docx-manipulation

Pass

Audited by Gen Agent Trust Hub on Mar 10, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of the python-docx library via pip install python-docx. This is a well-known and widely used library for Microsoft Word document manipulation.
  • [COMMAND_EXECUTION]: The skill's core functionality relies on generating and executing Python scripts to perform document operations. While this is the intended behavior of the skill, it represents a standard capability that must be monitored.
  • [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection. Specifically, the 'Mail Merge Pattern' and template filling capabilities ingest external data into document paragraphs without sanitization. If a document being processed or the data being injected contains malicious instructions, the agent might inadvertently follow them during subsequent processing steps.
  • Ingestion points: Document('existing.docx') and fill_template(template_path, replacements) ingest data from external files and dictionary objects.
  • Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the provided code examples.
  • Capability inventory: The skill utilizes doc.save() for file writing and requires an environment capable of executing Python code.
  • Sanitization: Simple string replacement (para.text.replace) is used for template filling, which does not sanitize against potential instruction injection within the replacement strings.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 10, 2026, 05:17 AM