to-markdown
Pass
Audited by Gen Agent Trust Hub on May 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill runs shell commands to perform document conversion, including path checking and file redirection. It includes instructions to suppress diagnostic output for cleaner results.
- [EXTERNAL_DOWNLOADS]: The skill recommends installing the
markitdownpackage from the Python Package Index (PyPI). This utility is an official project from a trusted technology organization. - [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection. Ingestion points: External files including PDFs, Word documents, images, and YouTube transcripts are processed by the tool. Boundary markers: The conversion workflow does not specify delimiters or explicit 'ignore instructions' warnings for the converted content. Capability inventory: The agent is authorized to execute shell commands and read/write files to the local system. Sanitization: Content extracted from documents is provided to the agent without sanitization or validation of the text body.
Audit Metadata