omni-model-explorer
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes model definitions (YAML) and AI context metadata retrieved from the Omni API.
- Ingestion points: Metadata is ingested via
omni models yaml-getandomni models get-topiccommands described inSKILL.md. - Boundary markers: No explicit delimiters are used to wrap external data, though results are intended for user inspection.
- Capability inventory: The skill utilizes the
omniCLI for reading and writing model configurations. - Sanitization: No specific sanitization of the retrieved YAML or context data is performed within the instructions.
- [EXTERNAL_DOWNLOADS]: The skill references the Omni CLI documentation and source code on GitHub.
- Finding: Documentation points to
github.com/exploreomni/cli, which is the official repository for the vendor's command-line interface.
Audit Metadata