PRD Writing Skill
The prd-write skill for Codex: turns an idea, prototype and screenshots into a PRD with roles, flows, events and acceptance
- 154 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- The skill writes to Feishu documents when you name a target document
- The skill generates images and HTML files in the project
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/prd-writing-skillDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- The skill writes to Feishu documents when you name a target document
- The skill generates images and HTML files in the project
This is third-party code. Review the repository files before installing.
What it does
prd-write drives PRD writing through fixed steps. First the skill inventories the source materials and marks the single document that may be modified, while other Feishu links stay read-only references. It then clarifies requirements (through Superpowers brainstorming or a local pass) and writes a short brief: goal, users, scope, main path, exceptions, assumptions and open questions. Next come a page and state inventory, a first visual prototype through imagegen, an interactive HTML demo and screenshots of its screens, which can optionally be redrawn for a cleaner look without inventing requirements. Before writing, the skill prepares a flowchart, a roles and permissions matrix, a status machine, a field dictionary, an analytics plan and acceptance checks. The final PRD covers background and goals, scope, roles, process, requirement descriptions, status rules, data and interfaces, analytics events, constraints and an acceptance checklist. Before writing back to Feishu it self-checks source coverage, consistency of flow, prototype and text, permissions and testable criteria.
Who it is for. Product managers and founders turning an idea, draft or app prototype into a working PRD with an agent.
Good fit when
- A raw app idea has to become a structured PRD
- An existing PRD draft is being rewritten into one structure
- An HTML prototype should become requirements with roles, statuses and events
- A PRD is prepared for Feishu with flowcharts and screenshots
Not a fit when
- You need a short one-pager: a seven-step process is excessive
- A product with no screens, such as a library or an API, has no use for screen steps
- The agent cannot generate images or open HTML: prototype steps are skipped or done by hand
Example request
Use prd-write to build a PRD for a doctor appointment app from this prototype and three screenshotsLimitations
The skill targets the Feishu ecosystem and optionally Superpowers. Prototype steps need imagegen and the ability to build an HTML demo. Screenshot and page-description rules live in references/, and evaluation scenarios live in the evals folder. Language is mixed: the README is English and some rules use Chinese terms. The seven steps with a prototype and screenshots noticeably lengthen PRD work, and the result depends on the agent's ability to generate images.
How to disable. Delete the prd-write skill folder from the agent skills directory. Created PRDs, prototypes and screenshots stay in the project.
FAQ
Which steps does the skill follow?
Source inventory, requirement clarification, page and state inventory, visual prototype, interactive HTML demo, screenshots, business artifacts and PRD writing with a self-check.
Why does the skill insist on one target document?
It is a safety rule: only one explicitly named document is edited and other Feishu links are read-only. When unsure, the skill stops and asks before writing back.
Is Superpowers required?
No, but with it installed the skill asks the agent to use using-superpowers and brainstorming first. Without it the clarification is a local pass.
Does the skill work in Claude Code?
The README talks about Codex and OpenSkills. SKILL.md has name and description fields, a format other agents read too, but the author describes no checks in Claude Code.
What does the self-check cover before writing?
Source coverage, consistency of flow, prototype and text, roles and permissions, status rules, analytics and testable acceptance criteria. The write to Feishu comes after that.
Discussion
No questions or reviews yet.
Related
SkillsA Claude Code method: a gated PRD, 47 skills, hooks and a Markdown fact base so the agent remembers decisions between sessions.
A Claude Code plugin that turns a goal into a graded PRD, a dependency-ordered task graph and execution gated by evidence
Idea stress test, PRD with a roadmap, design tokens, a launch plan and a phase-by-phase build in one skill for founders.
A set of 58 commands for product managers in Claude Code, Codex and Cursor: from an idea and a PRD to launch and growth analysis
A spec-driven delivery skill: PRDs, epics, tasks, GitHub Issues sync and parallel agents in git worktrees
A CLAUDE.md and AGENTS.md file with seven principles: the agent frames the problem and scope before it writes code
Comparison with similar tools
Cards that share the main search query, and how they differ.
| Tool | Official | Repository stars | Main difference |
|---|---|---|---|
| PRD Writing Skill (this card) | no | 154 | |
| create-prd | no | 376 | a short PRD-creation skill without prototype or Feishu steps |
| AI PRD Workflow | no | 310 | a general PRD workflow with an agent, without prototype work |
| prd-taskmaster | no | 610 | breaks a PRD into tasks for the agent |
README in short
The README lists seven steps from source inventory to PRD writing, says for each what the skill does and produces, and adds screenshot layout rules for apps and mini-programs. A PRD template, page description rules and Feishu whiteboard flowchart guidance sit alongside.
README badge
Original SKILL.md
--- name: prd-write description: Structured product requirement writing workflow for turning ideas, Feishu docs, PRD drafts, HTML prototypes, app/mini-program/PC demos, or screenshots into a complete PRD. Use when the user asks to write, update, rewrite, review-then-write, or directly publish a PRD/需求文档/产品方案, especially when the work should first inventory source materials, clarify requirements with brainstorming or Superpowers, create a requirement brief, design page/state inventory, build or inspect a prototype demo, then write a PRD with background, roles, permissions, business flowchart, requirement description, PV/UV analytics events, acceptance checklist, and Feishu-ready screenshots. --- # PRD Write ## Core Rule Run the workflow in this order: 1. Inventory source materials and mark the single write target; treat all other Feishu links as read-only references. 2. Clarify requirements with Superpowers `using-superpowers` + `brainstorming`, or a local brainstorming pass if Superpowers is unavailable. 3. Produce a requirement brief and get alignment before design-heavy work. 4. Create a page/state inventory before image generation so visual work does not invent workflow coverage. 5. Generate a first-pass visual prototype with imagegen from the clarified product description. 6. Build, read, or refine an interactive HTML demo based on that visual prototype to validate real interactions. 7. Verify the demo flow, capture key screens, and optionally redraw them with imagegen for a cleaner PRD presentation. 8. Prepare business artifacts: flowchart, status machine, roles/permissions, fields/data dependencies, analytics, and acceptance checks. 9. Write the PRD with the user's PRD template, self-check it, then update only the explicitly targeted document. Do not jump straight into PRD writing when the product path is still blurry. ## Writing Style: Direct And Concise Keep the original PRD structure and requirement coverage, but write only information that changes product understanding, implementation, testing, or decision-making. - Start sections with the business fact or requirement. Do not narrate how the document is organized, how readers should review it, or why a screenshot is placed there. - Omit meta prose such as “本 PRD 以 HTML 原型为依据”“文档先放完整页面总览”“便于评审确认页面结构” when the attachment, screenshot, caption, and following requirement already make that relationship obvious. - Do not announce the next section, summarize the previous section, or explain that a table/screenshot “用于说明” content that is already visible. - Prefer one direct sentence over a setup paragraph. Example: use “PC 端交互以随附 HTML 原型为准。” only when the source-of-truth relationship must be explicit; otherwise omit it. - Avoid stiff template language such as “对应需求如下”“用户可通过该功能实现”“系统将会进行”. Use direct product language: “支持…”, “点击后…”, “校验失败时…”. - After drafting, delete any sentence whose removal does not change scope, rule, interaction, data, exception, or acceptance meaning. ## Workflow ### 0. Source Inventory And Target Safety Before brainstorming or editing, list the available inputs and classify each item. - Target document: the one document/wiki/page that may be modified. - Read-only references: Feishu docs, PRD examples, flowcharts, meeting notes, screenshots, HTML demos, data tables, API docs, or prior drafts. - Prototype assets: current HTML files, screenshots, imagegen drafts, redrawn images, and generated artifacts. - External systems: CRM, payment, inventory, approval, messaging, data warehouse, or manual operations. - Constraints: deadline, platform, brand/design system, compliance, money/data risk, and permission boundaries. If there is any ambiguity about which Feishu document may be edited, stop and ask before writeback. ### 1. Requirement Brainstorming