umbraco-tree
Fail
Audited by Gen Agent Trust Hub on Feb 15, 2026
Risk Level: HIGHPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [Prompt Injection] (HIGH): The skill is vulnerable to Indirect Prompt Injection (Category 8) because it fetches external content via WebFetch and uses it to generate code without sanitization or boundary markers. Ingestion points: External documentation URLs provided in the skill; Boundary markers: None present to delimit fetched content from instructions; Capability inventory: The skill uses Write and Edit tools to generate .ts files on the local filesystem; Sanitization: No validation or filtering is mentioned before using fetched documentation to produce executable code.
- [Dynamic Execution] (MEDIUM): The skill performs runtime generation of TypeScript files (Category 10) based on external inputs, a pattern that becomes critical if the remote source is compromised.
Recommendations
- AI detected serious security threats
Audit Metadata