rollinggo-hotel-price-monitor

Warn

Audited by Gen Agent Trust Hub on Jun 6, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to use the rollinggo CLI tool by constructing shell commands that include user-provided parameters like hotel names, cities, and dates as command-line flags. Without explicit requirements for input sanitization or shell-safe quoting, this pattern is susceptible to command injection if an attacker provides malicious input.
  • [EXTERNAL_DOWNLOADS]: The skill configuration in SKILL.md and the reference guides (rollinggo-npx.md, rollinggo-uv.md) mandate the installation of the rollinggo package from NPM or PyPI at runtime. It specifically requests the latest version, which introduces a supply chain risk as the executed code is not pinned to a known-safe version and can change between executions.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jun 6, 2026, 03:56 PM
Security Audit — agent-trust-hub — rollinggo-hotel-price-monitor