The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes content from external Word documents and user-provided Markdown, which represents an indirect prompt injection surface. Malicious instructions embedded in these files could potentially influence agent behavior.\n
Ingestion points: inspect_docx.py (reads .docx structure) and export_markdown_docx.py (reads markdown from stdin).\n
Boundary markers: Absent. The skill does not explicitly use delimiters to isolate untrusted content from system instructions.\n
Capability inventory: The skill can execute file operations and shell commands (unzip, zip) as instructed in the documentation.\n
Sanitization: No explicit sanitization or filtering of processed text is performed.\n- [COMMAND_EXECUTION]: The skill's instructions describe the use of system utilities like unzip and zip to perform structural edits on the OOXML components of .docx files. This is a standard and expected behavior for advanced Word document manipulation.\n- [EXTERNAL_DOWNLOADS]: The skill requests the installation of the python-docx library. This is a well-known and standard library for Word document handling, and it is installed from the official Python Package Index (PyPI).

docx