data-extractor
Pass
Audited by Gen Agent Trust Hub on Mar 5, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns, direct prompt injections, or unauthorized data access commands were detected in the skill instructions or code snippets.\n- [EXTERNAL_DOWNLOADS]: The skill recommends installing the 'unstructured' library. This is a well-known, legitimate package for document partitioning and extraction from a trusted open-source origin.\n- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted document formats (PDF, Word, HTML, Email), creating a potential surface for indirect prompt injection.\n
- Ingestion points: Document files processed via
partition(),partition_pdf(), and other partitioning functions inSKILL.md.\n - Boundary markers: No explicit delimiters or 'ignore embedded instructions' warnings are provided in the example prompts.\n
- Capability inventory: Uses the
unstructuredlibrary for extraction; includes file reading, regex processing, and batch document handling.\n - Sanitization: The provided code examples do not demonstrate sanitization or validation of the content extracted from external documents.
Audit Metadata