analyzing-malicious-pdf-with-peepdf

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

• Third-party content exposure detected (high risk: 0.85). The required workflow runs peepdf/pdf-parser.py on a user-supplied PDF and then ingests the extracted PDF object text (including embedded JavaScript/streams) into the agent’s runtime context via result.stdout/content_preview and print(json.dumps(report...)), which is outsider-authored free text from the malicious PDF attachment.