The Agent Skills Directory

[SAFE]: The skill utilizes standard, well-maintained libraries such as pypdf, pdfplumber, and reportlab. All operations are performed on local files, and there are no network requests, hardcoded credentials, or unauthorized file system access.
[COMMAND_EXECUTION]: Documentation provides instructions for using reputable command-line tools like qpdf, pdftotext, and pdftk for PDF manipulation. These tools are standard for the skill's domain.
[PROMPT_INJECTION]: The skill ingests untrusted PDF data which presents an indirect prompt injection attack surface. Evidence: The forms.md instructions have the agent analyze extracted text and images from PDF files. Capability: The agent has the capability to write files and execute local scripts based on its analysis. Sanitization: No explicit sanitization of PDF content is performed. Note: This risk is inherent to document processing and is assessed as SAFE in the context of this skill's primary purpose.
[SAFE]: The script fill_fillable_fields.py includes a monkeypatch for a specific bug in the pypdf library. This is a targeted correction for a data-type mismatch and does not introduce malicious dynamic execution risks.

pdf