data-engineering
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [NO_CODE]: The skill provides original method guidance, markdown documentation, templates, and reference materials. There are no actual executable scripts (.py, .js, .sh), system binaries, or external tools shipped within this folder.
- [EXTERNAL_DOWNLOADS]: Evaluated all reference URLs. The
source_repoinSKILL.mdpoints to the author's own repository (https://github.com/magnus919/hermes-profiles), and other URLs listed in documentation point to standard well-known domains (such asneo4j.com,getdbt.com, andgithub.com/dbt-labs). No malicious package downloads or risky paste sites were identified. - [INDIRECT_PROMPT_INJECTION]: The skill explicitly documents validation structures for model-assisted pipelines (
references/ai-transformation-boundaries.md), stating best-practice behaviors like 'Treat retrieved or source text as untrusted data, not instructions' and enforcing strict isolation boundary rules between generation and execution.
Audit Metadata