patent-oa

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

• Third-party content exposure detected (low risk: 0.10). The workflow reads and parses user-provided external files (such as local PDF/text documents via ingest_playbook.py peek / read_document and case files via ingest_case.py), which are outsider-authored free-text ingestion points.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.70). The skill package utilizes Python scripts (book_to_skill_setup.py) to fetch code and installer instructions from GitHub at runtime (https://raw.githubusercontent.com/virgiliojr94/book-to-skill/master/README.md and https://github.com/virgiliojr94/book-to-skill.git), falling under the weak/unknown/suspicious community provenance tier because ownership and distribution of the third-party dependency are unverified.