菜单
Run multiple sprints in parallel — partition backlog cards, create git worktrees per sprint, spawn background subagents, merge when done. Use when you want to parallelise independent concerns across concurrent sprints.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
Primary entry point for a freshly cloned Vibin seed. Interviews the user, seeds the knowledge graph, scaffolds the stack, writes the permission profile, strips seed-meta, and hands off to the loop skill. Use when starting a new project, or when the user says "bootstrap" or "let's begin".
File a new work item as a card in wiki/backlog/ (v2 — card-is-spec, slug filename, proportional detail). Use whenever the user reports a bug, asks for a feature, or proposes new work mid-task — instead of inline-patching.
Run the Vibin v2 build loop — pull a work item into the sprint, then retrieve → do → verify → capture → review → close, mostly inline. Use to build the backlog or resume work.
Bring an existing Vibin project up to date with the latest seed by diffing the project's recorded seed commit against the latest Vibin on GitHub and applying the changes. Use when the user says "migrate", "migrate-vibin", "upgrade Vibin", "apply the latest seed changes", or after a new seed release. Reads the seed commit hash from .vibin-version, audits the project's actual state, applies the migrations added since, and reconciles seed-owned files while preserving local customizations.
Execute a pre-composed sprint — retrieve, do, verify, capture, commit, close. No planning, no user checkpoints. Used by the loop skill after planning, and by the sprint-runner agent for parallel sprints.
One-screen v2 status — current sprint tasks + run log tail, backlog cards, knowledge-graph size. Read-only.
| name | orchestrate |
| description | Run multiple sprints in parallel — partition backlog cards, create git worktrees per sprint, spawn background subagents, merge when done. Use when you want to parallelise independent concerns across concurrent sprints. |
Partition the backlog into independent sprint streams, set up isolated worktrees, and spawn a background subagent per sprint. You are a launcher and merger — the work happens inside each subagent, which runs its sprint to completion and exits.
wiki/knowledge/index.md and wiki/knowledge/project/the-rules.md.wiki/backlog/.Verify .vibin/sprint/ is in .gitignore before proceeding.
For each approved sprint:
git checkout -b sprint/<slug>
git worktree add .vibin/sprint/<slug> sprint/<slug>
git checkout -
Write wiki/sprint.md into the worktree:
# Sprint — current
## Sprint goal
<one-line goal for this sprint>
## Tasks
- [ ] <card-slug-1>
- [ ] <card-slug-2>
**Next:** start with <card-slug-1>
## Run log
- Sprint composed by orchestrator. Cards: <slugs>.
Copy the sprint's cards into the worktree's backlog:
cp wiki/backlog/<slug>.md .vibin/sprint/<sprint-slug>/wiki/backlog/
For each sprint, spawn a background sprint-runner agent (defined in
.claude/agents/sprint-runner.md) with this prompt:
Worktree root:
<absolute-path>/.vibin/sprint/<slug>Sprint branch:sprint/<slug>Execute the sprint. Follow your agent definition.
All agents run in background concurrently. Report which are running and their worktree paths.
While agents run, check progress:
cat .vibin/sprint/<slug>/wiki/sprint.md # run log
git log sprint/<slug> --oneline # commits
When a background subagent reports done, review its report (cards closed, atoms captured, blockers). Then:
git diff main...sprint/<slug> — summarise what changed.git checkout main
git merge --no-ff sprint/<slug> -m "sprint(<slug>): merge <goal>"
--no-ff to keep the sprint's commit history visible as a group.git worktree remove .vibin/sprint/<slug>
git branch -d sprint/<slug>
Merge one sprint at a time — fully resolve before starting the next merge.
git add -A — parallel worktrees share the repo.