zotero-paper-reader
Pass
Audited by Gen Agent Trust Hub on Jul 8, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its processing of untrusted academic data. \n
- Ingestion points: Paper metadata and full-text content are retrieved from Zotero APIs and processed via local files. \n
- Boundary markers: No specific delimiters or instructions to ignore embedded commands are used when analyzing paper content. \n
- Capability inventory: The bundled tool executes file writes to sync bibliography entries and performs network requests to Zotero services. \n
- Sanitization: External content is not sanitized or filtered before analysis.
- [COMMAND_EXECUTION]: The skill executes a local Python tool to perform Zotero operations. \n
- Evidence: Commands such as
uv run --script scripts/zotero_tool.pyare used to search libraries, download PDFs, and manage citations. This behavior aligns with the skill's stated purpose. - [EXTERNAL_DOWNLOADS]: The skill retrieves data and files from well-known academic services. \n
- Evidence: The tool downloads PDF files and metadata from the Zotero Web API. These downloads originate from a well-known service and are considered safe within the context of the skill.
Audit Metadata