omh-product-brief

Pass

Audited by Gen Agent Trust Hub on Sep 25, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructions include the execution of the omh CLI tool (omh runtime record) to log execution status and delegation results within the environment. This tool is a native component of the 'Oh My Hermes' (OMH) ecosystem associated with the skill vendor.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted 'product evidence' and other user-provided inputs to generate requirements documents.
  • Ingestion points: Untrusted data enters the agent context through the Required inputs section in SKILL.md, specifically via product evidence and problem and user fields.
  • Boundary markers: The skill lacks explicit structural delimiters (e.g., XML tags or specific markers) or 'ignore' instructions for the processed evidence, though safety rules recommend separating evidence from assumptions.
  • Capability inventory: The skill possesses shell command execution capabilities via the omh utility.
  • Sanitization: No explicit sanitization, filtering, or validation logic is defined for the external product evidence before it is incorporated into the planning process.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 25, 2026, 08:34 PM
Security Audit — agent-trust-hub — omh-product-brief