google-trends-ath-detector
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- Indirect Prompt Injection (LOW): The skill is susceptible to Indirect Prompt Injection (Category 8). It ingests 'Related Queries' from Google Trends, which is external, untrusted data. If a malicious actor influences Google's trending terms with prompt-injection-like strings, the agent might inadvertently process them as instructions. Evidence: 1. Ingestion: 'drivers_from_related_queries' (via scripts/trend_fetcher.py). 2. Boundary markers: Absent. 3. Capability inventory: Execution of Python scripts and CLI commands. 4. Sanitization: Not documented.
- External Downloads (LOW): The skill utilizes 'webdriver-manager' which automatically downloads and executes browser drivers (e.g., Chromedriver) from external sources. This is standard behavior for Selenium automation but introduces a dependency on external binary downloads.
- Command Execution (LOW): The skill requires the execution of Python scripts and CLI commands for data fetching. While these are local to the skill, users should audit 'scripts/trend_fetcher.py' to ensure it doesn't perform unauthorized system operations or network calls.
- Credentials Unsafe (SAFE): No hardcoded API keys or secrets were found. The skill mentions a 'login mode' for 2FA, but this is an interactive process where the user provides credentials manually to the browser session.
Audit Metadata