omni-model-builder
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill uses the official Omni CLI (
omni) to interact with the Omni platform. All external references (such as documentation links) point to the vendor's official GitHub repository atgithub.com/exploreomni/cli. - [SAFE]: The skill implements a robust 'Safe Development Workflow' that mandates creating branches (
omni models create-branch) for any modifications, followed by validation (omni models validate) and query testing (omni query run) before merging. - [SAFE]: The skill explicitly forbids automatic merging or promotion of changes to the production model ('HARD STOP'), requiring the agent to wait for a user-initiated command to ship or merge code.
- [SAFE]: Authentication is handled safely by instructing the agent to hand off to the user for login operations (
omni config login) when unauthorized, rather than attempting to handle credentials or tokens directly. - [SAFE]: The skill includes comprehensive documentation for handling various modeling tasks (views, topics, relationships, etc.) and provides troubleshooting steps for common validation errors.
Audit Metadata