bcms-content

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

• Third-party content exposure detected (high risk: 0.85). The CLI ingests outsider-authored free text from the operator’s command-line --data '<json>' / --data-file <path> (arbitrary JSON/string content), parses it as JSON/prose, and then includes it in the LLM-visible stdout via printResult(...) after create-entry/update-entry (JSON output becomes LLM context).