Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add chaunsin/agent-skills --skill prd-to-specgit clone --depth 1 https://github.com/chaunsin/agent-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/chaunsin/agent-skills/prd-to-spec)<a href="https://agentmods.dev/skills/chaunsin/agent-skills/prd-to-spec"><img src="https://agentmods.dev/badge/skills/chaunsin/agent-skills/prd-to-spec/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/chaunsin/agent-skills/prd-to-spec"><img src="https://agentmods.dev/badge/skills/chaunsin/agent-skills/prd-to-spec.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00077 | $0.02835 |
| Opus 5.5 | $0.00031 | $0.01134 |
| Sonnet 5.5 | $0.00015 | $0.00567 |
| Haiku 4.5 | $0.00008 | $0.00283 |
Grade A, and why
prd-to-spec scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 23d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 192 lines — stays where its author put it; the contents beside it link to each section on GitHub.
产品设计文档到技术规范的转换
以50年代码开发工作经验及架构师身份从代码工程范式、可维护性、性能、质量、安全等角度出发,生成具体代码详细技术规范。根据(PRD)文档内容,补全代码领域潜在行业知识、代码风格、技术规范,并澄清潜在的问题,来指导工程师或人工智能如何构建。
何时使用
- 根据
/prd生成得文档,弥合它与实施落地之间的差距。 - 当未开始实时编码,方案调整阶段,prd文档更新对应spec同步更新时。
- 你希望在编码开始之前就把架构决策记录下来。
- 多个开发人员/智能体将实现此功能,因此需要一份共享合约。
- 在落地实施之前,生成产品需求文档 (PRD) 之后,您需要验证技术可行性以及问题。
步骤1:找到PRD
可以通过以下方式之一找到输入 PRD:
Provide the PRD to convert:
A. File path (e.g., docs/prd/prd-priority-system.md)
B. Accessible web service URL
C. Paste PRD content directly
D. Auto-detect: scan docs/ directory for recent PRDs
如果自动检测,则列出可用的 PRD 并让用户选择:
**Found PRDs in docs/prd/:**
1. docs/prd/prd-priority-system.md (2024-03-15)
2. docs/prd/prd-user-auth.md (2024-03-10)
Which PRD should I convert? [1/2]
如果没有找到PRD则引导用户执行/prd 技能并生成PRD文档,然后在执行后续操作。
步骤2:信息收集(可选)
结合项目工程上下文收集信息,补充实际落地代码之间的隐藏关键差距信息
- 现有架构——当前系统的结构方式
- 技术栈——已使用的语言、框架、依赖库、代码管理工具。
- 模式——文件组织、公共包、通用方法、组件、帮助函数、错误处理方法
- 代码规范——命名约定、代码风格、规范、质量标准、设计模式
- 数据库——当前模式、迁移工具、ORM
- API 风格— REST/GraphQL/gRPC、身份验证方法、响应格式
- 测试——测试框架、覆盖模式、测试工具
收集方式如下:
- 用户提供得prompt、文档、图片、链接、可访问资源等。
- 检索架构设计文档
docs/architecture/ADR-[number]-[title].md文档,存在则参考设计。 - 当前已有的项目实现,优先读取CLAUDE.md、AGENTS.md、README.md等文档内容,或分析现有代码仓库实现。
- 知识库、.docs/、.spec/、plan/ 、prd/检索
- 检索skills以及rules
.claude/skills/**/SKILL.md,.codex/skills/**/SKILL.md,.**/skills/**/SKILL.md、.claude/rules/*、.codes/rules/*等可用得技能、规则, 提取最佳实践、约定规范、案例等。 - 外部知识检索,当收集到的信息不足、了解最新技术方案与技术调研,如下:
- Context7 MCP(
mcp__context7__resolve-library-id,mcp__context7__query-docs):当 MCP 服务器连接时首选,返回结构化文档MCP检索。当MCP不可用但ctx7 CLI 可用则作为备用方案 - WebSearch/WebFetch。
- 其他可访问知识检索MCP、skills、知识库检索等。
- Context7 MCP(
对收集得信息进行分析并归纳总结,用于后面澄清与决策。
步骤3:澄清问题
当产品需求内容存在潜在问题、矛盾、缺失,或对技术决策存在歧义时才提出问题,一次只提一个问题。
- 需求模糊,交互流程缺失、逻辑矛盾,边界不清晰。
- 数据流转生命周期是否完整,是否闭环,是否矛盾。
- 错误处理、限流控值、服务降级、安全保证,兜底方案。
- 需求不合理处,技术瓶颈,实现风险。
- 复用现有实现,如不能复用,在原有的接口、页面、组件、方法、对象,以及功能模块紧邻相关方法上进行修改或扩展。
- 澄清内容矛盾点,例如一个需求功能提供了多份资料来源:
- 文档与文档之间描述不一致情况问题。
- 文档与截图,UI图,原型图所呈现不一致内容矛盾点(图中展示5个字段文档中为6个字段)。
- 现有代码与文档之间的矛盾点。
- 补充prd文档、需求中没有提到的隐形知识与规则,根据 信息收集 丰富上下文。例:
- 平台支持了国际化,默认人是知道的但是文档没有体现。
- 增加了新的支付方式,文档并没有提出账户流水处理方式。
- 默认样式颜色、布局、弹窗、组件、间距、字体、内容过长缩略展示策略等。
- 接口参数的规则规范,字段命令都是采用驼峰还是下滑线。
- 具体领域知识的默认、约定俗成规则,比如文件命名不能采用中文等。
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 23d ago Changed · +1 lines 2a7d202f0b9c
- 1mo ago First seen · 191 lines · 77 tokens per session scan A 3172054f3811
prd-to-spec is a skill published in the GitHub repository chaunsin/agent-skills (2 stars, last pushed 24d ago), licensed Apache-2.0. It adds 77 tokens to every session and 2,835 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-11.
Other skills, from other repositories
new
Create a new project to start development quickly.
channel-manager
Configure IM platform channels (Feishu, WeCom, Weixin, Discord, Telegram, DingTalk, QQ) for openclacky. Uses browser automation for navigation; guides the user to paste credentials and perform UI steps. Trigger on: "channel setup", "setup feishu", "setup wecom", "setup weixin", "setup wechat", "setup discord", "setup…
yida-data-management
A tool for querying, adding, and updating Yida form records, child tables, process records, and task-center data. It keeps ordinary forms and approval workflows on their separate data interfaces.
yida-canvas-data-binding
A method for connecting a YidaCodeCanvas page to real data from Yida forms, connectors, or same-site APIs. YidaCodeCanvas is the component used to run custom React pages inside Yida.
yida-connector-safe-actions
A method for turning existing API definitions into callable actions in an existing Yida HTTP connector. It reads frontend API wrappers and backend controller definitions to determine the request details.
yida-get-schema
A command for reading the field IDs, subtable paths, and behavior settings of a Yida form. Yida is a platform for building online forms and business workflows.