The Agent Skills Directory

[SAFE]: The skill uses industry-standard and reputable Python libraries for its core functionality, such as python-pptx for PowerPoint, openpyxl for Excel, and PyMuPDF for PDF processing.
[DATA_EXFILTRATION]: The presence of google-generativeai as a dependency indicates the skill may send document content to Google's Gemini API for summarization or analysis. This targets a well-known service and is consistent with the skill's stated purpose of document analysis.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted data from external files, creating a potential surface for indirect prompt injection.
Ingestion points: Files are read via tools/pptx_ops.py, tools/excel_ops.py, and tools/pdf_page_editor.py (referenced in SKILL.md).
Boundary markers: None. The instructions do not define clear boundaries or provide warnings to the agent about ignoring instructions found within the processed documents.
Capability inventory: The skill has file system access (read/write) and potential network access via official Google SDKs.
Sanitization: There is no evidence of sanitization or content filtering mentioned in the skill instructions.

document-processor