shape-up-analysis

📁 tachun/shape-up-analysis-skills 📅 13 days ago
2
总安装量
1
周安装量
#66369
全站排名
安装命令
npx skills add https://github.com/tachun/shape-up-analysis-skills --skill shape-up-analysis

Agent 安装分布

cursor 1
claude-code 1
antigravity 1

Skill 文档

Shape Up Pitch & Codebase Analysis

Analyze Shape Up pitches against actual codebase to produce comprehensive kickoff documents with risk assessment and confidence rating.

References

Load as needed for detailed guidance:

  • references/shape-up-methodology.md – Shape Up concepts, pitch anatomy, betting criteria
  • references/risk-assessment.md – Risk categorization, rabbit hole identification, confidence scoring
  • references/codebase-analysis.md – Systematic code review approach, health signals, impact mapping

Examples

See for expected quality:

  • examples/example-pitch.md – Well-formed Shape Up pitch
  • examples/example-output.md – Complete analysis output demonstrating format and depth

Required Inputs

  1. Pitch document – Shape Up format (problem, appetite, solution, rabbit holes, no-gos)
  2. Codebase access – Available in current repository

If pitch not provided, ask user before proceeding.

Dual Perspective Analysis

Analyze as both Senior PM and Senior Developer:

PM lens:

  • Is problem clearly defined or solution disguised as problem?
  • Who benefits and why now?
  • What does “good enough” look like?
  • What’s measurable?

Dev lens:

  • What does affected code actually look like?
  • What complexity hides in “simple” requirements?
  • What technical debt will we encounter?
  • What’s missing from pitch that code reveals?

Principles:

  • Clarity over optimism—don’t sugarcoat
  • Time is limited—honor the appetite
  • Be constructive—risks need mitigations
  • Challenge assumptions—mark as ✅ Safe / ⚠️ Risky / ❓ Unknown

Process

Phase 1: Understand Pitch

  1. Read pitch document carefully
  2. Extract: problem, appetite, solution, rabbit holes, no-gos
  3. Identify explicit scope boundaries
  4. Note assumptions (stated and implied)
  5. Flag missing elements

Phase 2: Codebase Exploration

  1. Identify affected code areas using Glob/Grep
  2. For each area:
    • Read relevant files
    • Assess health (tests, complexity, docs)
    • Note dependencies and integrations
    • Flag legacy or fragile code
  3. Map pitch requirements to code paths
  4. Find hidden complexity pitch doesn’t mention

Phase 3: Risk Analysis

  1. Cross-reference pitch assumptions with codebase reality
  2. Categorize:
    • Known risks: Identified problems with clear impact
    • Rabbit holes: Could explode in scope
    • Unknowns: Need spikes to assess
  3. Rate each: likelihood (High/Med/Low) × impact (High/Med/Low)
  4. Propose mitigations

Phase 4: Synthesis

  1. Rate bet confidence (1-5 stars)
  2. List blocking questions
  3. Suggest spikes, experiments, decisions
  4. Note accepted trade-offs

Output Format

Generate project-kickoff-analysis.md:

# Project Kickoff – Pitch & Codebase Analysis

## 1. Project Summary (TL;DR)
One paragraph: problem, why now, who benefits. No jargon.

---

## 2. Understanding the Problem (PM View)

### 2.1 Problem Statement
- Real user/business problem?
- Clearly defined or solution disguised as problem?

### 2.2 Why Now?
- What triggered this pitch?
- Cost of not doing this?

### 2.3 Success Criteria
- How we know it worked
- What's measurable
- What's "good enough"

---

## 3. Proposed Solution (Critical Review)

### 3.1 Solution Overview
- Proposed solution summary
- Explicitly in scope
- Explicitly out of scope

### 3.2 Assumptions
| Assumption | Type | Status |
|------------|------|--------|
| Description | Technical/Product/User | ✅/⚠️/❓ |

---

## 4. Risks & Rabbit Holes

### 4.1 Known Risks
| Risk | Likelihood | Impact | Mitigation |
|------|------------|--------|------------|
| Description | H/M/L | H/M/L | Strategy |

### 4.2 Potential Rabbit Holes
List with concrete examples:
- What looks simple but isn't
- What could explode in scope
- Hidden dependencies

---

## 5. Codebase Review (Dev View)

### 5.1 High-Level Architecture
Brief tech stack overview relevant to this pitch.

### 5.2 Relevant Areas
| Area | Why It Matters | Confidence |
|------|---------------|------------|
| `path/to/module` | Reason | 👍/😐/🚨 |

**Legend:** 👍 Well understood, 😐 Needs exploration, 🚨 High risk

### 5.3 Code Health Signals
- Test coverage by area
- Complexity hotspots
- Technical debt markers
- Documentation gaps

---

## 6. Constraints & Boundaries

### 6.1 Time Box
- Fixed appetite: X weeks
- Must fit vs. can drop

### 6.2 Technical Constraints
- Required/forbidden tech
- Performance/security limits

---

## 7. Open Questions

| Question | Type | Can Proceed? | Risk if Unanswered |
|----------|------|--------------|-------------------|
| Specific question | Product/Tech/Data | Yes/No | Low/Med/High |

---

## 8. Bet Confidence

**Overall: ⭐⭐⭐☆☆** (X/5)

**Positive factors:**
- List

**Negative factors:**
- List

**What would increase confidence:**
- List

**What would kill the bet:**
- List

---

## 9. Suggested Next Steps

### Spikes to Run
- [ ] Spike name (timebox): Specific question to answer

### Experiments to Validate
- [ ] Experiment: What to test

### Decisions Needed Before Building
- [ ] Decision: What needs resolving

---

## 10. Final Notes
- Uncomfortable truths
- Trade-offs we're accepting

Quality Checklist

Before finalizing output:

  • Every risk has mitigation strategy
  • Code areas have specific paths, not vague descriptions
  • Confidence ratings backed by evidence
  • Open questions prioritized by blocker status
  • Spikes are timeboxed with specific questions
  • No vague or open-ended next steps