faion-researcher（法安研究员）

Entry point: /faion-net — invoke this skill for automatic routing to the appropriate domain.

Research Domain Skill

Communication: User's language. Docs: English.

Purpose

Orchestrate all research and discovery activities for product/startup development. Coordinates 2 specialized sub-skills for comprehensive research coverage.

Context Discovery

Auto-Investigation

Check for existing research and product context:

Read existing assets:

• Check landing page for current positioning/messaging
• Read constitution.md for product context
• Check any existing research docs

Discovery Questions

Use AskUserQuestion to understand research needs.

Q1: Research Stage

question: "What stage of research are you in?"
header: "Stage"
multiSelect: false
options:
  - label: "Exploring ideas (no specific direction)"
    description: "Generate and evaluate new ideas"
  - label: "Have an idea, need validation"
    description: "Test if idea is worth pursuing"
  - label: "Committed, need market intelligence"
    description: "Understand market, competitors, pricing"
  - label: "Building, need user insights"
    description: "Understand users for better product"

Routing:

• "Exploring ideas" → Skill(faion-market-researcher) → idea-generation
• "Need validation" → Skill(faion-user-researcher) → problem-validation
• "Market intelligence" → Skill(faion-market-researcher) → competitors, TAM, pricing
• "User insights" → Skill(faion-user-researcher) → personas, JTBD, pain points

Q2: Research Focus (if market intelligence)

question: "What do you need to understand?"
header: "Focus"
multiSelect: true
options:
  - label: "Market size (TAM/SAM/SOM)"
    description: "How big is the opportunity?"
  - label: "Competitors & alternatives"
    description: "Who else solves this problem?"
  - label: "Pricing benchmarks"
    description: "What do customers pay for similar?"
  - label: "Industry trends"
    description: "Where is the market heading?"

Routing:

• "Market size" → tam-sam-som-analysis
• "Competitors" → competitor-analysis
• "Pricing" → pricing-research
• "Trends" → trend-analysis

Q3: User Research Type (if user insights)

question: "What user insights do you need?"
header: "Insights"
multiSelect: true
options:
  - label: "Who are my users? (personas)"
    description: "User profiles and segments"
  - label: "What problems do they have?"
    description: "Pain points and frustrations"
  - label: "What are they trying to achieve?"
    description: "Jobs-to-be-done, goals"
  - label: "Would they pay for a solution?"
    description: "Willingness to pay, value perception"

Routing:

• "Personas" → persona-building
• "Problems" → pain-point-research
• "Achieve" → jobs-to-be-done
• "Pay" → value-proposition + pricing-research

Q4: Evidence Standards

question: "What would convince you the research is solid?"
header: "Evidence"
multiSelect: false
options:
  - label: "Quick directional insights"
    description: "Enough to make a decision, move fast"
  - label: "Backed by data sources"
    description: "Links to research, reports, data"
  - label: "Primary research (interviews/surveys)"
    description: "Talk to real users/customers"

Context impact:

• "Quick" → Web search, secondary research, fast output
• "Backed by data" → Include sources, citations, methodology
• "Primary research" → Interview guides, survey design, synthesis

Architecture

faion-researcher (orchestrator)
├── faion-market-researcher (22 methodologies)
│   ├── Market sizing (TAM/SAM/SOM)
│   ├── Competitor analysis
│   ├── Pricing research
│   ├── Trend analysis
│   ├── Niche evaluation
│   ├── Business model planning
│   └── Idea generation
└── faion-user-researcher (21 methodologies)
    ├── Persona building
    ├── User interviews
    ├── Pain point research
    ├── Jobs-to-be-done
    ├── Problem validation
    ├── Value proposition design
    └── Survey design

Total: 43 methodologies across 2 sub-skills

Quick Decision Tree

Sub-Skills (2)

faion-market-researcher

Focus: Market & business intelligence

Files: 22 | Location: faion-market-researcher/

faion-user-researcher

Focus: User research & validation

Files: 21 | Location: faion-user-researcher/

Agents (2)

Details: agent-invocation.md

Research Modes (9)

Core Workflows

1. Idea Discovery

Context → Generate Ideas → Select → Pain Research → Niche Eval → Results
Sub-skills: market-researcher (ideas, niche) + user-researcher (pains)

2. Product Research

Parse Project → Read Docs → Select Modules → Sequential Execution → Summary
Sub-skills: market-researcher (market, competitors, pricing) + user-researcher (personas, validate)

3. Project Naming

Gather Concept → Generate Names → Select → Check Domains → Results
Sub-skill: market-researcher (naming)

Details: workflows.md

Quick Reference

Output Files

All outputs go to .aidocs/product_docs/:

Integration

Entry Point

Invoked via /faion-net when intent is research-related:

if intent in ["idea", "research", "market", "competitors", "naming", "personas", "pricing"]:
    invoke("faion-researcher")

Sub-Skill Invocation

# Market intelligence
invoke("faion-market-researcher", mode="market")

# User research
invoke("faion-user-researcher", mode="personas")

# Full research (sequential)
invoke("faion-market-researcher", mode="market")
invoke("faion-market-researcher", mode="competitors")
invoke("faion-user-researcher", mode="personas")
invoke("faion-user-researcher", mode="validate")

Next Steps

After research complete, offer:

• "Create GTM Manifest?" → faion-marketing-manager
• "Create spec.md?" → faion-sdd
• "Start development?" → faion-software-developer

Rules

1. Sequential execution - Run agents ONE BY ONE (not parallel)
2. Sub-skill routing - Market topics → market-researcher, User topics → user-researcher
3. Source citations - All research must include URLs
4. Data quality - If not found → "Data not available"

*faion-researcher v2.0* *Orchestrator with 2 sub-skills* *Total: 43 methodologies | Agents: 2 | Modes: 9*