markdown-exporter
Pass
Audited by Gen Agent Trust Hub on Mar 7, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill configuration installs the
md-exporterPython package from the official PyPI registry.\n- [COMMAND_EXECUTION]: The skill tools invoke themarkdown-exportercommand-line utility to perform file processing and conversion tasks.\n- [PROMPT_INJECTION]: The skill processes external Markdown input, which introduces a surface for indirect prompt injection:\n - Ingestion points: Markdown files are read as input for all tools, including conversions to PDF, HTML, and script files (via
md_to_codeblock).\n - Boundary markers: The skill does not implement boundary markers or instructions to isolate the untrusted input from the agent's logic.\n
- Capability inventory: The skill has the capability to write various file formats to the local system, including the extraction of code blocks into executable script files (e.g., .py, .js, .sh).\n
- Sanitization: There is no evidence of sanitization or content validation to prevent the inclusion of malicious instructions or scripts within the Markdown input.
Audit Metadata