Skills
Navigate
skills/viralcode/openwhale/nodetool/Gen Agent Trust Hub

nodetool

Fail

Audited by Gen Agent Trust Hub on May 14, 2026

Risk Level: HIGHREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONCREDENTIALS_UNSAFE
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill instructs the agent to download and immediately execute installation scripts from a remote GitHub repository (nodetool-ai/nodetool) using curl | bash and irm | iex patterns. This execution of unverified remote code bypasses local security inspections and integrity checks.
  • [COMMAND_EXECUTION]: The documentation provides specific instructions to use "non-interactive mode" via the -y and -Yes flags. This is a concealment technique that directs the agent to skip all user confirmation prompts and safety warnings during the software installation process.
  • [CREDENTIALS_UNSAFE]: The skill includes a built-in command (nodetool settings show) specifically designed to display "secrets," which risks exposing sensitive API keys, tokens, or other credentials to the agent's context and output logs.
  • [COMMAND_EXECUTION]: The skill exposes capabilities for managing background jobs and proxy daemons (nodetool jobs start, nodetool proxy-daemon), which represent high-risk operations involving system services and network configuration.
Recommendations
  • HIGH: Downloads and executes remote code from: https://raw.githubusercontent.com/nodetool-ai/nodetool/refs/heads/main/install.sh - DO NOT USE without thorough review
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
May 14, 2026, 12:14 AM
Security Audit — agent-trust-hub — nodetool