How to Use the /write-prd Command to Create a Product Requirements Document in phuryn/pm-skills

The /write-prd command automates Product Requirements Document creation through a four-step conversational workflow that transforms feature ideas into structured markdown files.

The /write-prd command provides a structured workflow for generating comprehensive Product Requirements Documents (PRDs) from simple feature descriptions or detailed problem statements. Defined in pm-execution/commands/write-prd.md, this command orchestrates the create-prd skill located at pm-execution/skills/create-prd/SKILL.md to produce standardized documentation. Understanding how to use the /write-prd command enables product managers to move from concept to structured PRD through an interactive, guided process.

Invoking the /write-prd Command

You can trigger the workflow using natural language inputs ranging from brief feature names to detailed problem descriptions. The command accepts plain text entries or uploaded documents such as PDFs, research briefs, or strategy decks.

/write-prd SSO support for enterprise customers

/write-prd Users are dropping off during onboarding — we need to fix step 3

The Four-Step Workflow

According to the source code in pm-execution/commands/write-prd.md, the command executes a structured pipeline that guides you from initial idea to finalized document.

Step 1: Understand the Feature

The system parses any input format—whether a feature name, problem statement, user request, or uploaded file—to establish the foundation for the PRD. This initial parsing occurs in lines 20‑27 of the command definition.

Step 2: Gather Context

The command initiates targeted conversational prompts to capture critical missing information. It queries for:

• The core user problem
• Target users and personas
• Success metrics
• Constraints and limitations
• Prior art and competitive analysis
• Scope preferences

This contextual gathering is implemented in lines 31‑39 of pm-execution/commands/write-prd.md.

Step 3: Generate the PRD

The workflow invokes the create-prd skill to populate an eight-section markdown template. The resulting document includes:

1. Executive Summary
2. Background
3. Objectives
4. Target Users
5. User Stories
6. Solution Overview
7. Open Questions
8. Timeline

The template structure and generation logic reside in pm-execution/skills/create-prd/SKILL.md.

Step 4: Review and Iterate

After generating the draft, the system presents options to refine scope, run a pre-mortem, split the work into user stories, or create stakeholder updates. Finally, it saves the document as PRD‑<feature>.md in your workspace (lines 101‑108).

Practical Input Examples

The command handles various input styles depending on your documentation maturity:

Simple feature requests:

/write-prd Implement dark mode for dashboard

Problem-oriented statements:

/write-prd Checkout abandonment spikes at the payment step

Document-based inputs: Attach research documents or strategy decks directly after the command text to seed the context.

Source Code Structure

Understanding the /write-prd command requires familiarity with its underlying architecture:

• pm-execution/commands/write-prd.md: Contains the complete workflow definition, invocation examples, and step-by-step logic.
• pm-execution/skills/create-prd/SKILL.md: Defines the eight-section PRD template and generation instructions used by the command.
• pm-execution/README.md: Provides overview context for how this command integrates with the broader execution suite.

Summary

• The /write-prd command resides in pm-execution/commands/write-prd.md and automates PRD generation through a conversational interface.
• It follows a four-step workflow: understand the feature, gather context, generate the PRD, and review iteratively.
• The command produces an eight-section markdown document covering Executive Summary through Timeline according to the template in pm-execution/skills/create-prd/SKILL.md.
• Input flexibility allows plain text descriptions or uploaded documents, with the system adapting to extract necessary context.
• The generated file is saved as PRD‑<feature>.md after the review and iteration phase.

Frequently Asked Questions

What file format does the /write-prd command generate?

The command generates a markdown file named PRD‑<feature>.md containing eight standardized sections including Executive Summary, Background, Objectives, Target Users, User Stories, Solution Overview, Open Questions, and Timeline. The template structure is defined in pm-execution/skills/create-prd/SKILL.md.

Can I use the /write-prd command with existing research documents?

Yes. The command accepts uploaded briefs, PDFs, research docs, or strategy decks in addition to plain text descriptions. During the "Understand the Feature" step (lines 20‑27 of pm-execution/commands/write-prd.md), it parses these documents to extract the foundational context needed for the PRD.

What happens after the PRD is generated?

After generation, the workflow enters a review phase where you can refine scope, run a pre-mortem analysis, break requirements into user stories, or create stakeholder updates. This iterative process occurs in lines 101‑108 of the command definition before finalizing the markdown file.

Do I need to answer all the context-gathering questions?

While the command prompts for specific details—including target users, success metrics, constraints, and prior art—you provide only the information you have available. The conversational interface adapts to fill gaps during the "Gather Context" step (lines 31‑39).

{
  "mcpServers": {
    "instagit": {
      "command": "npx",
      "args": ["-y", "instagit@latest"]
    }
  }
}