google-trends-ath-detector
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill retrieves data from Google Trends, specifically 'related queries' and 'rising' terms, which are then included in the analysis. This creates a surface for indirect prompt injection where adversarial search trends could influence the agent's behavior. 1. Ingestion points: The fetch_related_queries function in scripts/trend_fetcher.py retrieves search data from trends.google.com. 2. Boundary markers: The skill does not use explicit boundary markers or instructions to the agent to ignore potentially malicious content within the search data. 3. Capability inventory: The skill possesses file system read access and performs network operations. 4. Sanitization: No sanitization or filtering is performed on the ingested search terms before they are processed.
- [DATA_EXFILTRATION]: The script scripts/trend_fetcher.py accesses the user's home directory to locate the Downloads folder to read CSV files exported by the browser. Accessing the user's home directory for file retrieval is a sensitive operation. Evidence: The GoogleTrendsCrawler.init method in scripts/trend_fetcher.py checks Path.home() / 'Downloads' to set the download_dir.
- [EXTERNAL_DOWNLOADS]: The skill uses the webdriver-manager library to automatically download the appropriate browser driver binary from official repositories. Evidence: scripts/trend_fetcher.py calls ChromeDriverManager().install() to fetch the driver executable.
Audit Metadata