gdelt-mentions-fetch

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

• Third-party content exposure detected (high risk: 0.70). The skill explicitly fetches and parses public GDELT endpoints (lastupdate.txt, masterfilelist.txt, and *.mentions.CSV.zip from http://data.gdeltproject.org/gdeltv2) as shown in SKILL.md and scripts/gdelt_mentions_fetch.py, and the agent is expected to read and act on those remote file contents (including previewed/decompressed lines) which could materially influence subsequent selection, validation, or actions.