pp-adguard-home

Pass

Audited by Gen Agent Trust Hub on Jun 28, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches the adguard-home-pp-cli tool from the @mvanhorn/printing-press-library npm package.
  • [REMOTE_CODE_EXECUTION]: Installs binaries via go install from the vendor's repository at github.com/mvanhorn/printing-press-library.
  • [COMMAND_EXECUTION]: Executes the adguard-home-pp-cli binary to manage AdGuard Home settings, including TLS configuration, filtering rules, and DHCP leases. It also manages local configuration files in the user's home directory (~/.config/adguard-home-pp-cli/config.toml).
  • [DATA_EXFILTRATION]: The skill uses a tool that supports a --deliver webhook:<url> flag, which can be used to send command results to external URLs.
  • [PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection by processing and displaying data from AdGuard Home logs and client lists.
  • Ingestion points: REST API outputs from AdGuard Home (query logs, client info).
  • Boundary markers: Not explicitly defined for tool output in the instructions.
  • Capability inventory: Binary execution of adguard-home-pp-cli, network requests via webhooks, and local file system writes.
  • Sanitization: Use of --agent and --json flags to ensure machine-readable, non-interactive output.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 28, 2026, 01:05 PM
Security Audit — agent-trust-hub — pp-adguard-home