The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes external shell commands to facilitate the conversion of legacy document formats (.doc, .wps) into the modern .docx format.
Evidence: The SofficeConverter.convert_to_docx method in scripts/wfp_core.py uses subprocess.run to call the LibreOffice soffice binary. The execution is limited to a specific headless conversion command with controlled arguments.
[DYNAMIC_EXECUTION]: The skill employs dynamic module loading within its testing infrastructure to discover and run unit tests.
Evidence: In scripts/wfp_tests.py, the main function uses __import__(__name__) to load the current module into the unittest loader. This is a standard Python pattern for self-contained test scripts.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user-provided document files, which establishes a surface for indirect prompt injection if the agent interprets content from these files as instructions.
Ingestion points: WordProcessor.format_document in scripts/wfp_core.py reads content from various document formats (.docx, .txt, .md).
Boundary markers: The skill does not implement specific delimiters or warnings to isolate processed document content from the agent's instruction context.
Capability inventory: The skill has the ability to write to the file system to save documents and execute the LibreOffice binary for format conversion.
Sanitization: The skill performs filename sanitization using RE_SAFE_FILENAME_CHARS and strips Markdown metadata through _clean_markdown to normalize input data.

doc-format