一键导入
sdlc
The project's software-delivery workflow — how changes move from idea to merged on main. Use when starting a new task, committing, or opening/updating a PR.
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。
菜单
The project's software-delivery workflow — how changes move from idea to merged on main. Use when starting a new task, committing, or opening/updating a PR.
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。
基于 SOC 职业分类
Seed the auto-engineer SDLC toolkit into a project. Asks for the target project path, detects the tech stack and GitHub configuration, collects any missing values interactively, then writes customized skill files and Docker infrastructure into that project. Use when the user says "seed a project", "set up auto-engineer", "install the loop", or invokes `/seed`.
Two-way sync between a seeded project and this auto-engineer repo's templates — covers skills, Dockerfile, and scripts. Detects drift in either direction, presents per-file diffs, and lets you pull from template, push to template, or skip. Harness-internal tool — never seeded into target projects. Invoked as /sync.
Autonomously drive project work end-to-end — pick an unblocked issue, plan it, implement, push a PR, resolve CI and review-bot feedback, merge when green, then repeat. Use when the user says "auto-engineer", "auto-pilot", "go run the loop", or invokes `/auto-engineer`.
Drive a large body of work end-to-end — either from a single scoping issue or a fuzzy topic ("ship GA", "harden the review app") — by discovering scope in the backlog + codebase, filing sub-issues, iterating with the user, planning parallel workstreams, then spawning `/auto-engineer` subagents to ship the whole epic autonomously. Use when the user says "auto-manager", "manage this epic", "orchestrate /auto-engineer across…", or invokes `/auto-manager`.
File a new tracker issue with the correct priority and area labels applied up front. Use whenever an agent needs to capture a new unit of work — follow-ups from a merged PR, TODOs discovered while coding, or manually-requested issues. Invoked as `/file-issue`.
Autonomously drive project work end-to-end — pick an unblocked issue, plan it, implement, push a PR, resolve CI and review-bot feedback, merge when green, then repeat. Use when the user says "auto-engineer", "auto-pilot", "go run the loop", or invokes `/auto-engineer`.
| name | sdlc |
| description | The project's software-delivery workflow — how changes move from idea to merged on main. Use when starting a new task, committing, or opening/updating a PR. |
Read docs/agent-playbooks/sdlc.md first for repo-level branch, commit, push, and PR policy.
Read docs/agent-playbooks/pr-review.md when the task includes CI interpretation or review response.
Use this skill when the task involves:
Create a branch per unit of work. Conventional patterns:
m<issue-number>-<slug> for issue-tracked work<verb>-<noun> for untracked changes (e.g. fix-login-timeout, add-metrics-endpoint)Always branch from an up-to-date main:
git checkout main && git pull
git checkout -b <branch>
Add, Fix, Remove, not Added, Fixed).Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
main.Use the standard Summary + Test plan body structure:
## Summary
- <bullet describing what changed and why>
## Test plan
- [ ] <concrete test step>
- [ ] <edge case to verify>
Closes #<N>
Open via mcp__github__create_pull_request. No "generated with Claude" footer.
wait-for-pr for Claude-specific polling and automation.docs/agent-playbooks/pr-review.md.This project has no review bots or CI tests, so every PR must get an independent review before merge. Before requesting merge:
Agent tool with a prompt framing it as a Senior prompt engineer reviewing the diff for correctness, security, performance, and adherence to project conventions.gh pr diff <N>); instruct it to report blocking issues, in-scope nits, and out-of-scope suggestions separately, and to keep its reply under 400 words.Never merge a PR for this project without completing the self-review cycle.
This repo is a toolkit factory. There are two copies of most skills:
| Path | What it is |
|---|---|
templates/skills/<skill>/SKILL.md | The canonical source shipped to user projects via seed |
.claude/skills/<skill>/SKILL.md | The local instance used to develop this repo |
Issues and feature work always target templates/. The local instance (.claude/skills/) is only touched when the task is explicitly about the development harness itself.
Before editing any skill file, confirm it lives under templates/skills/. If you find yourself reaching for .claude/skills/, stop and re-read the issue — you almost certainly have the wrong file.
Exception — seed: The seed skill does not seed itself. Its canonical source lives at .claude/skills/seed/SKILL.md (there is no templates/skills/seed/). Issue work targeting seed goes to .claude/skills/seed/.
main..claude/skills/<skill>/ for issue work — always edit the corresponding templates/skills/<skill>/ file instead (except seed, which has no template copy).