Skills Reference

All skills with category grouping and $name usage syntax.

Canonical Workflow

1. $deep-interview - clarify intent, boundaries, and non-goals when the task is still fuzzy.
2. $ralplan - approve the plan and tradeoffs before execution; completion/skipping requires recorded Architect review evidence followed by Critic review evidence, not just PRD/test-spec files. Scholastic is available as a separate advisory ontology reviewer/persona for concept-heavy planning evidence.
3. $prometheus-strict - harden high-risk plans with Metis clarification, Momus critique, and Oracle synthesis.
4. $ultragoal - execute and verify through durable goal/ledger checkpoints; launch $team only when parallel lanes are warranted.
5. $code-review then $ultraqa - finish with merge-readiness and adversarial QA gates.

Execution Modes

• $team - coordinated multi-agent execution for the approved plan
• $ralph - persistence loop until completion
• $autopilot - strict autonomous loop over $deep-interview -> $ralplan -> $ultragoal (+ $team if needed) -> $code-review -> $ultraqa, returning to planning when review or QA is not clean and never leaving ralplan until Architect/Critic consensus evidence is recorded
• $ultrawork - max parallel execution
• $ultragoal - durable multi-goal planning and execution over .omx/ultragoal artifacts plus Codex goal handoff
• $visual-verdict - structured visual QA loop for screenshot/reference matching
• $ecomode - token-efficient model routing
• $ultraqa - test/verify/fix loop
• $swarm - team compatibility facade

Planning & Clarification

• $deep-interview - requirements clarification before planning or execution
• $ralplan - planner/architect/critic consensus planning with durable Architect→Critic approval evidence before execution handoff; may cite separate Scholastic advisory evidence for concept-heavy plans
• $plan - supporting strategic planning when you are not using the canonical path
• $prometheus-strict - clean-room interview-driven planner that writes durable plans under .omx/plans/prometheus-strict/ when warranted and hands off to $ultragoal/$team; credit: Inspired by OMO Prometheus (code-yeongyu/oh-my-openagent), reimplemented from concept under MIT.
• $best-practice-research - bounded official/upstream best-practice evidence wrapper when current external guidance affects correctness

Research and Planning Boundaries

• $best-practice-research is the ordinary pre-planning wrapper for external official docs, upstream references, standards, release notes, and best-practice checks. It produces cited evidence for a caller; it is not the architecture plan.
• $autoresearch is for a bounded validator-gated research deliverable or artifact. Use it when the research output itself must pass an explicit evaluator before the loop can finish.
• $autoresearch-goal is the Codex goal-mode version for research missions that need durable goal focus plus professor/critic-style validation.
• $ralplan is where architecture and implementation planning are synthesized from requirements and prior evidence. Autoresearch findings gathered before planning should feed into $ralplan as evidence, not appear as a final architecture component unless the user explicitly asks for ongoing research automation.

Agent Shortcuts

• $analyze
• $deepsearch
• $tdd
• $build-fix
• $code-review
• $security-review
• $design - maintain repo-local DESIGN.md
• $visual-ralph - measured visual-reference implementation loop
• $frontend-ui-ux (deprecated; use $design or $visual-ralph)
• $git-master
• $review

Utilities

• $cancel, $doctor, $help, $note, $trace
• $skill
• $hud, $omx-setup, $configure-notifications
• $ralph-init
• $worker (internal) - team worker protocol / lifecycle

Usage Examples

$deep-interview "clarify the auth change"
$ralplan "approve the auth plan and review tradeoffs"
$prometheus-strict "stress-test the plan before durable execution"
$ultragoal "carry the approved plan to completion"
$team 3:executor "execute the active Ultragoal story in parallel"
$ralph "carry the approved plan to completion with the explicit legacy loop"
$best-practice-research "find current official best practices for this API"
$code-review "review current branch"
$doctor

Visual Verdict Guidance

• Use for visual tasks with reference image(s) + generated screenshot.
• Pass threshold target: 90+.
• Use pixel diff / pixelmatch overlays only as secondary debugging support.