omni-ai-eval

Pass

Audited by Gen Agent Trust Hub on Nov 11, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the omni CLI to perform various tasks including configuration management (omni config), authentication (omni whoami), and AI evaluation (omni ai-eval). These are standard operations for the tool's intended purpose.
  • [EXTERNAL_DOWNLOADS]: The documentation references the official GitHub repository for the Omni CLI (https://github.com/exploreomni/cli) and the vendor's official documentation (https://docs.omni.co). These are trusted resources belonging to the skill's author.
  • [INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it processes untrusted natural language data. 1. Ingestion points: Prompts are ingested from the evals/files/cases.jsonl file and via the prompt_text field in CLI commands within SKILL.md. 2. Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in provided instructions. 3. Capability inventory: The skill can execute shell commands via the omni CLI and perform network requests to the vendor's domain. 4. Sanitization: No sanitization of the ingested prompt text is mentioned before evaluation. Note: This surface is inherent to its primary function and considered safe.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 11, 2026, 09:26 PM