consultancy-practices
About
This skill provides structured practices for software quality consulting, helping developers advise clients or establish consultancy workflows. It emphasizes listening-first approaches, impact/effort prioritization, and knowledge transfer. Key features include engagement type guidance and success metric tracking for quality assessments.
Quick Install
Claude Code
Recommended/plugin add https://github.com/majiayu000/claude-skill-registrygit clone https://github.com/majiayu000/claude-skill-registry.git ~/.claude/skills/consultancy-practicesCopy and paste this command in Claude Code to install this skill
Documentation
Consultancy Practices
<default_to_action> When consulting on quality:
- LISTEN FIRST: Understand their context before prescribing solutions
- DISCOVER: What's the pain? What have they tried? What are constraints?
- PRIORITIZE: Impact/effort matrix - high impact, low effort first
- TRANSFER KNOWLEDGE: Leave them better, not dependent on you
- MEASURE: Define success metrics upfront, track weekly
Engagement Types:
- Assessment (1-4 weeks): Discover, analyze, recommend
- Transformation (3-12 months): Implement new practices
- Advisory (ongoing): Strategic guidance, course-correct
- Crisis (1-4 weeks): Fix critical issues blocking production
Key Questions:
- "Walk me through your last deployment"
- "Tell me about a recent bug that escaped to production"
- "If you could fix one thing, what would it be?" </default_to_action>
Quick Reference Card
The Consulting Process
| Phase | Duration | Goal | Deliverable |
|---|---|---|---|
| Discovery | Week 1-2 | Understand context | Interview notes, observations |
| Analysis | Week 2-3 | Identify root causes | Impact/effort matrix |
| Recommendations | Week 3-4 | Present findings | Report with roadmap |
| Implementation | Month 2-6+ | Execute changes | Working system, trained team |
| Transition | Final month | Ensure self-sufficiency | Handover docs |
Impact/Effort Matrix
| Priority | What | Action |
|---|---|---|
| High Impact, Low Effort | Quick wins | Do first |
| High Impact, High Effort | Major initiatives | Plan carefully |
| Low Impact, Low Effort | Nice-to-haves | If time permits |
| Low Impact, High Effort | Distractions | Skip |
Common Patterns
"We Need Test Automation"
What they say: "We need test automation" What they mean: "Manual testing is too slow/expensive"
Discovery: How long is regression? What's deployment frequency?
Typical Finding: They need faster feedback, not "automation"
Recommendation:
- Unit tests for new code (TDD)
- Smoke tests for critical paths
- Keep exploratory for discovery
- Build automation incrementally
"Fix Our Quality Problem"
What they say: "We have too many bugs" What they mean: "Something is broken but we don't know what"
Discovery: Where found? What types? When introduced?
Typical Finding: No test strategy, testing too late, poor feedback loops
Recommendation:
- Shift testing left
- Improve coverage on critical paths
- Speed up CI/CD feedback
- Better requirements/acceptance criteria
"We Want to Scale Quality"
What they say: "Growing fast, quality can't keep up" What they mean: "Can't hire enough QA fast enough"
Discovery: Current QA:Dev ratio? Where's QA spending time?
Typical Finding: QA is bottleneck - manual regression, gatekeeping
Recommendation:
- Make QA strategic, not tactical
- Developers own test automation
- QA focuses on exploratory, risk analysis
- Use agentic approaches for scale
Anti-Patterns
| Anti-Pattern | Problem | Better |
|---|---|---|
| Cookie-Cutter | Same solution everywhere | Context-specific recommendations |
| Tool Pusher | Recommend expensive tools | Tools that solve actual problems |
| Process Nazi | Impose rigid process | Lightweight, fits their culture |
| Permanent Fixture | Never leave, create dependency | Work toward them not needing you |
| Blame Game | Point fingers at people | Fix systems, not blame people |
Difficult Situations
"We already tried that" → "Tell me what you tried and what didn't work" (learn from their experience)
"Our context is special" → "Help me understand what makes yours special" (they might be right, or making excuses)
"We don't have budget/time" → "What's the cost of not fixing this? Let's start small" (show ROI)
"That won't work here" → "What specific constraints? Let's adapt" (find what WILL work)
Agent Integration
// Automated codebase assessment
const assessment = await Task("Assess Codebase", {
scope: 'client-project/',
depth: 'comprehensive',
reportFormat: 'executive-summary'
}, "qe-quality-analyzer");
// Returns: { qualityScore, testCoverage, technicalDebt, recommendations }
// ROI analysis for quality initiatives
const roi = await Task("Calculate ROI", {
currentState: { defectEscapeRate: 0.15, mttr: 48 },
proposedImprovements: ['test-automation', 'ci-cd-pipeline'],
timeframe: '6-months'
}, "qe-quality-analyzer");
// Returns: { estimatedCost, estimatedSavings, paybackPeriod }
Agent Coordination Hints
Memory Namespace
aqe/consultancy/
├── assessments/* - Client assessments
├── recommendations/* - Prioritized recommendations
├── roi-analysis/* - ROI calculations
└── progress/* - Implementation tracking
Fleet Coordination
const consultingFleet = await FleetManager.coordinate({
strategy: 'client-engagement',
agents: [
'qe-quality-analyzer', // Assess current state
'qe-regression-risk-analyzer', // Risk assessment
'qe-quality-gate', // Define quality gates
'qe-deployment-readiness' // Deployment maturity
],
topology: 'hierarchical'
});
Related Skills
- quality-metrics - Metrics for client reporting
- risk-based-testing - Client risk assessment
- holistic-testing-pact - Comprehensive strategy
Remember
Good consulting is about empowering teams, not creating dependency. Your success is measured by them not needing you anymore - while still wanting to work with you again.
Best compliment: "We've got this now, but when we tackle X next year, we're calling you."
Be honest. Be helpful. Be context-driven. Leave them better.
GitHub Repository
Related Skills
consultancy-practices
OtherThis skill applies proven software quality consultancy practices for advising clients or establishing consultancy workflows. It provides structured approaches for assessment, transformation, and advisory engagements, emphasizing listening, prioritization, and knowledge transfer. Key features include quick reference guidance and integration with quality analysis agents.
content-collections
MetaThis skill provides a production-tested setup for Content Collections, a TypeScript-first tool that transforms Markdown/MDX files into type-safe data collections with Zod validation. Use it when building blogs, documentation sites, or content-heavy Vite + React applications to ensure type safety and automatic content validation. It covers everything from Vite plugin configuration and MDX compilation to deployment optimization and schema validation.
creating-opencode-plugins
MetaThis skill provides the structure and API specifications for creating OpenCode plugins that hook into 25+ event types like commands, files, and LSP operations. It offers implementation patterns for JavaScript/TypeScript modules that intercept and extend the AI assistant's lifecycle. Use it when you need to build event-driven plugins for monitoring, custom handling, or extending OpenCode's capabilities.
sglang
MetaSGLang is a high-performance LLM serving framework that specializes in fast, structured generation for JSON, regex, and agentic workflows using its RadixAttention prefix caching. It delivers significantly faster inference, especially for tasks with repeated prefixes, making it ideal for complex, structured outputs and multi-turn conversations. Choose SGLang over alternatives like vLLM when you need constrained decoding or are building applications with extensive prefix sharing.
