syncfusion-dotnet-word

Pass

Audited by Gen Agent Trust Hub on Nov 16, 2026

Risk Level: SAFEDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [DYNAMIC_EXECUTION]: The skill employs a mechanism to generate and run temporary C# scripts (.csx) for document automation tasks. This dynamic execution is governed by instructions that mandate the use of pre-defined templates and restrict generated code to specific APIs documented in the skill's reference files.
  • [INDIRECT_PROMPT_INJECTION]: The skill includes functionality to ingest and process external data formats such as Word, HTML, XML, Markdown, and JSON.
  • Ingestion points: Untrusted data enters the skill's context when opening or converting files from these formats.
  • Boundary markers: There are no explicit markers or prompt-level instructions defined in the snippets to isolate the contents of these files from the AI agent's instructions.
  • Capability inventory: The skill can perform file system operations (write), execute shell commands via dotnet-script, and interact with the workspace.
  • Sanitization: The skill relies on the underlying DocIO library for parsing, but does not specify additional sanitization for document content.
  • [COMMAND_EXECUTION]: The skill uses the dotnet script command to execute the dynamically generated document processing tasks. This execution is restricted to the skill's own internal script directory to prevent interference with workspace files.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 16, 2026, 07:51 PM