claude-skills-reference/engineering-team/tech-stack-evaluator/HOW_TO_USE.md

How to Use the Technology Stack Evaluator Skill

The Technology Stack Evaluator skill provides comprehensive evaluation and comparison of technologies, frameworks, and complete technology stacks for engineering teams.

Quick Start Examples

Example 1: Simple Technology Comparison

Conversational (Easiest):

Hey Claude—I just added the "tech-stack-evaluator" skill. Can you compare React vs Vue for building a SaaS dashboard?

What you'll get:

• Executive summary with recommendation
• Comparison matrix with scores
• Top 3 pros and cons for each
• Confidence level
• Key decision factors

---

Example 2: Complete Stack Evaluation

Hey Claude—I just added the "tech-stack-evaluator" skill. Can you evaluate this technology stack for a real-time collaboration platform:
- Frontend: Next.js
- Backend: Node.js + Express
- Database: PostgreSQL
- Real-time: WebSockets
- Hosting: AWS

Include TCO analysis and ecosystem health assessment.

What you'll get:

• Complete stack evaluation
• TCO breakdown (5-year projection)
• Ecosystem health scores
• Security assessment
• Detailed recommendations

---

Example 3: Migration Analysis

Hey Claude—I just added the "tech-stack-evaluator" skill. We're considering migrating from Angular.js (1.x) to React. Our codebase:
- 75,000 lines of code
- 300 components
- 8-person development team
- Must minimize downtime

Can you assess migration complexity, effort, risks, and timeline?

What you'll get:

• Migration complexity score (1-10)
• Effort estimate (person-months and timeline)
• Risk assessment (technical, business, team)
• Phased migration plan
• Success criteria

---

Example 4: TCO Analysis

Hey Claude—I just added the "tech-stack-evaluator" skill. Calculate total cost of ownership for AWS vs Azure for our workload:
- 50 EC2/VM instances (growing 25% annually)
- 20TB database storage
- Team: 12 developers
- 5-year projection

Include hidden costs like technical debt and vendor lock-in.

What you'll get:

• 5-year TCO breakdown
• Initial vs operational costs
• Scaling cost projections
• Cost per user metrics
• Hidden costs (technical debt, vendor lock-in, downtime)
• Cost optimization opportunities

---

Example 5: Security & Compliance Assessment

Hey Claude—I just added the "tech-stack-evaluator" skill. Assess the security posture of our current stack:
- Express.js (Node.js)
- MongoDB
- JWT authentication
- Hosted on AWS

We need SOC2 and GDPR compliance. What are the gaps?

What you'll get:

• Security score (0-100) with grade
• Vulnerability analysis (CVE counts by severity)
• Compliance readiness for SOC2 and GDPR
• Missing security features
• Recommendations to improve security

---

Example 6: Cloud Provider Comparison

Hey Claude—I just added the "tech-stack-evaluator" skill. Compare AWS vs Azure vs GCP for machine learning workloads:
- Priorities: GPU availability (40%), Cost (30%), ML ecosystem (20%), Support (10%)
- Need: High GPU availability for model training
- Team: 5 ML engineers, experienced with Python

Generate weighted decision matrix.

What you'll get:

• Weighted comparison matrix
• Scores across all criteria
• Best performer by category
• Overall recommendation with confidence
• Pros/cons for each provider

---

Input Formats Supported

1. Conversational Text (Easiest)

Just describe what you want in natural language:

"Compare PostgreSQL vs MongoDB for a SaaS application"
"Evaluate security of our Express.js + JWT stack"
"Calculate TCO for migrating to microservices"

2. Structured JSON

For precise control over evaluation parameters:

{
  "comparison": {
    "technologies": ["React", "Vue", "Svelte"],
    "use_case": "Enterprise dashboard",
    "weights": {
      "performance": 25,
      "developer_experience": 30,
      "ecosystem": 25,
      "learning_curve": 20
    }
  }
}

3. YAML (Alternative Structured Format)

comparison:
  technologies:
    - React
    - Vue
  use_case: SaaS dashboard
  priorities:
    - Developer productivity
    - Ecosystem maturity

4. URLs for Ecosystem Analysis

"Analyze ecosystem health for these technologies:
- https://github.com/facebook/react
- https://github.com/vuejs/vue
- https://www.npmjs.com/package/react"

The skill automatically detects the format and parses accordingly!

---

Report Sections Available

You can request specific sections or get the full report:

Available Sections:

1. Executive Summary (200-300 tokens) - Recommendation + top pros/cons
2. Comparison Matrix - Weighted scoring across all criteria
3. TCO Analysis - Complete cost breakdown (initial + operational + hidden)
4. Ecosystem Health - Community size, maintenance, viability
5. Security Assessment - Vulnerabilities, compliance readiness
6. Migration Analysis - Complexity, effort, risks, timeline
7. Performance Benchmarks - Throughput, latency, resource usage

Request Specific Sections:

"Compare Next.js vs Nuxt.js. Include only: ecosystem health and performance benchmarks. Skip TCO and migration analysis."

---

What to Provide

For Technology Comparison:

• Technologies to compare (2-5 recommended)
• Use case or application type (optional but helpful)
• Priorities/weights (optional, uses sensible defaults)

For TCO Analysis:

• Technology/platform name
• Team size
• Current costs (hosting, licensing, support)
• Growth projections (user growth, scaling needs)
• Developer productivity factors (optional)

For Migration Assessment:

• Source technology (current stack)
• Target technology (desired stack)
• Codebase statistics (lines of code, number of components)
• Team information (size, experience level)
• Constraints (downtime tolerance, timeline)

For Security Assessment:

• Technology stack components
• Security features currently implemented
• Compliance requirements (GDPR, SOC2, HIPAA, PCI-DSS)
• Known vulnerabilities (if any)

For Ecosystem Analysis:

• Technology name or GitHub/npm URL
• Specific metrics of interest (optional)

---

Output Formats

The skill adapts output based on your environment:

Claude Desktop (Rich Markdown)

• Formatted tables with visual indicators
• Expandable sections
• Color-coded scores (via markdown formatting)
• Decision matrices

CLI/Terminal (Terminal-Friendly)

• ASCII tables
• Compact formatting
• Plain text output
• Copy-paste friendly

The skill automatically detects your environment!

---

Advanced Usage

Custom Weighted Criteria:

"Compare React vs Vue vs Svelte.
Priorities (weighted):
- Developer experience: 35%
- Performance: 30%
- Ecosystem: 20%
- Learning curve: 15%"

Multiple Analysis Types:

"Evaluate Next.js for our enterprise SaaS platform.
Include: TCO (5-year), ecosystem health, security assessment, and performance vs Nuxt.js."

Progressive Disclosure:

"Compare AWS vs Azure. Start with executive summary only."

(After reviewing summary)
"Show me the detailed TCO breakdown for AWS."

---

Tips for Best Results

1. Be Specific About Use Case: "Real-time collaboration platform" is better than "web app"

2. Provide Context: Team size, experience level, constraints help generate better recommendations

3. Set Clear Priorities: If cost is more important than performance, say so with weights

4. Request Incremental Analysis: Start with executive summary, then drill into specific sections

5. Include Constraints: Zero-downtime requirement, budget limits, timeline pressure

6. Validate Assumptions: Review the TCO assumptions and adjust if needed

---

Common Questions

Q: How current is the data? A: The skill uses current data sources when available (GitHub, npm, CVE databases). Ecosystem metrics are point-in-time snapshots.

Q: Can I compare more than 2 technologies? A: Yes! You can compare 2-5 technologies. More than 5 becomes less actionable.

Q: What if I don't know the exact data for TCO analysis? A: The skill uses industry-standard defaults. Just provide what you know (team size, rough costs) and it will fill in reasonable estimates.

Q: Can I export reports? A: Yes! The skill can generate markdown reports that you can save or export.

Q: How do confidence scores work? A: Confidence (0-100%) is based on:

• Score gap between options (larger gap = higher confidence)
• Data completeness
• Clarity of requirements

Q: What if technologies are very close in scores? A: The skill will report low confidence and highlight that it's a close call, helping you understand there's no clear winner.

---

Need Help?

If results aren't what you expected:

1. Clarify your use case - Be more specific about requirements
2. Adjust priorities - Set custom weights for what matters most
3. Provide more context - Team skills, constraints, business goals
4. Request specific sections - Focus on what's most relevant

Example clarification:

"The comparison seemed to favor React, but we're a small team (3 devs) with no React experience. Can you re-evaluate with learning curve weighted at 40%?"

The skill will adjust the analysis based on your refined requirements!