acceptance-criteria
Skill1 starsChanged 7 months agoMITView on GitHub
Knowledge pack for writing Given/When/Then acceptance criteria that are unambiguous and automatable. Auto-loaded by product-agent (writing) and test-agent (consuming).
16 other projects use a version of this.
Adopted by
Most recently taken up by
- celestialdust/achilles-skillstheir own version, first seen Nov 1, 2026
- MiguelSavignano/claude-code-systemtheir own version, first seen Nov 1, 2026
- Happy-Technologies-LLC/happy-platform-skillstheir own version, first seen Nov 1, 2026
- Tngc93/product-management-skillstheir own version, first seen Oct 20, 2026
- Sunnyshinnny776/pm-agentic-cookbooktheir own version, first seen Oct 20, 2026
- prabhakar349/claude-ddd-redesigntheir own version, first seen Oct 9, 2026
Other versions16 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 16
- Happy-Technologies-LLC/happy-platform-skills★ 41same name
- jdm4pku/RE-Skills★ 26same name
- MiguelSavignano/claude-code-system★ 6same name
- HamzaSheikh768/Agents-Skills-Lap★ 3same name
- designedbythanh/ai-skills-library-product-design★ 2same name
- garthcodes/claude-skills★ 2same name
- smarzban/agent-sdlc★ 2same name
- Sunnyshinnny776/pm-agentic-cookbook★ 2same name
- celestialdust/achilles-skills★ 1same name
- KonstantinVCH/QA-agent-skills★ 1same name
6 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository12 other files it gives its agents.
- ai-native-designSkill
- api-contract-firstSkill
- backend-developmentSkill
- checkpoint-commitSkill
- ddt-baseline-syncSkill
- ddt-brief-builderSkill
- decision-gateSkill
- delivery-packageSkill
- efficiency-metricsSkill
- frontend-developmentSkill
- relaySkill
- validation-loopSkill
The file
Files
Acceptance Criteria
Triggers
- product-agent 启动 / /prd 命令
- test-agent 启动 / /test 命令
Core Principles
- 每个用户故事至少 1 条 happy-path + 1 条 edge-case
- 可自动化测试 是硬约束——如果测试工程师无法在不问产品的情况下把它变成代码,它就不合格
- Non-Goals 和验收标准等重要——明确边界防止实现漂移
Format
Given <前置状态 / 上下文>
When <用户动作 或 系统事件>
Then <可观测结果,含可量化断言>示例(登录场景)
# Happy-path
Given 用户已注册(email: alice@example.com,password: 已设置)
When 用户提交正确的 email + password
Then HTTP 200,body.token 存在(JWT 格式),有效期 1 小时
# Edge-case:密码错误
Given 用户已注册
When 用户提交正确 email + 错误 password,连续 5 次
Then 第 5 次返回 HTTP 429,body.error.code = "too_many_attempts",锁定 15 分钟
# Edge-case:账号不存在
Given 系统中无此 email 的用户
When 用户提交该 email
Then HTTP 404,body.error.code = "user_not_found"Do
- 使用可观测的结果(HTTP 状态码 + 响应字段断言、页面上出现的精确文本)
- 数值要具体(响应时间 < 500ms,而不是"响应快")
- 错误路径要给出确切错误码或错误文案(而不是"报错")
- 边界值全覆盖:0 / 1 / 最大值 / 负数 / 空字符串 / null / 超长输入
Don't
- 写"系统应该正确处理 X"("正确"无客观判定)
- 写"用户体验要好"(不可测)
- 写"根据需要返回相关数据"(糊弄词,定义不清)
- 写"性能要足够快"(没有数值的性能要求无效)
可测试性判定表
| 验收标准特征 | 可测 | 不可测 → 怎么修 |
| 含具体 HTTP 状态码 | ✅ | 补充状态码 |
| 含 body 字段断言 | ✅ | 指定 body.field == "value" |
| 含时间数值(< Xms) | ✅ | 写出具体毫秒数 |
| "成功" / "失败" | ❌ | 改为 HTTP 2xx / 4xx + body |
| "响应快" / "体验好" | ❌ | 改为 p95 < 500ms |
| "根据需要" | ❌ | 明确条件与结果 |
Self-Test(交给 test-agent 前自问)
- 我能只看这条验收标准写出 test 代码吗?
- "通过 / 不通过" 的判定是客观的吗?
- 边界值(0 / 1 / 最大值 / 负数 / 空 / null)都覆盖了吗?
- 每个用户故事至少有 1 条 happy-path + 1 条 edge-case 吗?
- 非目标(Non-Goals)≥ 3 条吗?
Templates & References
templates/prd.template.md(内含验收标准填写示例)skills/api-contract-first/SKILL.md(错误码与验收标准对应关系)
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.

