omh-toolbelt-readiness

Pass

Audited by Gen Agent Trust Hub on Sep 25, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process user-provided data such as 'user request' and 'target context' to determine necessary toolsets.
  • Ingestion points: Untrusted data enters the agent context through the 'user request' and 'target context' fields defined in the inputs section.
  • Boundary markers: The skill includes instructions to 'Separate prepared guidance from observed platform... evidence' and to establish 'not-evidence boundaries', which serve as logical constraints within the prompt.
  • Capability inventory: The skill invokes a local CLI utility omh for state management (omh runtime record). No direct file-system modifications or external network requests are specified within the provided instructions.
  • Sanitization: There is no mention of programmatic sanitization, filtering, or escaping for the user-provided text before it is interpolated into the workflow analysis.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 25, 2026, 08:34 PM
Security Audit — agent-trust-hub — omh-toolbelt-readiness