PRD Generation from User Story

Converts docs/UserStory.md into docs/PRD.md with structured functional requirements.

Purpose

Bridges the gap between user-focused stories and implementation-ready requirements by converting narrative descriptions into technical requirements.

Workflow

1. Read inputs

- Read ralph/docs/templates/prd.md.template - follow its structure exactly - Read docs/UserStory.md - parse all sections: Problem, Users, Value, Stories, Criteria, Constraints

1. Extract and convert

- User stories → Functional requirements (WHAT the system must do) - Success criteria + Constraints → Non-functional requirements (HOW it performs) - Group by functional area (CLI, API, Storage, etc.)

1. Generate PRD.md

- Follow template structure and PARSER REQUIREMENT comments - Project Overview: Combine problem statement + value proposition - Functional Requirements: Organized by feature area - Non-Functional Requirements: Performance, usability, platform - Story Breakdown: Include (depends: STORY-XXX) when story B requires story A to complete first - Out of Scope: Copy from UserStory.md

1. Validate against template

- Verify output matches ralph/docs/templates/prd.md.template structure - Check PARSER REQUIREMENT elements are present (headings, checkboxes, story format)

1. Save

- Backup existing PRD.md as PRD.md.bak if overwriting - Write to docs/PRD.md - Suggest: make ralphprdjson

Conversion Examples

User Story → Functional Requirement:

• "As a developer, I want to add tasks from CLI" → "CLI command: task add "description""

Success Criteria → Non-Functional:

• "Operations in < 5 seconds" → "Performance: All CLI operations complete in < 5 seconds"

Constraints → Non-Functional:

• "Must work on Linux, macOS, Windows" → "Platform Support: Cross-platform compatibility"

Usage

make ralph_prd_md

Next Steps

make ralph_prd_json   # Generate prd.json from PRD.md
make ralph_run        # Start Ralph loop