creating-flashcards
Pass
Audited by Gen Agent Trust Hub on Mar 6, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTIONEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local system commands including 'pandoc' for file format conversion and runs node.js and Python scripts located within the plugin's internal directory to handle document parsing and data formatting.
- [PROMPT_INJECTION]: The skill acts as an ingestion point for untrusted external data by processing user-provided EPUB and PDF files. This represents a potential surface for indirect prompt injection where malicious instructions embedded in the source documents could attempt to manipulate the agent's extraction or formatting behavior.
- [EXTERNAL_DOWNLOADS]: The skill performs an 'npm install' to set up dependencies for a local PDF-to-markdown utility. These downloads are directed at standard package registries to support internal skill functionality.
- [REMOTE_CODE_EXECUTION]: The instructions specify the generation of a temporary Python pipeline script at runtime to optimize large-scale translation tasks. This script is used to facilitate batch communication with a local LM Studio API endpoint.
Audit Metadata