glmv-resume-screen

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly accepts arbitrary resume URLs (see SKILL.md "Resume (URL)" and the usage examples) and the script scripts/resume_screen.py adds each URL as a "file_url" item in the model request payload (content -> {"type":"file_url","file_url":{"url": ...}}), meaning the model will ingest and interpret untrusted third-party webpage/file content as part of its screening workflow, exposing it to indirect prompt injection risks.