The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill provides instructions to download and install the 'markitdown' package from Microsoft's official GitHub repository and PyPI, which are established and trusted sources.
[COMMAND_EXECUTION]: The script scripts/generate_schematic.py executes a secondary Python script using subprocess.run. The command is constructed using a list of arguments rather than a shell string, which is a secure practice that prevents shell injection vulnerabilities.
[PROMPT_INJECTION]: Documentation in SKILL.md includes instructions for the agent to automatically generate scientific schematics for new documents by default. While this influences the agent's intended behavior, it is presented as a functional enhancement for document preparation.
[PROMPT_INJECTION]: The skill's primary function involves processing content from untrusted external files (such as PDFs, DOCX, and XLSX) using the markitdown library. This creates a surface for indirect prompt injection where malicious instructions embedded in a processed document could influence the agent's behavior.
Ingestion points: Files are ingested via md.convert() in scripts/batch_convert.py, scripts/convert_literature.py, and scripts/convert_with_ai.py.
Boundary markers: No explicit boundary markers or 'ignore embedded instruction' warnings are used when processing the extracted text.
Capability inventory: The skill has capabilities for file writing (Path.write_text), network operations (requests.post to OpenRouter), and subprocess execution (subprocess.run).
Sanitization: Content is converted to Markdown but no additional sanitization or instruction filtering is applied to the extracted text before it enters the agent's context.

markitdown