The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses Python scripts scripts/fill_docx.py and scripts/embed_evidence.py to process Word documents and convert images/PDFs. These scripts are invoked via defined commands to automate file manipulation tasks.\n- [INDIRECT_PROMPT_INJECTION]: The skill provides an attack surface for indirect prompt injection by processing external user-provided data such as document templates, CSV mapping files, and evidence files.\n
Ingestion points: The skill reads user-provided DOCX templates, CSV data, and image/PDF evidence files (e.g., fill_docx.py reads CSV mapping files).\n
Capability inventory: The skill performs file read and write operations using python-docx, pymupdf, and PIL.\n
Boundary markers: No explicit delimiters or instructions are provided to the agent to differentiate between system instructions and potentially malicious data embedded in the user-supplied materials.\n
Sanitization: There is no evidence of sanitization or content validation for the data interpolated into the document templates.

application-bundle