devexpress-office-file-api-word-processing
Pass
Audited by Gen Agent Trust Hub on Oct 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill enables the agent to read and process external Word documents, which could contain instructions intended to influence agent behavior.
- Ingestion points: Document content is loaded via
RichEditDocumentServer.LoadDocumentand read usingDocument.GetTextor field result inspection (documented inreferences/getting-started.mdandreferences/text-and-paragraphs.md). - Boundary markers: The instructions do not explicitly mandate the use of delimiters when processing extracted text, though they emphasize structured API usage.
- Capability inventory: The skill provides extensive capabilities to write files (
SaveDocument), export content to PDF/HTML, and modify document structures. - Sanitization: A dedicated reference (
references/safe-document-processing.md) provides detailed instructions on mitigating these risks by enforcing loading limits, stripping macros/OLE objects, and sanitizing metadata.
Audit Metadata