umbraco-skill-author
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides instructions for running dotnet test and local Python scripts to validate code and lint manifests.
- [EXTERNAL_DOWNLOADS]: Guidance encourages the agent to fetch documentation from Umbraco's official domain to ensure code matches current source of truth.
- [INDIRECT_PROMPT_INJECTION]: 1. Ingestion points: External documentation from docs.umbraco.com (references/authoring-steps.md). 2. Boundary markers: None specified for external content. 3. Capability inventory: File system writes, code generation, and test execution (dotnet test). 4. Sanitization: No explicit validation or filtering mentioned for fetched content.
- [DYNAMIC_EXECUTION]: The skill uses generate-examples.py to dynamically assemble executable code from assets and manifests during the build process.
Audit Metadata