Skills
skills/aradotso/security-skills/k7-total-security-deployment/Gen Agent Trust Hub

k7-total-security-deployment

Warn

Audited by Gen Agent Trust Hub on Jul 5, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill fetches installation packages from an external domain for multiple operating systems.
  • Evidence: Invoke-WebRequest -Uri "https://example.com/k7-agent-16.0.1195-win-x64.msi" in SKILL.md.
  • Evidence: curl -O https://example.com/k7-agent-16.0.1195-macos.pkg in SKILL.md.
  • Evidence: wget https://example.com/k7-agent-16.0.1195-amd64.deb in SKILL.md.
  • [COMMAND_EXECUTION]: The skill performs installation and management tasks requiring administrative privileges through shell commands.
  • Evidence: msiexec /i k7-agent.msi /qn in SKILL.md.
  • Evidence: sudo installer -pkg k7-agent-16.0.1195-macos.pkg -target / in SKILL.md.
  • Evidence: sudo dpkg -i k7-agent-16.0.1195-amd64.deb in SKILL.md.
  • Evidence: sudo systemctl restart k7-agent in SKILL.md.
  • [PROMPT_INJECTION]: The skill facilitates the ingestion of untrusted security data into AI model contexts for analysis.
  • Ingestion points: The skill uses k7-console --list-threats and k7-console --view-logs to retrieve data for AI integration in SKILL.md.
  • Boundary markers: No explicit delimiters or instructions are specified to isolate external log data from the AI's core instructions.
  • Capability inventory: The skill has access to system-level operations including file quarantine, policy enforcement, and network connectivity testing through the k7-console utility.
  • Sanitization: There is no documented validation or filtering process for the logs and threat data before they are analyzed by the AI.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jul 5, 2026, 11:27 PM
Security Audit — agent-trust-hub — k7-total-security-deployment