/wave-execute
EXECUTE a lean wave — plan story DAG + file-ownership batches via CLI, dispatch full-sdc per story, fan-in check, hand off merge to @devops. Use when: wave-execute, run wave, epic wave, parallel stories SDC.
$ npx -y skills add SynkraAI/aiox-core --skill wave-execute --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
- Fires itselfAuto-invocation. Claude auto-loads it when your prompt matches the work.Auto-invocation is when the right skill fires by itself at the right moment, driven by a FLOW.md router and a hook, instead of you invoking it by name. It is the difference between a skill being installed and a skill actually getting used.Read the full definition →
- You can call itInvoke it directly when you want it.
- Slash command
/wave-execute
Context preview
The summary Claude sees to decide when to auto-load this skill.
EXECUTE a lean wave — plan story DAG + file-ownership batches via CLI, dispatch full-sdc per story, fan-in check, hand off merge to @devops. Use when: wave-execute, run wave, epic wave, parallel stories SDC.
SKILL.md
wave-execute.SKILL.mdname: wave-execute
description: >
EXECUTE a lean wave — plan story DAG + file-ownership batches via CLI,
dispatch full-sdc per story, fan-in check, hand off merge to @devops.
Use when: wave-execute, run wave, epic wave, parallel stories SDC.
user-invocable: true
argument-hint: "{wave-id} --stories s1.md,s2.md [yolo|interactive] [--dry-run] [--no-confirm]"
agent: aiox-master
depends_on: ["full-sdc"]wave-execute (lean EXECUTE)
Run a **wave of stories**: deterministic plan (DAG + file partition) → dispatch `/full-sdc` (or `aiox sdc` + skills) per story → fan-in → `@devops` merge-back.
**Not in scope:** cockpit `wave launch`, conductor loops, product worktree WL registry, companion spawn surface.
Depends on skill **full-sdc** and CLI `aiox wave` / `aiox sdc`.
Invocation
/wave-execute {wave-id} --stories path1.md,path2.md [yolo|interactive] [--dry-run] [--no-confirm]
# pure CLI plan
aiox wave plan --stories path1.md,path2.md --wave-id {wave-id} --mode yolo --save
aiox wave next {wave-id}Yolo/parallel dispatch requires a positive explicit `AIOX_MODEL_BUDGET_CEILING_USD`. Each child intent is scanned and bound to its existing story before a model executor is invoked.
CLI (mechanical)
# From epic directory (C3) — preferred
aiox wave from-epic --epic-dir docs/framework/epics/core-super-update \
--filter 'CORE-SU.C' --wave-id CORE-SU-C --mode yolo
# Or explicit paths
aiox wave plan --stories a.md,b.md,c.md --wave-id WAVE-1 --mode yolo --save
# Advance / next batch (auto-completes stories already Done / sdc completed)
aiox wave advance WAVE-1
aiox wave next WAVE-1
# After a child full-sdc finishes or fails
aiox wave mark WAVE-1 {story-id} --status completed
aiox wave mark WAVE-1 {story-id} --status failed --notes "qg breaker"
# Report
aiox wave report WAVE-1
aiox wave status WAVE-1
# Per story (child full-sdc)
aiox sdc plan {story} --mode yolo
aiox sdc next {story}
# … run skill for phase …
aiox sdc verify {story} {phase} --markState: `.aiox/waves/{wave-id}/state.json` Controller: `wave-run.js` + `dispatch-adapter.js` (C2) + `epic-glue.js` (C3)
EXECUTE stages
Stage 1 — Preflight (CLI computes)
1. Resolve story paths (must exist). 2. Before automated dispatch, validate the shared budget ceiling and scan each story-bound child intent; any rejection blocks that child before model use. 3. `aiox wave plan … --save` — do **not** re-invent the DAG by grepping in prose. 4. Judge the plan: Ready stories only for develop-heavy waves; flag Draft that still need validate; halt if `executor == quality_gate` when those fields exist. 5. `--dry-run` → print plan and **stop**.
Stage 2 — Confirm
Show batches (parallel vs sequenced by file ownership). Get human OK unless `--no-confirm` or mode `yolo`.
Stage 3 — Dispatch batches
For each batch **in order**:
1. For **each story in the batch** (parallel when multi-agent available):
- Run **full-sdc** execute protocol (skill `full-sdc` / `/aiox-full-sdc`)
- Preferred: `spawn_subagent` with `aiox-master` (or dedicated coordinator) prompt:
`Execute full-sdc on {story-path} mode={mode}. Use aiox sdc plan/next/verify. Do not git push.`
- Sequential fallback: run full-sdc inline one story at a time
2. Wait for all stories in the batch to reach SDC `completed` (or HALT). 3. On a blocked/failed story: do **not** dispatch dependents still waiting on it; continue independent later batches only if their deps are satisfied. 4. Cascade: if story B `dependsOn` A and A failed → mark B blocked in the wave report (do not fake green).
Stage 4 — Fan-in
Before merge:
- Re-check File List overlap across branches/stories that ran in parallel (partition should prevent; verify).
- Surface conflicts to human + `@devops`; never silent overwrite.
Stage 5 — Merge-back
Hand off to **`@devops` only** (push/PR/merge exclusive). Provide handoff:
handoff:
from: wave-execute
to: devops
wave_id: "{wave-id}"
stories: ["…"]
branches: ["feat/…"]
next_action: "create/merge PRs per repo policy"Partition rules (same as CLI)
- Topological sort on `depends_on` (deps outside the wave = already satisfied).
- Within a ready set: max non-overlapping File List subset runs in one batch; overlapping remainder in later batches.
- Empty File List → treated as non-conflicting (still may be sequential by deps).
Modes
| Mode | Behavior | |------|----------| | `interactive` | Confirm plan; pause between batches | | `yolo` | Auto-confirm plan; autonomous full-sdc children | | `--dry-run` | Plan only |
Blocking conditions
- Wave plan `errors` non-empty (cycle)
- Story path missing
- full-sdc integrity HALT (Done before close)
- QG circuit breaker on a story
- Fan-in conflict unresolved
Strip checklist
- [x] No cockpit-only spawn commands required
- [x] CLI First plan/partition
- [x] full-sdc children only (no invented parallel AC)
- [x] @devops exclusive merge
Read more
name: wave-execute
description: >
EXECUTE a lean wave — plan story DAG + file-ownership batches via CLI,
dispatch full-sdc per story, fan-in check, hand off merge to @devops.
Use when: wave-execute, run wave, epic wave, parallel stories SDC.
user-invocable: true
argument-hint: "{wave-id} --stories s1.md,s2.md [yolo|interactive] [--dry-run] [--no-confirm]"
agent: aiox-master
depends_on: ["full-sdc"]wave-execute (lean EXECUTE)
Run a **wave of stories**: deterministic plan (DAG + file partition) → dispatch `/full-sdc` (or `aiox sdc` + skills) per story → fan-in → `@devops` merge-back.
**Not in scope:** cockpit `wave launch`, conductor loops, product worktree WL registry, companion spawn surface.
Depends on skill **full-sdc** and CLI `aiox wave` / `aiox sdc`.
Invocation
/wave-execute {wave-id} --stories path1.md,path2.md [yolo|interactive] [--dry-run] [--no-confirm]
# pure CLI plan
aiox wave plan --stories path1.md,path2.md --wave-id {wave-id} --mode yolo --save
aiox wave next {wave-id}Yolo/parallel dispatch requires a positive explicit `AIOX_MODEL_BUDGET_CEILING_USD`. Each child intent is scanned and bound to its existing story before a model executor is invoked.
CLI (mechanical)
# From epic directory (C3) — preferred
aiox wave from-epic --epic-dir docs/framework/epics/core-super-update \
--filter 'CORE-SU.C' --wave-id CORE-SU-C --mode yolo
# Or explicit paths
aiox wave plan --stories a.md,b.md,c.md --wave-id WAVE-1 --mode yolo --save
# Advance / next batch (auto-completes stories already Done / sdc completed)
aiox wave advance WAVE-1
aiox wave next WAVE-1
# After a child full-sdc finishes or fails
aiox wave mark WAVE-1 {story-id} --status completed
aiox wave mark WAVE-1 {story-id} --status failed --notes "qg breaker"
# Report
aiox wave report WAVE-1
aiox wave status WAVE-1
# Per story (child full-sdc)
aiox sdc plan {story} --mode yolo
aiox sdc next {story}
# … run skill for phase …
aiox sdc verify {story} {phase} --markState: `.aiox/waves/{wave-id}/state.json` Controller: `wave-run.js` + `dispatch-adapter.js` (C2) + `epic-glue.js` (C3)
EXECUTE stages
Stage 1 — Preflight (CLI computes)
1. Resolve story paths (must exist). 2. Before automated dispatch, validate the shared budget ceiling and scan each story-bound child intent; any rejection blocks that child before model use. 3. `aiox wave plan … --save` — do **not** re-invent the DAG by grepping in prose. 4. Judge the plan: Ready stories only for develop-heavy waves; flag Draft that still need validate; halt if `executor == quality_gate` when those fields exist. 5. `--dry-run` → print plan and **stop**.
Stage 2 — Confirm
Show batches (parallel vs sequenced by file ownership). Get human OK unless `--no-confirm` or mode `yolo`.
Stage 3 — Dispatch batches
For each batch **in order**:
1. For **each story in the batch** (parallel when multi-agent available):
- Run **full-sdc** execute protocol (skill `full-sdc` / `/aiox-full-sdc`)
- Preferred: `spawn_subagent` with `aiox-master` (or dedicated coordinator) prompt:
`Execute full-sdc on {story-path} mode={mode}. Use aiox sdc plan/next/verify. Do not git push.`
- Sequential fallback: run full-sdc inline one story at a time
2. Wait for all stories in the batch to reach SDC `completed` (or HALT). 3. On a blocked/failed story: do **not** dispatch dependents still waiting on it; continue independent later batches only if their deps are satisfied. 4. Cascade: if story B `dependsOn` A and A failed → mark B blocked in the wave report (do not fake green).
Stage 4 — Fan-in
Before merge:
- Re-check File List overlap across branches/stories that ran in parallel (partition should prevent; verify).
- Surface conflicts to human + `@devops`; never silent overwrite.
Stage 5 — Merge-back
Hand off to **`@devops` only** (push/PR/merge exclusive). Provide handoff:
handoff:
from: wave-execute
to: devops
wave_id: "{wave-id}"
stories: ["…"]
branches: ["feat/…"]
next_action: "create/merge PRs per repo policy"Partition rules (same as CLI)
- Topological sort on `depends_on` (deps outside the wave = already satisfied).
- Within a ready set: max non-overlapping File List subset runs in one batch; overlapping remainder in later batches.
- Empty File List → treated as non-conflicting (still may be sequential by deps).
Modes
| Mode | Behavior | |------|----------| | `interactive` | Confirm plan; pause between batches | | `yolo` | Auto-confirm plan; autonomous full-sdc children | | `--dry-run` | Plan only |
Blocking conditions
- Wave plan `errors` non-empty (cycle)
- Story path missing
- full-sdc integrity HALT (Done before close)
- QG circuit breaker on a story
- Fan-in conflict unresolved
Strip checklist
- [x] No cockpit-only spawn commands required
- [x] CLI First plan/partition
- [x] full-sdc children only (no invented parallel AC)
- [x] @devops exclusive merge
🌐 README por idioma: EN | PT | ES | ZH Devolvendo às pessoas o poder de criar — Framework open source de orquestração de IA que devolve o controle a quem tem coragem de construir.
Repo: SynkraAI/aiox-core
Other skills on aiox-core.
- /aiox-master
Activate Orion (aiox-master) for AIOX Master Orchestrator & Framework Developer. Use when you need comprehensive expertise across all domains, framework component creation/modification, workflow orchestration, or running tasks that don't...
Open skill - /analyst
Activate Atlas (analyst) for Business Analyst. Use for market research, competitive analysis, user research, brainstorming session facilitation, structured ideation workshops, feasibility studies, industry trends analysis, project discov...
Open skill - /architect
Activate Aria (architect) for Architect. Use for system architecture (fullstack, backend, frontend, infrastructure), technology stack selection (technical evaluation), API design (REST/GraphQL/tRPC/WebSocket), security architecture, perf...
Open skill - /data-engineer
Activate Dara (data-engineer) for Database Architect & Operations Engineer. Use for database design, schema architecture, Supabase configuration, RLS policies, migrations, query optimization, data modeling, operations, and monitoring
Open skill - /dev
Activate Dex (dev) for Full Stack Developer. Use for code implementation, debugging, refactoring, and development best practices
Open skill - /devops
Activate Gage (devops) for GitHub Repository Manager & DevOps Specialist. Use for repository operations, version management, CI/CD, quality gates, and GitHub push operations. ONLY agent authorized to push to remote repository.
Open skill

