coverage-check

Pass

Audited by Gen Agent Trust Hub on Sep 4, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill queries the GDELT Project API (https://api.gdeltproject.org/api/v2/doc/doc). This is a well-known public data service for monitoring global news media. The request is performed with appropriate timeouts and error handling.
  • [COMMAND_EXECUTION]: The skill uses argparse to handle user input for queries and configuration. The inputs are passed to a requests call as parameters, which is a safe way to handle external input without shell injection risks.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes news headlines and metadata from the GDELT API. It includes a clear warning that 'nothing does not mean unreported' and distinguishes between 'no measurement' and 'no results.' It performs structural analysis (clustering) rather than executing instructions found in the data.
  • [DATA_EXFILTRATION]: The skill does not access sensitive local files or environment variables. It only sends the user-provided search query to the GDELT API. It includes logic to strip common tracking parameters from URLs to improve deduplication and privacy.
  • [SAFE]: The code follows best practices for secret management by explicitly stating it requires no API keys or authentication, and it implements exponential backoff for rate-limiting compliance.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 4, 2026, 10:21 AM
Security Audit — agent-trust-hub — coverage-check