paper-reading-assistant
Pass
Audited by Gen Agent Trust Hub on Apr 2, 2026
Risk Level: SAFE
Full Analysis
- [DATA_EXPOSURE]: The skill includes Python scripts that access local filesystem paths to read PDF files and write extracted metadata to JSON files. This is consistent with its stated purpose of batch processing research papers.- [INDIRECT_PROMPT_INJECTION]: The skill defines templates that interpolate content extracted from academic papers directly into AI prompts (e.g., '{paper_text}'). While this creates a surface where text within a PDF could attempt to influence the agent's behavior, the skill does not grant the agent dangerous capabilities like network access or system command execution based on that input.- [EXTERNAL_DOWNLOADS]: Mentions the use of external tools such as Zotero, Paperpile, and Semantic Reader for reference management, which are standard tools in academic research environments.
Audit Metadata