before-building
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill does not contain any malicious patterns, credentials, or unauthorized execution commands. It follows security best practices by disabling model tool invocation (
disable-model-invocation: true) and explicitly instructing the agent to avoid reading files or using external search tools. - [INDIRECT_PROMPT_INJECTION]: The skill includes an ingestion surface for user-provided ideas. Ingestion points: User input via the build trigger or /before-building command. Boundary markers: No delimiters are defined for user-provided content. Capability inventory: No active capabilities are available, as tool invocation is disabled and file access is forbidden. Sanitization: No input sanitization or filtering is implemented.
Audit Metadata