skills/lovstudio/skills/lov-fill-form/Gen Agent Trust Hub

lov-fill-form

Pass

Audited by Gen Agent Trust Hub on Oct 18, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The scripts/fill_form.py script executes system utilities (textutil on macOS and libreoffice or soffice as a fallback) to convert legacy .doc files to the .docx format. These operations are performed securely using argument lists in subprocess.run(), which prevents shell injection vulnerabilities even when handling user-provided file paths.
  • [EXTERNAL_DOWNLOADS]: The skill requires the python-docx library, a well-known and standard package for programmatically interacting with Microsoft Word documents. Installation instructions utilize standard package management via python3 -m pip.
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses a processing surface for indirect prompt injection because it ingests and parses content from user-provided Word templates. This is an inherent risk of the skill's primary purpose.
  • Ingestion points: The scripts/fill_form.py script reads template files via python-docx to identify table labels and paragraph text.
  • Boundary markers: Detected fields are presented to the agent as a list and JSON structure. No explicit instructions are given to the agent to disregard instructions potentially embedded within the document text.
  • Capability inventory: The skill has the capability to write documents to the local filesystem using the doc.save() method.
  • Sanitization: The script performs basic whitespace normalization on label text but does not implement sanitization or escaping of document content.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 18, 2026, 04:28 PM