feature-prepare
Skill2 starsChanged 5 months agoMITView on GitHub
将用户的功能想法整理成轻量 feature 批次,生成业务项目 features/<版本>/feature_list.json、specs/FEAT-xxx.md 和 tasks/FEAT-xxx-TASK.md 草案。适用于用户要开发新页面、新功能、功能优化,想先明确目标、主要行为和验收标准,再交给 feature-worktree 开发。
In this repository12 other files it gives its agents.
- bug2csvSkill
- bugbatch-prepareSkill
- bugfix-comment-contextSkill
- bugfix-doneSkill
- bugfix-worktreeSkill
- debug-tag-addSkill
- debug-tag-clearSkill
- feature-doneSkill
- feature-worktreeSkill
- plan-csv-executeSkill
- planSkill
- plan-to-issues-csvSkill
The file
Files
Feature Prepare
目的
把自然语言需求整理成可执行 feature 规格。初版保持轻量,不做复杂 rubric、依赖图或证据目录。
输出放在要开发的业务项目根目录:
features/<版本>/
feature_list.json
specs/
FEAT-001.md
tasks/
FEAT-001-TASK.md输入
需要:
- 功能需求描述
- 业务项目路径
- 版本名;如果用户没给,用当前日期或一个短 slug
feature_list.json
初版字段:
{
"features": [
{
"feature_id": "FEAT-001",
"title": "笔记详情页",
"status": "ready",
"spec_file": "specs/FEAT-001.md",
"task_file": "tasks/FEAT-001-TASK.md",
"branch": "",
"worktree_path": "",
"commit": "",
"ai_verify_result": "",
"block_reason": ""
}
]
}状态:
draft, ready, needs_info, in_progress, ai_verified, apk_built, human_verified, merged, integration_verified, worktree_removed, blocked
spec.md
每个 feature 的 spec 初版只要求三块:
# FEAT-001 <标题>
## 目标
<这个功能要解决什么>
## 主要行为
- <入口/操作/跳转/状态变化>
## 验收标准
- <什么结果算完成>如果需求是页面开发,主要行为里可简要写页面入口、主要区块和关键交互。不要把所有视觉细节都写死,除非用户明确要求。
TASK.md
生成空执行日志:
# FEAT-001 TASK
## 当前状态
ready
## 修改记录
## AI 自证验证
- 验证计划:
- 执行命令:
- 结果:
- 证据:
- 不能验证的部分:
## 用户验证
## 阻塞/风险
## 下一步
- 使用 `$feature-worktree FEAT-001` 开始开发。人工审核点
生成后提醒用户主要审核 spec.md:
- 目标是否正确
- 主要行为是否漏掉关键入口/交互
- 验收标准是否认可
用户不需要逐个审核 TASK,TASK 是后续执行日志。
完成回复
报告:
feature_list.json路径- spec 路径
- task 路径
- 哪些 feature 是
ready - 哪些 feature 需要补信息
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

