grill-with-docs
Pass
Audited by Gen Agent Trust Hub on Aug 28, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The analyzed files consist entirely of configuration and high-level instructions without any implementation logic, scripts, or binary executables.
- [SAFE]: There are no indicators of prompt injection, data exfiltration patterns, or obfuscated content. The skill operates within a restricted environment as indicated by the 'disable-model-invocation' setting.
- [INDIRECT_PROMPT_INJECTION]: The skill is intended to process user designs and plans. While this represents a data ingestion surface, the lack of powerful capabilities (such as shell execution, network access, or file system writes) in the provided files means there is no significant risk of exploitation. No specific sanitization or boundary markers are defined in these metadata files, but the overall posture remains safe.
Audit Metadata