ocr

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly requires extracting and quoting "exact text" from local images/PDFs and instructs "When quoting, quote from OCR text", which would force the LLM to reproduce any secrets found in those documents verbatim.

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

• Third-party content exposure detected (medium risk: 0.30). The runtime workflow runs ocr extract/open on a user-provided local image/PDF path and then ingests the OCR-extracted “content” from that document text as agent context.