syncfusion-wpf-docx-editor
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates a workflow where the AI agent can modify user project files. This creates a potential surface for indirect prompt injection if a user processes malicious content that influences the agent's code generation or file writing behavior. While the skill includes safeguards such as requiring user confirmation and explicit file paths, the underlying capability is present.\n
- Ingestion points: User project files (e.g., MainWindow.xaml.cs) as described in the 'Workflow' section of SKILL.md.\n
- Boundary markers: The skill explicitly instructs the agent to stop and ask for a delivery mode (replace file vs. direct chat) and mandates user confirmation and a specific file path before making changes.\n
- Capability inventory: The skill allows for replacing code in project files (Option 1 in SKILL.md), though it does not spawn arbitrary shell processes.\n
- Sanitization: No specific technical sanitization or validation of the code content or destination paths is defined, relying entirely on the agent following the provided confirmation rules.
Audit Metadata