paddleocr-doc-parsing
Pass
Audited by Gen Agent Trust Hub on Apr 22, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill instructions in
SKILL.mdguide the agent to extract thePADDLEOCR_ACCESS_TOKENdirectly from the user's chat messages if environment variables are missing. This encourages the practice of sharing sensitive credentials in conversation history. - [PROMPT_INJECTION]: The skill is susceptible to Indirect Prompt Injection because it ingests untrusted document content and presents the full extracted text to the agent without sanitization.
- Ingestion points:
scripts/vl_caller.pyprocesses external files and URLs for OCR extraction. - Boundary markers: Absent. The extracted content is provided directly to the agent's context.
- Capability inventory: The skill can perform network requests (
scripts/lib.py) and file system operations (scripts/vl_caller.py,scripts/split_pdf.py). - Sanitization: No sanitization or filtering is performed on the text extracted from documents.
Audit Metadata