process-detection

Pass

Audited by Gen Agent Trust Hub on Sep 29, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The documentation outlines the use of standard system utilities for process monitoring, such as ps and lsof on Unix, and PowerShell cmdlets like Get-CimInstance and Get-NetTCPConnection on Windows. These operations are essential for the skill's primary purpose of auto-instrumentation and process discovery.
  • [INDIRECT_PROMPT_INJECTION]: The skill describes an architecture that ingests potentially untrusted data from the operating system, specifically process command lines and environment metadata, which serves as an attack surface for indirect injection.
  • Ingestion points: Process listing commands including ps ax (Unix) and Get-CimInstance Win32_Process (Windows) provide command-line strings to the agent context.
  • Boundary markers: The documentation does not specify the use of delimiters or specific instructions to ignore embedded content in command lines.
  • Capability inventory: The system includes capabilities for process termination (stopProcesses), process adoption (windows.OpenProcess), and active network polling (Get-NetTCPConnection).
  • Sanitization: The implementation uses case-insensitive substring matching and filtering (strings.ToLower()) to correlate processes to projects, though specific prompt-level sanitization for ingested command strings is not detailed.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 29, 2026, 09:07 PM