umbraco-preview-app-provider
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill uses the
WebFetchtool to retrieve documentation from official Umbraco domains (docs.umbraco.com). This is a legitimate functional requirement for providing up-to-date guidance on backoffice extensions. - [INDIRECT_PROMPT_INJECTION]: The skill defines a workflow where external documentation is ingested and used to generate code. This creates a surface for indirect prompt injection where malicious content in the documentation could attempt to influence the agent's output. However, the ingestion is limited to trusted official vendor documentation, significantly reducing the risk profile.
- Ingestion points: Documentation URLs fetched via
WebFetchinSKILL.md. - Boundary markers: None explicitly defined in the workflow for the tool output.
- Capability inventory: The skill has
WriteandEditpermissions to modify the local file system. - Sanitization: None explicitly defined; the skill relies on the agent's internal reasoning to interpret the fetched documentation.
Audit Metadata