gdelt-gkg-fetch

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.80). The skill explicitly fetches and parses public third-party endpoints (lastupdate.txt and masterfilelist.txt from GDELT_GKG_BASE_URL) and downloads/inspects *.gkg.csv.zip payloads (see scripts/gdelt_gkg_fetch.py and SKILL.md/OpenClaw templates), and those untrusted public file contents and previews are read and used to select files, validate structure, and produce JSON results that can drive subsequent agent actions.