Skills
skills/majiayu000/claude-skill-registry/adguard-home/Gen Agent Trust Hub

adguard-home

Fail

Audited by Gen Agent Trust Hub on Mar 17, 2026

Risk Level: HIGHCOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCREDENTIALS_UNSAFEDATA_EXFILTRATION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructions involve executing administrative system commands with root privileges via SSH.
  • Evidence: Use of sudo systemctl restart AdGuardHome and sudo journalctl -u AdGuardHome in SKILL.md via SSH connection.
  • [REMOTE_CODE_EXECUTION]: The skill facilitates the download and execution of binary updates from remote sources.
  • Evidence: The command sudo ./AdGuardHome --update in SKILL.md performs an automated software update at runtime.
  • [CREDENTIALS_UNSAFE]: The skill requires and handles sensitive credentials in plaintext via environment variables and CLI arguments.
  • Evidence: Use of ADGUARD_USER, ADGUARD_PASS, and ADGUARD_SSH_USER environment variables in SKILL.md.
  • [DATA_EXFILTRATION]: The skill provides instructions to access and read sensitive configuration files that may contain secrets or internal network details.
  • Evidence: The command sudo cat /opt/AdGuardHome/AdGuardHome.yaml in SKILL.md exposes the server's primary configuration file.
  • [EXTERNAL_DOWNLOADS]: The skill allows the agent to fetch data from arbitrary external URLs.
  • Evidence: python scripts/adguard_api.py add-filter --url "https://example.com/blocklist.txt" in SKILL.md.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Mar 17, 2026, 08:32 AM