paper-workflow-orchestrator

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

• Third-party content exposure detected (high risk: 0.90). The skill may silently invoke .trae/skills/authoritative-data-harvester/scripts/run.py to populate crawled_data/ with external/public web data (see "外部资源获取 (Optional)" in SKILL.md and the harvester check in scripts/run_all.py), and those crawled_data/ files are then scanned and used by the data-cleaning and micro-unit generation pipelines—allowing untrusted third-party content to influence processing and downstream actions.