← Back to Skills Marketplace
PRD Iterative
by
Galaxy-Operations-Guide
· GitHub ↗
· v1.0.0
· MIT-0
123
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install prd-iterative
Description
多轮迭代PRD写作流程。用法:用户说"写PRD"、"帮我做一个产品的PRD"、"多轮迭代PRD"、"PRD rounds"时触发。输入:产品主题 + 背景素材 + 开发平台。流程:派发writer写稿 → reviewer审阅 → 反馈打回 → 循环N轮 → 最终交付。
Usage Guidance
这个技能看起来是按描述实现的,但在安装/使用前请注意:
- 在启动前确认并指定 Writer/Reviewer 模型名称及交付路径(技能已强调必须由用户确认)。
- 若提供的“素材”包含网页链接,技能会抓取这些链接的公开内容——不要传送包含敏感或需要认证的私密链接/凭据。
- 确认交付路径和文件写入权限,避免写入系统敏感目录;如不希望在本地写文件,可指定一个安全的输出目录或询问手动导出。
- 复核所选模型是否为您可接受的模型(避免无意调用付费或未授权模型)。
总体来看,这个技能内部一致且无明显过权限要求,但请按上面建议在首次运行时谨慎确认输入与输出路径。
Capability Analysis
Type: OpenClaw Skill
Name: prd-iterative
Version: 1.0.0
The skill defines a legitimate multi-round workflow for generating Product Requirements Documents (PRDs) using a writer-reviewer agent pattern. It includes safeguards such as mandatory user confirmation for output paths and model selection, and its instructions in SKILL.md are strictly focused on document iteration and file management within user-specified directories. No indicators of data exfiltration, malicious execution, or prompt injection attacks were found.
Capability Assessment
Purpose & Capability
技能名与描述是“多轮迭代 PRD 写作”,SKILL.md 仅描述如何收集信息、调用 writer/reviewer 模型、抓取链接素材(若有)并将版本文件写入用户指定路径;这些都是与写 PRD 直接相关的能力,没有请求与目的无关的凭据或外部服务。
Instruction Scope
指令明确要求在素材为链接时抓取内容,并把生成的 MD 文件写入用户指定路径;这超出了纯文本处理的范围但与功能相关。注意点:技能会进行网络抓取(如果用户提供链接),并会在主机上写入文件——用户应确认并授权目标路径,且避免提供包含敏感信息或需要凭证的私有链接。指令没有要求读取其它系统文件或环境变量。
Install Mechanism
这是一个 instruction-only 技能,没有安装步骤、没有下载或执行外部二进制,风险最低。
Credentials
技能不要求任何环境变量、凭据或配置路径。其默认输出路径仅作提示,并明确要求用户确认实际交付路径。未见不相称的秘密或外部凭据请求。
Persistence & Privilege
always:false,且技能没有指示修改其他技能或系统范围配置。技能允许由代理调用(默认行为),但没有过度持久化或权限升级行为。
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install prd-iterative - After installation, invoke the skill by name or use
/prd-iterative - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release - 多轮迭代PRD写作流程,3-5轮Writer/Reviewer循环交付完整PRD
Metadata
Frequently Asked Questions
What is PRD Iterative?
多轮迭代PRD写作流程。用法:用户说"写PRD"、"帮我做一个产品的PRD"、"多轮迭代PRD"、"PRD rounds"时触发。输入:产品主题 + 背景素材 + 开发平台。流程:派发writer写稿 → reviewer审阅 → 反馈打回 → 循环N轮 → 最终交付。 It is an AI Agent Skill for Claude Code / OpenClaw, with 123 downloads so far.
How do I install PRD Iterative?
Run "/install prd-iterative" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is PRD Iterative free?
Yes, PRD Iterative is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does PRD Iterative support?
PRD Iterative is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created PRD Iterative?
It is built and maintained by Galaxy-Operations-Guide (@galaxy-operations-guide); the current version is v1.0.0.
More Skills