run

星标0

分支0

更新时间2026年7月15日 07:34

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.

安装

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。

来源

dxlbnl

dxlbnl/list

打开 GitHub 仓库查看创作者相关仓库

下载

Run — sprint execution

Execute every open task in wiki/sprint.md to completion. The sprint is already composed — go straight to work.

Per card

Read the card from wiki/backlog/<slug>.md — it is the spec. (The retrieval hook will suggest relevant atoms.)
Retrieve — read the suggested atoms + their links before touching any file. Always read wiki/knowledge/project/the-rules.md. The knowledge gate blocks work until current knowledge is read.
Do it — write the minimum code that satisfies the card's acceptance criteria. Right-size tests to the project's testing policy (from the-rules or testing atoms).
Verify — run the project's check/test commands. Fix red before closing.
Capture — write any durable, reusable learning as a knowledge atom (update an existing atom if one covers the topic — one canonical home). Routine edits → one-line "nothing durable."
Commit immediately — stage the files changed for this card (explicit paths — never git add -A) and commit <type>(<slug>): <title>. One commit per card, no exceptions.
Close — delete the card, tick the task in wiki/sprint.md, append one line to the run log.

Repeat for every open task. If a hard blocker appears (missing context, unresolvable dependency, real fork), log it in the run log and continue with the next card — do not stop the sprint.

Sprint close

When all tasks are done:

Retro briefly — what went well, what dragged, what recurred. Durable lessons → atoms; recurring failure-shapes → wiki/knowledge/project/known-issues.md.
Archive to wiki/sprint-archive/NNNN-slug.md — preserve the run log as the spine, enrich with goal, closed cards, and atom links.
Clear wiki/sprint.md for the next sprint.
Commit the close — stage archive + cleared sprint + any atoms captured in the retro, commit chore(sprint-close): retro + archive NNNN-slug.

Rules

All file paths relative to the project root (or worktree root if running as sprint-runner).
git add -A is never allowed — stage explicit paths only.
Knowledge lives in atoms, never the run log.
New work discovered mid-card → /intake a card; do not inline-patch.

同仓库更多 Skills

同仓库

bootstrap

dxlbnl/list

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".

2026-07-150

intake

dxlbnl/list

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.

2026-07-150

loop

dxlbnl/list

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.

2026-07-150

migrate-vibin

dxlbnl/list

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.

2026-07-150

orchestrate

dxlbnl/list

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.

2026-07-150

status

dxlbnl/list

One-screen v2 status — current sprint tasks + run log tail, backlog cards, knowledge-graph size. Read-only.

2026-07-150

Run — sprint execution

Execute every open task in wiki/sprint.md to completion. The sprint is already composed — go straight to work.

Per card

Read the card from wiki/backlog/<slug>.md — it is the spec. (The retrieval hook will suggest relevant atoms.)

Retrieve — read the suggested atoms + their links before touching any file. Always read wiki/knowledge/project/the-rules.md. The knowledge gate blocks work until current knowledge is read.

Do it — write the minimum code that satisfies the card's acceptance criteria. Right-size tests to the project's testing policy (from the-rules or testing atoms).

Verify — run the project's check/test commands. Fix red before closing.

Capture — write any durable, reusable learning as a knowledge atom (update an existing atom if one covers the topic — one canonical home). Routine edits → one-line "nothing durable."

Commit immediately — stage the files changed for this card (explicit paths — never git add -A) and commit <type>(<slug>): <title>. One commit per card, no exceptions.

Close — delete the card, tick the task in wiki/sprint.md, append one line to the run log.

Repeat for every open task. If a hard blocker appears (missing context, unresolvable dependency, real fork), log it in the run log and continue with the next card — do not stop the sprint.

Sprint close

When all tasks are done:

Retro briefly — what went well, what dragged, what recurred. Durable lessons → atoms; recurring failure-shapes → wiki/knowledge/project/known-issues.md.

Archive to wiki/sprint-archive/NNNN-slug.md — preserve the run log as the spine, enrich with goal, closed cards, and atom links.

Clear wiki/sprint.md for the next sprint.

Commit the close — stage archive + cleared sprint + any atoms captured in the retro, commit chore(sprint-close): retro + archive NNNN-slug.

Rules

All file paths relative to the project root (or worktree root if running as sprint-runner).

git add -A is never allowed — stage explicit paths only.

Knowledge lives in atoms, never the run log.

New work discovered mid-card → /intake a card; do not inline-patch.

name	run
description	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.