary markers: No boundary markers are present in the scripts to isolate log content or prevent the agent from interpreting instructions embedded in the logs.
  • Capability inventory: The skill reads local files and uses regular expressions to extract content, which is then outputted as JSON or displayed to the agent.
  • Sanitization: Log content extracted by the scripts is not sanitized, although 'scripts/agent.py' truncates script excerpts to 300 characters.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 7, 2026, 06:47 PM