devexpress-office-file-api-presentation
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides functionality to read and extract content from external PowerPoint files (.pptx, .pptm, etc.), which represents a potential attack surface.
- Ingestion points: Files are loaded into the agent context via the
Presentationclass constructors (file streams, byte arrays) as seen inSKILL.mdandreferences/getting-started.md. - Boundary markers: The provided examples for text extraction (e.g., in
references/advanced-features.md) do not include specific delimiters or warnings to the agent to ignore instructions that might be embedded in the document text. - Capability inventory: The skill possesses capabilities for file system modification (
SaveDocument,ExportToPdf,Savefor images) and printing (Print), which could be targets for instructions found in processed documents. - Sanitization: There is no explicit mention of sanitizing or escaping extracted text before it is used in subsequent operations or prompts.
Audit Metadata