devextreme-button
Pass
Audited by Gen Agent Trust Hub on Aug 18, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill includes tools to fetch documentation from external sources, which creates a potential surface for indirect prompt injection. This is proactively mitigated within the skill's instructions.
- Ingestion points: The
devexpress_docs_searchanddevexpress_docs_get_contenttools (SKILL.md). - Boundary markers: The skill contains an explicit warning instructing the agent to treat documentation as untrusted reference data rather than instructions (SKILL.md).
- Capability inventory: The agent generates UI component code and configurations based on the framework and API details (SKILL.md).
- Sanitization: Instructions mandate that external content should never be executed as commands or allowed to override existing system or user instructions (SKILL.md).
- [EXTERNAL_DOWNLOADS]: The skill references documentation and demo URLs hosted on
js.devexpress.com. These belong to the primary vendor and are standard resources for the skill's functionality. - [COMMAND_EXECUTION]: The skill references specific documentation-fetching tools. These tools are used solely for information retrieval to support code generation and do not provide a path for arbitrary shell command execution.
Audit Metadata