literature-reading
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's core function involves processing and interpreting external, potentially untrusted data from scholarly publications.
- Ingestion points: The workflow (Steps 1 and 2) requires the agent to read and map local MinerU publications, including Markdown content and metadata.json.
- Boundary markers: The instructions do not define specific delimiters or include warnings to ignore instructions that might be embedded within the source documents.
- Capability inventory: The skill leverages tools for reading local files, searching, viewing images, and writing new Markdown files to the local filesystem.
- Sanitization: There are no provisions for sanitizing or validating the text content extracted from the parsed papers before the agent processes it.
Audit Metadata