syncfusion-dotnet-smart-data-extraction

Pass

Audited by Gen Agent Trust Hub on Oct 6, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection because it is intended to parse and process untrusted external data (PDFs and images) while maintaining the capability to modify critical project files.
  • Ingestion points: The skill processes user-supplied PDF documents and image formats (JPEG, PNG) as described in SKILL.md and README.md.
  • Boundary markers: The instructions do not specify any delimiters or safety markers to differentiate between the document's content and the agent's instructions during the processing phase.
  • Capability inventory: The skill is instructed to modify .csproj files and completely overwrite the contents of Program.cs based on user requests (SKILL.md, Step 2 and Step 4).
  • Sanitization: There are no documented procedures for sanitizing or validating the content of the documents before the agent generates code based on their structure or requirements.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 6, 2026, 10:39 AM