跳到正文
cccskills
免费在 GitHub 公开

roadmap-generator

Generate phased implementation roadmaps with Epic/Story/Task breakdown, effort estimates, and validation checkpoints. Use when cto-architect needs to create actionable technical roadmaps from architecture designs.

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(3)

  • SKILL.md7.3 KB
  • estimation-guide.md6.7 KB
  • generator.py15.1 KB

SKILL.md(原文)

安装前可以查看交给 AI 代理的指令内容。

Roadmap Generator

Creates structured implementation roadmaps that transform architecture designs into actionable plans with clear phases, milestones, and validation gates.

When to Use

  • After architecture design is complete
  • When translating strategic goals into execution plans
  • Creating project timelines with realistic effort estimates
  • Breaking down large initiatives into manageable phases

Roadmap Structure

Three-Phase Framework

Every roadmap follows MVP → Scale → Advanced phases:

Phase 1: MVP (Foundation)
├── Goal: Prove core value proposition
├── Duration: Typically 6-12 weeks
├── Success Criteria: Core functionality working
└── Exit Gate: User validation / metrics threshold

Phase 2: Scale (Growth)
├── Goal: Handle production load, improve reliability
├── Duration: Typically 8-16 weeks
├── Success Criteria: Performance SLAs met
└── Exit Gate: Load testing passed / uptime targets

Phase 3: Advanced (Optimization)
├── Goal: Competitive features, optimization
├── Duration: Ongoing
├── Success Criteria: Feature parity / differentiation
└── Exit Gate: Business metrics achieved

Epic/Story/Task Hierarchy

Epic

High-level capability that delivers business value.

  • Duration: 2-6 weeks
  • Format: "Enable [user] to [capability]"
  • Example: "Enable customers to receive real-time order notifications"

Story

User-facing functionality within an epic.

  • Duration: 2-5 days
  • Format: "As a [user], I want to [action] so that [benefit]"
  • Example: "As a customer, I want push notifications so I know when my order ships"

Task

Technical work to complete a story.

  • Duration: 2-8 hours
  • Format: "[Verb] [object] [context]"
  • Example: "Implement WebSocket connection handler for notification service"

Effort Estimation Framework

T-Shirt Sizing

SizeStory PointsTypical DurationCharacteristics
XS12-4 hoursTrivial change, no unknowns
S20.5-1 daySimple, well-understood
M31-2 daysSome complexity, minor unknowns
L53-5 daysComplex, some research needed
XL81-2 weeksVery complex, significant unknowns
XXL13+> 2 weeksShould be broken down

Estimation Guidelines

  1. Include buffer: Add 20-30% for unknowns
  2. Account for context switching: Multiply by 1.3 for teams juggling priorities
  3. New technology penalty: Add 50% for unfamiliar tech
  4. Integration tax: Add 20% for each external system integration

Team Velocity Calculation

Effective Capacity = Team Size × Working Days × Focus Factor × Skill Factor

Focus Factor:
- Dedicated team: 0.8
- 50% allocated: 0.4
- Ad-hoc: 0.2

Skill Factor:
- Expert team: 1.0
- Experienced: 0.8
- Learning curve: 0.5

Validation Checkpoints

Phase Gate Reviews

Each phase ends with a validation gate:

## Phase 1 Exit Gate

### Go Criteria
- [ ] Core features functional in staging
- [ ] Unit test coverage > 70%
- [ ] Integration tests passing
- [ ] Performance baseline established
- [ ] Security review completed

### No-Go Signals
- Critical bugs unresolved
- Core functionality missing
- Team capacity insufficient for Phase 2
- Business requirements changed significantly

### Decision
[ ] GO - Proceed to Phase 2
[ ] CONDITIONAL GO - Proceed with identified risks
[ ] NO-GO - Address blockers first

Mid-Phase Checkpoints

  • Week 2: Architecture validation
  • Week 4: Integration milestone
  • Week 6: Feature complete checkpoint
  • Weekly: Sprint review and course correction

Dependency Management

Dependency Types

TypeDescriptionMitigation
TechnicalService A needs Service BStart B early, use mocks
TeamNeed DevOps for deploymentReserve capacity early
ExternalThird-party API integrationValidate early, have fallback
DataNeed production data for testingCreate synthetic data

Critical Path Identification

  1. List all dependencies
  2. Identify longest chain
  3. Mark critical path items
  4. Add buffer to critical path (25%)
  5. Parallelize non-critical work

Output Format

Roadmap Document Structure

# Implementation Roadmap: [Project Name]

**Created**: [Date]
**Owner**: [Team/Person]
**Architecture Reference**: [Link to architecture doc]

## Executive Summary
[2-3 sentences on approach and timeline]

## Timeline Overview
[Visual timeline or Gantt representation]

## Phase 1: MVP (Weeks 1-8)

### Phase Goals
- [Goal 1]
- [Goal 2]

### Epics

#### Epic 1.1: [Name]
**Duration**: [X weeks]
**Dependencies**: [List]

| Story | Points | Owner | Dependencies |
|-------|--------|-------|--------------|
| [Story 1] | M | [Team] | None |
| [Story 2] | L | [Team] | Story 1 |

**Tasks for Story 1**:
- [ ] [Task 1] (S)
- [ ] [Task 2] (M)

### Phase 1 Exit Criteria
- [ ] [Criterion 1]
- [ ] [Criterion 2]

### Phase 1 Risks
| Risk | Probability | Impact | Mitigation |
|------|-------------|--------|------------|
| [Risk] | High/Med/Low | High/Med/Low | [Action] |

---

## Phase 2: Scale (Weeks 9-16)
[Same structure as Phase 1]

---

## Phase 3: Advanced (Weeks 17+)
[Same structure as Phase 1]

---

## Resource Requirements

### Team Composition
| Role | Phase 1 | Phase 2 | Phase 3 |
|------|---------|---------|---------|
| Backend | 2 FTE | 3 FTE | 2 FTE |
| Frontend | 1 FTE | 2 FTE | 1 FTE |
| DevOps | 0.5 FTE | 1 FTE | 0.5 FTE |

### Infrastructure Costs
| Phase | Monthly Cost | Notes |
|-------|-------------|-------|
| Phase 1 | $X | Development environment |
| Phase 2 | $Y | Staging + production |
| Phase 3 | $Z | Full scale |

---

## Success Metrics

| Metric | Phase 1 Target | Phase 2 Target | Phase 3 Target |
|--------|---------------|----------------|----------------|
| [Metric 1] | [Value] | [Value] | [Value] |
| [Metric 2] | [Value] | [Value] | [Value] |

---

## Appendix

### Assumptions
- [Assumption 1]
- [Assumption 2]

### Open Questions
- [Question 1]
- [Question 2]

Anti-Patterns to Avoid

The Big Bang

Planning everything in detail upfront without iteration. Fix: Plan Phase 1 in detail, Phase 2 at epic level, Phase 3 at theme level.

The Feature Factory

Listing features without connecting to business outcomes. Fix: Every epic must connect to a measurable business goal.

The Happy Path

No buffer, no risk mitigation, assumes everything goes perfectly. Fix: Add 25-30% buffer, identify top 5 risks with mitigations.

The Kitchen Sink

Including every possible feature in MVP. Fix: Ruthlessly cut to core value proposition only.


Python Utilities

This skill includes Python utilities for roadmap generation:

  • generator.py - Core roadmap generation logic
  • templates/ - Output templates (markdown, JSON)

See generator.py for programmatic roadmap generation.


References

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

Detect common technical and organizational anti-patterns in proposals, architectures, and plans. Use when strategic-cto-mentor needs to identify red flags before they become problems.

中文简介正在准备中。当前显示原文说明。

alirezarezvani/claude-cto-team1182025年12月18日 更新

Recommend architecture patterns (monolith, microservices, serverless, modular monolith) based on scale, team size, and constraints. Use when cto-architect needs to select the right architectural approach for a new system or migration.

中文简介正在准备中。当前显示原文说明。

alirezarezvani/claude-cto-team1182025年12月18日 更新

Identify and challenge implicit assumptions in plans, proposals, and technical decisions. Use when strategic-cto-mentor needs to surface hidden assumptions and wishful thinking before they become costly mistakes.

中文简介正在准备中。当前显示原文说明。

alirezarezvani/claude-cto-team1182025年12月18日 更新

Generate targeted clarifying questions (2-3 max) that challenge vague requirements and extract missing context. Use after request-analyzer identifies clarification needs, before routing to specialist agents. Helps cto-orchestrator avoid delegating unclear requirements.

中文简介正在准备中。当前显示原文说明。

alirezarezvani/claude-cto-team1182025年12月18日 更新

Infrastructure and development cost estimation for technical projects. Use when planning budgets, evaluating build vs buy decisions, or projecting TCO for architecture choices.

中文简介正在准备中。当前显示原文说明。

alirezarezvani/claude-cto-team1182025年12月18日 更新

Transform clarified user requests into structured delegation prompts optimized for specialist agents (cto-architect, strategic-cto-mentor, cv-ml-architect). Use after clarification is complete, before routing to specialist agents. Ensures agents receive complete context for effective work.

中文简介正在准备中。当前显示原文说明。

alirezarezvani/claude-cto-team1182025年12月18日 更新

查看 alirezarezvani 的全部技能

报告此技能的问题