alterlab-pdf-extract
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied document content and interpolates it directly into a prompt sent to an LLM. A malicious document containing adversarial instructions could attempt to override the extraction logic. Ingestion points:
extract_to_table.pyreads and converts documents. Boundary markers: Uses basic instructional text but lacks strict structural delimiters for document content. Capability inventory: Reads local files, writes output to files, and makes network requests to API providers. Sanitization: No specific escaping or validation of document content before prompt interpolation. - [COMMAND_EXECUTION]: The skill uses a Python script to perform document processing and data extraction.
- [EXTERNAL_DOWNLOADS]: Requires standard external packages including
markitdownandopenaifor document conversion and LLM interaction.
Audit Metadata