aml
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is a standard configuration and instruction set for a declarative modeling language (AML). All operations are consistent with its stated purpose of data modeling.
- [COMMAND_EXECUTION]: The skill uses the
anfraCLI tool for legitimate purposes, such as validating AML code (anfra validate) and inspecting database schemas (anfra query --lang sql). The instructions include specific guidance to prevent misuse of the query tool for data retrieval instead of modeling. - [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data in the form of
.amlfiles from the repository. However, it treats these as structured configuration files for theanfratool. The capability to execute SQL is limited to schema exploration and is a documented feature of the environment.
Audit Metadata