agentleFS
youshuang123youshuang123/product-factory/skills/agent-blueprint/SKILL.md

agent-blueprint

Skill0 starsChanged 9 days agoMITView on GitHub

将 PRD、Agent 点子或现有系统需求转成架构蓝图:判断是否需要 Agent,映射工具/知识/观察/行动/权限五要素,选择最小运行机制,定义工具契约、权限、预算与验收证据,必要时画七层架构图。用于开发前的架构推导或评审,产出 factory/blueprint.md;不自动生成代码、调用外部工具或部署。

1 other project uses a version of this.

Adopted by

Most recently taken up by

  • Other versions1 project uses a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 1

    1. Starboy1968/claude-skills★ 0same name

    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 repository6 other files it gives its agents.

The file

Files
SKILL.md

Agent 架构蓝图

从业务需要推导架构,不从组件清单反推需求。

单独使用(不走工厂流程)时:跳过 factory/ 路径与 factory_gate.py,把同样的内容写进项目已有的文档位置;需要用户确认或授权的地方,仍要在对话中明确取得。

Agent = 模型(决策) + Harness(你构建的代码)
Harness = 工具 + 知识 + 观察 + 行动接口 + 权限

Harness 不是必须采用的某个框架,而是围绕模型的这五类能力与控制。

输入与读取

  • 工厂流程中:输入是已签字的 factory/prd.md,产出写到 factory/blueprint.md。PRD 缺失或未签字时,建议先用 prd-author;用户坚持只给一句话时,读 需求澄清 后做初版,并把假设标清楚。
  • 已有代码:先读相关需求、现有架构与接口,再读 组件决策,只处理本次涉及的能力。
  • 涉及工具、权限、记忆、并行或复用示例代码:读 运行边界。
  • 需要分层图:读 七层架构。
  • 写方案文档:用 蓝图模板,按任务删减;小任务不强制产出大文档。

推导步骤

  1. 确认用户要的是澄清、方案、评审还是已授权实现。蓝图任务到方案交付为止;“最终要上线”不等于本次有部署授权。
  2. 提取硬事实:目标用户、一次输入到结果的闭环、数据/系统来源、红线和成功标准,引用 PRD 的 AC 编号。PRD 中的矛盾先指出,例如禁止数据出域却指定外部模型。
  3. 比较普通程序、固定工作流、动态 Agent 与混合方案。确定性步骤交给代码;只有运行时确实需要推理选择工具时,才增加 Agent 自主性。
  4. 把需求映射到五要素:需要哪些工具、知识从哪来、如何观察结果、行动用什么接口、谁有权批准。每个外部动作都要有观察证据和权限边界。
  5. 选最少必要组件。每个组件写需求依据、缺少它的后果、维护成本和更简单的替代。默认只有循环 + 工具;不因“完整”而加团队、MCP、记忆或调度。
  6. 给出模块/数据流、工具契约、模型与上下文策略、失败降级、成本/时间/步数上限和验收方法。规模较大时用七层视角检查遗漏;小方案可合并层级或标“不适用”。
  7. 列出需用户确认的关键选择和下一阶段输入(交给技术适配)。用户已批准方案并要求实现时,不重复索要同一批准,只处理新增的实质分歧。已有项目做增量设计,不为蓝图推翻技术栈。

方案是设计,不是已运行产品;测试计划不是测试结果。未运行、未测量、未确认的内容要明确标注。不要因为生成了骨架或满足了几条清单就称“生产级”。

交付与签字

工厂流程中,写完 factory/blueprint.md(删除模板标记)后,向用户给出审批摘要:Agent/工作流结论、选中与放弃的组件、新增的外部依赖与费用、权限与预算边界、需要确认的问题;请用户签字(blueprint_signoff)。

功能范围

本 Skill 负责五要素建模与需求驱动选型,不包含代码生成器,也不依赖仓库外文件。见 功能范围 与 许可证。

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.