docx-perfect

Warn

Audited by Gen Agent Trust Hub on Mar 13, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes command-line Python execution (python -c) to extract text from Word documents and manage file versions.
  • [COMMAND_EXECUTION]: The workflow involves running a provided Python script (scripts/template.py) that performs local file operations and document styling.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes content from external Word files without sanitization or boundary markers.
  • Ingestion points: Document text is read via the docx library as described in 'SKILL.md' and 'references/workflow.md'.
  • Boundary markers: None identified; document content is not delimited or isolated from the agent's core instructions.
  • Capability inventory: The skill has the ability to read/write local files and execute Python scripts based on processed document data.
  • Sanitization: None identified; all document text is treated as legitimate content for analysis.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 13, 2026, 06:59 AM