一键导入
alloy-plan
Use when turning context or a clear user request into an implementation-ready plan with acceptance criteria, files, commands, gates, risks, and handoff.
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
Use when turning context or a clear user request into an implementation-ready plan with acceptance criteria, files, commands, gates, risks, and handoff.
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
Use only when the user explicitly allows bounded unattended execution of an approved Alloy task.
Use when encountering a bug, regression, failing command, or unexpected behavior before attempting a fix.
Use before planning when requirements are ambiguous, large, user-visible, or likely to cause rework unless decisions are locked.
Use when implementing an approved Alloy plan. Maintains progress.md, enforces TDD/debug discipline, and stops on blocked gates.
Use when a task needs broad source understanding before planning or execution.
Automated end-to-end QA — read acceptance criteria (Azure DevOps or manual), generate Playwright test plan, execute with screenshots/video/GIF, produce acceptance report. Uses Playwright CLI for deterministic, reportable execution.
| name | alloy-plan |
| description | Use when turning context or a clear user request into an implementation-ready plan with acceptance criteria, files, commands, gates, risks, and handoff. |
Planner uses this skill to write the Builder handoff contract. Do not implement code while using this skill.
.alloy/tasks/<id>/context.md exists.Write:
.alloy/tasks/<task-id>/plan.md
The file must start with:
---
id: <task-id>
approved: false
---
If the user approves in chat, update the frontmatter or add a progress handoff note before /execute.
# <task-id>: Plan
## Goal
## Non-Goals
## Acceptance Criteria
- AC-1:
- AC-2:
## Source Reality
- Files inspected:
- Existing conventions:
- Generated/runtime contracts:
- Tests or scripts available:
## File Scope
- Create:
- Modify:
- Delete:
- Avoid:
## Implementation Tasks
1. Task:
- Files:
- Test first:
- Expected gate update:
2. Task:
## Verification Commands
- Command:
- Purpose:
- Expected result:
## Progress Gates
```markdown
## Gate
- [ ] tdd_red
- [ ] green
- [ ] review
- [ ] verified
## Output Density
Do not write vague tasks like "update docs" or "fix tests". Include exact paths, exact commands, and what each command proves.
## Subagent Policy
Use one-shot Explorer only when the plan needs broad source discovery. The handoff must include the exact question, search scope, and expected output. Do not install or reference persistent subagent files.
## Final Chat Output
Return:
- task id
- plan path
- acceptance criteria count
- files in scope
- required verification commands
- unresolved risks
- next command: `Review the plan, set approved: true, then run /execute.`
## Upstream (fusionType `inline-append`)
This skill distills obra/superpowers v5.1.0 `brainstorming` + `writing-plans`. The verbatim upstream copies live under `vendor/skills/superpowers/5.1.0/` in the Alloy source repo; read them for the full brainstorming dialogue discipline and plan-writing depth when a plan is large or high-risk.