omh-model-optimization

Pass

Audited by Gen Agent Trust Hub on Sep 25, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides instructions for the agent to use framework-specific CLI tools, such as omh coding model-route, omh model-chains set, and omh runtime record. These tools are intended for model identification, routing configuration, and status tracking within the 'Oh My Hermes' environment.\n- [INDIRECT_PROMPT_INJECTION]: The skill defines a research-based workflow that ingests data from external "official documentation" and "community harness findings."\n
  • Ingestion points: SKILL.md (instructions to gather external research and release notes).\n
  • Boundary markers: The agent is instructed to label research findings (official, official-client, observed, community) to maintain provenance, which provides basic context separation.\n
  • Capability inventory: Framework-specific CLI tool execution and local file path access for reference rules.\n
  • Sanitization: The skill includes data validation rules for pricing (requiring documented sources) but does not define explicit content sanitization for general research findings.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 25, 2026, 08:34 PM
Security Audit — agent-trust-hub — omh-model-optimization