omh-capability-toggle
Pass
Audited by Gen Agent Trust Hub on Nov 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes the
omh capability-policycommand to modify the local OMH configuration. This is the primary function of the skill and uses the vendor's own CLI tool for configuration management. - [INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it processes natural language user input to drive command execution.
- Ingestion points: The skill reads user prompts (e.g., 'turn off memory', 'enable coding orchestration') from the agent context.
- Boundary markers: No specific technical delimiters are defined, but the instructions require the agent to use canonical IDs for families.
- Capability inventory: The skill has the capability to execute shell commands via the
omhtool and read files viaskill_view. - Sanitization: The skill implements sanitization by instructing the agent to never guess IDs, list options when ambiguous, and use a completion checklist to verify the correct canonical ID is used before execution.
Audit Metadata