docx-processing
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill possesses a data ingestion surface intended for document templating and mail merge operations.
- Ingestion points: The skill is designed to process data from JSON, CSV, database, and API sources as specified in Phase 1 and the Mail Merge implementation snippet (SKILL.md).
- Boundary markers: There are no explicit instructions or delimiters provided to prevent the agent from following instructions potentially embedded within the ingested data sources.
- Capability inventory: The skill utilizes file system write operations (doc.save and tpl.save) and shell command execution (subprocess.run) across its implementation patterns (SKILL.md).
- Sanitization: No explicit data sanitization or escaping mechanisms are described for the interpolated data beyond standard Jinja2 rendering provided by docxtpl.
- [COMMAND_EXECUTION]: The skill utilizes the subprocess module to perform document format conversions.
- Evidence: The 'Conversion Strategies' section in SKILL.md demonstrates the use of subprocess.run to invoke 'libreoffice' for converting DOCX files to PDF. This is a standard and expected implementation for the skill's stated purpose.
Audit Metadata