reasoning
Control omi reasoning effort level to balance depth, rigor, latency, and token cost.
Purpose
Use this skill to set an explicit reasoning posture before planning, execution, or verification-heavy turns.
Trigger
- User asks for faster vs deeper reasoning
- Session needs tighter quality gates or lower latency
- Team wants predictable planning/verification effort across stages
Supported Effort Levels
lowmediumhighxhigh
Workflow
- Identify requested effort or infer
mediumas default. - Explain impact on planning depth, verification strictness, and cost/latency.
- Persist/update
.iflow/state/reasoning.jsonwhen filesystem tools are available. - Recommend a next command (
/omi:mode,/omi:team, or/omi:loop) based on posture.
Output Template
## Reasoning
- requested:
- applied:
## Profile
| Dimension | Value |
| --- | --- |
## Next Command
- ...