detecting-malicious-npm-packages

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

• Third-party content exposure detected (medium risk: 0.30). The required runtime workflow reads outsider-authored free text from the contents of npm-distributed tarballs/JS source (e.g., scripts/agent.py downloads dist.tarball via npm view … dist.tarball and then parses package.json scripts and scans extracted *.js text for regex matches), which is attacker-controlled.