gaozhong-english-answer-parser
Pass
Audited by Gen Agent Trust Hub on Aug 8, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides instructions for the agent to use standard document processing tools like
pdfplumberandpdftoppm. No network activity, credential exposure, or persistent system modifications are requested. - [PROMPT_INJECTION]: The skill is designed to ingest untrusted data from user-uploaded PDF and image files (Ingestion Point). While there is no explicit sanitization step for the extracted text, the skill provides clear boundary instructions to ignore non-answer explanatory text. The risk of indirect prompt injection is significantly mitigated by a capability inventory that strictly limits output to a predefined Markdown template and explicitly forbids the generation of code blocks or files.
Audit Metadata