跳到正文
cccskills
免费在 GitHub 公开

prd-workflow

This skill should be used when the user asks to "plan a feature", "create PRD", "interview me about requirements", "define stories", "write a spec", "break down feature into tasks", "create implementation plan", "help me scope this feature", "what questions should I answer", or discusses feature planning. Covers the PRD interview process, spec structure, story atomization, and 4-phase workflow from idea to Dex tasks.

中文简介正在准备中。当前显示原文说明。

查看安装方法

包含的文件(1)

  • SKILL.md4.8 KB

SKILL.md(原文)

安装前可以查看交给 AI 代理的指令内容。

PRD Workflow - Feature Planning

Current branch: !git branch --show-current 2>/dev/null || echo "not in git repo" Existing plans: !ls plans/ 2>/dev/null | head -5 || echo "none"

The PRD (Product Requirements Document) workflow transforms ideas into atomic, implementable stories through structured interview, research, and specification.

When to Use PRD Workflow

  • Starting a new feature from scratch
  • Turning a vague idea into concrete tasks
  • Need thorough requirements gathering
  • Want AI-assisted research and review

The 4-Phase Approach

PhaseNamePurpose
1Input ClassificationIdentify feature name and type
2Interview + ExplorationGather requirements, then run research/expert agents
3Spec WriteCreate specification with Implementation Stories section
4Dex HandoffParse stories into Dex tasks

Key Change: Research and expert agents run after interview completes (blocking). Findings are reviewed before spec writing.

Starting the Workflow

/prd "Add user authentication"           # From idea
/prd path/to/existing/code               # From codebase
/prd                                     # Interactive discovery

Phase 2: Interview + Exploration

Commitment: "I will ask 8-10+ questions covering: core problem, success criteria, MVP scope, technical constraints, UX flows, edge cases, error states, and tradeoffs."

Interview (8-10+ questions)

Cover these areas:

  • Core problem, success criteria, MVP scope
  • Technical systems, data models, existing patterns
  • UX/UI flows, error states, edge cases
  • Tradeoffs, compromises, priorities

Focus on non-obvious details that will bite during implementation.

After Interview, Run Agents (blocking)

Spawn ALL agents in parallel (user waits ~2-3 min):

Research Agents:

  • prd-codebase-researcher - Existing patterns, files to modify
  • git-history-analyzer - Prior attempts, why patterns evolved
  • prd-external-researcher - Best practices, code examples

Expert Agents:

  • architecture-strategist - System design concerns
  • security-sentinel - Auth, data exposure risks
  • spec-flow-analyzer - User journeys, missing flows
  • pattern-recognition-specialist - Codebase consistency

Review Findings

Summarize key findings. Ask clarifying questions if gaps revealed.

Phase 3: Spec Structure

Write to plans/<feature>/spec.md with this structure:

# <Feature> Specification

## Overview
## Problem Statement
## Success Criteria
## User Stories
## Detailed Requirements
## Technical Design
## Edge Cases & Error Handling
## Open Questions
## Out of Scope

## Implementation Stories

### Story 1: <Title>
**Category:** functional|ui|integration|edge-case|performance
**Skills:** <skill-name>, <skill-name>
**Blocked by:** none
**Acceptance Criteria:**
- [ ] Criterion 1
- [ ] Criterion 2

## Research Findings
## Expert Review Findings

Story Atomization Rules

Commitment: "Each story will have ≤7 acceptance criteria, touch ≤3 files, be independently testable, and have no 'and' in title."

Each story must be completable in ONE task (~15-30 min):

  • Max 7 acceptance criteria - If more, split the story
  • Max 3 files touched - If more, consider splitting
  • No "and" in title - "User can X and Y" = two stories
  • Independently testable - Can verify in isolation
  • Cleanly revertible - Can undo without cascade

Phase 4: Dex Handoff

Use dex plan to automatically create tasks from the spec:

dex plan plans/<feature>/spec.md

This automatically:

  • Creates parent task from spec title
  • Analyzes Implementation Stories section
  • Generates subtasks with proper hierarchy
  • Sets blocked-by relationships from "Blocked by:" lines

Verify tasks created:

dex status
dex list

After PRD Completion

Use Dex for execution:

dex status              # Dashboard view
dex list --ready        # See unblocked tasks
dex start <id>          # Claim a task
/complete <id>          # Run reviewers and complete

Command Reference

/prd "feature description"     # Start from idea
/prd path/to/code             # Start from codebase
/cancel-prd                   # Cancel in-progress PRD

Related Commands

  • /complete - Complete task with reviewer workflow
  • dex list - View pending tasks
  • dex sync - Sync with GitHub issues

评价

还没有评价。欢迎分享你的使用感受。

评论

无需登录即可发表。欢迎分享使用感受、提问,或说明在哪些环境下无法运行。

还没有评论。来写第一条吧。

提交后会立即公开。请勿填写邮箱、电话等个人信息。URL 最多 2 个。

0 / 1000 字

此表单受 reCAPTCHA 保护,适用 Google 的隐私权政策和服务条款。

同一仓库的技能

This skill should be used when the user asks about "parallel agents", "background tasks", "run_in_background", "non-blocking agents", "check agent progress", "TaskOutput", "retrieve agent results", or discusses running multiple agents concurrently. Covers patterns for launching agents in background, monitoring progress, and retrieving results.

中文简介正在准备中。当前显示原文说明。

SomtoUgeh/somto-dev-toolkit22026年8月26日 更新

Creates and manages Biome GritQL custom lint rules to enforce coding patterns. Use when creating linter rules, enforcing code conventions, preventing anti-patterns, or when the user mentions Biome, GritQL, custom lint rules, or AST-based linting.

中文简介正在准备中。当前显示原文说明。

SomtoUgeh/somto-dev-toolkit22026年8月26日 更新

Transform brain dumps into polished blog posts in Nick Nisi's voice. Use when the user wants to write a blog post with scattered ideas, talking points, and conclusions that need organization into a cohesive narrative with Nick's conversational, authentic, and thoughtful tone.

中文简介正在准备中。当前显示原文说明。

SomtoUgeh/somto-dev-toolkit22026年8月26日 更新

This skill should be used when implementing features from a PRD spec, tracking implementation progress, or resuming work across sessions. Covers Dex task management for PRD story execution, daily workflows, and skill auto-loading.

中文简介正在准备中。当前显示原文说明。

SomtoUgeh/somto-dev-toolkit22026年8月26日 更新

This skill should be used when the user asks for "browser tests", "playwright tests", "end-to-end testing", "test user flows", "E2E coverage", "integration tests for UI", "page object pattern", "/e2e command", or discusses automated browser testing. Covers the E2E test loop workflow, Playwright patterns, page objects, and selector strategies.

中文简介正在准备中。当前显示原文说明。

SomtoUgeh/somto-dev-toolkit22026年8月26日 更新

gwt

免费

Manage git worktrees using a centralized worktrees folder. Use for parallel development, PR reviews, or isolated work.

中文简介正在准备中。当前显示原文说明。

SomtoUgeh/somto-dev-toolkit22026年8月26日 更新

查看 SomtoUgeh 的全部技能

报告此技能的问题