Persona MCP

# Context Detection - Smart Expert Matching ## How Context Detection Works ### Overview The system analyzes conversation context to suggest the most appropriate expert(s) for your needs. This document explains the detection logic and how to get the best expert matches. --- ## Keyword-Based Detection ### Engineering Keywords | Keywords | Suggested Expert | |----------|------------------| | React, Next.js, frontend, component | Frontend Engineer (107), React Expert (102) | | API, backend, server, database | Backend Engineer (106), Database Engineer (114) | | Docker, Kubernetes, CI/CD, deploy | DevOps Engineer (110) | | Security, auth, OWASP, vulnerability | Security Engineer (113) | | Python, ML, model, training | Data Scientist (402), ML Engineer (401) | | LLM, language model, prompt, RAG | LLM Engineer (410), Prompt Engineer (408) | | Test, QA, automation, Cypress | QA Lead (501), Test Automation Engineer (502) | | Performance, load, latency, optimize | Performance Engineer (120), Performance Tester (503) | ### Business Keywords | Keywords | Suggested Expert | |----------|------------------| | Strategy, market, competition | Strategy Consultant (301) | | Product, roadmap, feature, prioritize | Product Manager (305) | | Startup, funding, investor, pitch | VC Partner (310), Startup Advisor (313) | | Revenue, P&L, forecast, budget | CFO (309), Financial Analyst (312) | | Growth, acquisition, retention, funnel | Growth Manager (307) | | Innovation, disrupt, new market | Innovation Director (302) | ### Design Keywords | Keywords | Suggested Expert | |----------|------------------| | UX, user research, usability | UX Designer (201), UX Researcher (205) | | UI, visual, design system | UI Designer (202) | | Brand, logo, identity | Brand Designer (207) | | Animation, motion, interaction | Interaction Designer (204) | | Accessibility, WCAG, a11y | Accessibility Tester (507) | ### Leadership Keywords | Keywords | Suggested Expert | |----------|------------------| | Team, leadership, culture | Leadership Coach (801) | | Executive, C-level, board | Executive Coach (802) | | Negotiation, deal, contract | Negotiation Expert (903), Legal Advisor (901) | | Conflict, communication, feedback | Team Coach (803) | --- ## Intent-Based Detection ### Problem Types → Expert Mapping ``` "I have a bug in..." └── Fullstack Developer (101) or relevant specialist "How do I design..." └── UX Designer (201) + UI Designer (202) "Help me decide..." └── Strategy Consultant (301) or relevant domain expert "I need to hire..." └── HR Director (317) + Leadership Coach (801) "Is this secure..." └── Security Engineer (113) "How should I test..." └── QA Lead (501) + relevant specialist "Analyze this data..." └── Data Scientist (402) "Review this architecture..." └── Solution Architect (108) "Improve performance of..." └── Performance Engineer (120) "I'm presenting to..." └── Executive Coach (802) + Pitch Coach (315) ``` --- ## Multi-Expert Detection ### Scenarios Requiring Multiple Experts ``` SCENARIO: "Building a new feature" SUGGESTED CHAIN: ├── Product Manager (305): Define requirements ├── UX Designer (201): Design experience ├── Frontend Engineer (107): Implement └── QA Lead (501): Test SCENARIO: "Launching a startup" SUGGESTED CHAIN: ├── Startup Advisor (313): Strategy ├── Product Manager (305): Product ├── CFO (309): Finance └── Legal Advisor (901): Legal structure SCENARIO: "Scaling infrastructure" SUGGESTED CHAIN: ├── Solution Architect (108): Architecture ├── DevOps Engineer (110): Infrastructure ├── Security Engineer (113): Security └── SRE (111): Reliability SCENARIO: "AI/ML product launch" SUGGESTED CHAIN: ├── AI Product Manager (411): Strategy ├── LLM Engineer (410): Implementation ├── MLOps Engineer (406): Deployment └── Security Engineer (113): AI security ``` --- ## Confidence Levels ### High Confidence (Direct Match) ``` User: "Review my React component" Detection: React Expert (102) Confidence: HIGH Reason: Direct technology mention ``` ### Medium Confidence (Domain Match) ``` User: "Help me with my frontend" Detection: Frontend Engineer (107), React Expert (102), Vue Specialist (103) Confidence: MEDIUM Reason: Domain match, multiple specialists possible Response: "What technology stack are you using?" ``` ### Low Confidence (Broad Query) ``` User: "I need help with my app" Detection: Multiple possibilities Confidence: LOW Reason: Insufficient context Response: "Could you describe: 1. What type of app (web/mobile/backend)? 2. What specific challenge? 3. What technology stack?" ``` --- ## Improving Detection Accuracy ### Be Specific ``` VAGUE: "Help with my code" BETTER: "Help with Python FastAPI authentication" → Backend Engineer (106) + Security Engineer (113) VAGUE: "Design advice" BETTER: "UX design for e-commerce checkout flow" → UX Designer (201) + Interaction Designer (204) VAGUE: "Business question" BETTER: "Should we acquire this competitor?" → Strategy Consultant (301) → CFO (309) → Legal Advisor (901) ``` ### Include Context ``` MINIMAL: "Python help" BETTER: "I'm building a machine learning pipeline in Python and need help with feature engineering" → Data Scientist (402) + ML Engineer (401) MINIMAL: "Database question" BETTER: "PostgreSQL query optimization for a table with 100M rows, getting slow reads" → Database Engineer (114) + Performance Engineer (120) ``` ### State Your Goal ``` TASK-FOCUSED: "Write a test" GOAL-FOCUSED: "Ensure my payment flow is reliable and secure" → QA Lead (501) + Security Tester (504) + API Tester (506) TASK-FOCUSED: "Make a chart" GOAL-FOCUSED: "Visualize revenue trends for board presentation" → Data Scientist (402) + Executive Coach (802) ``` --- ## Manual Expert Selection ### Direct Activation ``` @strategic-oracle @fullstack-developer @ux-designer @cfo ``` ### Domain Request ``` "I need an engineering expert for..." "Connect me with a business strategist" "I want to talk to a design expert" ``` ### Chain Request ``` "Chain strategy-consultant and cfo" "Run product-manager then ux-designer then frontend-engineer" "Parallel: security-engineer and legal-advisor" ``` --- ## Expert Activation The system will automatically suggest experts based on your query. For best results: 1. Describe your challenge clearly 2. Include relevant technology/domain 3. State your goal or desired outcome 4. Mention any constraints Example: ``` "I'm building a B2B SaaS product. We have React frontend, Node.js backend, PostgreSQL database. Currently 10k users, growing 20% monthly. Need help with scaling architecture while maintaining security compliance for enterprise customers." → Suggested: Solution Architect (108) → Security Engineer (113) → DevOps Engineer (110) → Database Engineer (114) ```