gdelt-doc-search

Pass

Audited by Gen Agent Trust Hub on Mar 15, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [SAFE]: No malicious patterns, persistence mechanisms, or privilege escalation attempts were detected in the provided skill scripts or markdown files.
  • [EXTERNAL_DOWNLOADS]: Fetches data from the GDELT Project's official DOC 2.0 API endpoint (https://api.gdeltproject.org/api/v2/doc/doc). This is a well-known service for academic and research data retrieval.
  • [COMMAND_EXECUTION]: The scripts/gdelt_doc_search.py script performs network operations and file writes using standard library modules like urllib and pathlib. It does not execute arbitrary shell commands or external binaries.
  • [DATA_EXFILTRATION]: The skill facilitates the transfer of API response data to local file paths as specified by user-provided arguments (--output). It also reads configuration settings from environment variables, following standard operational patterns.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface because it ingests data from an external API and presents it to the agent.
  • Ingestion points: scripts/gdelt_doc_search.py (via urllib.request.urlopen for API responses).
  • Boundary markers: Absent.
  • Capability inventory: Local file write capability and console output.
  • Sanitization: Absent; the script returns raw API content or JSON data without filtering.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 15, 2026, 01:59 AM