Vibecoding Lg MCP A2a
ttimes λ°μ΄λΈμ½λ© 컨νΌλ°μ€ λΌμ΄λΈ μ½λ© - 리μμΉ / λ³΄κ³ μ μμ± Multi Agent νλ‘μ νΈ
Installation
npx vibecoding-lg-mcp-a2aAsk AI about Vibecoding Lg MCP A2a
Powered by Claude Β· Grounded in docs
I know everything about Vibecoding Lg MCP A2a. Ask me about installation, configuration, usage, or troubleshooting.
0/500
Reviews
Documentation
TTimes Guide Coding - Multi-Agent Report Generation System
μΉ κ²μκ³Ό λ³΄κ³ μ μμ±μ μλννλ LangGraph κΈ°λ° λ©ν° μμ΄μ νΈ μμ€ν
MCP & A2A λ‘ μ€μ Multi Agents λ§λ€μ΄λ³΄κΈ° νλ‘μ νΈ

κ°μ
λ³Έ νλ‘μ νΈλ μ¬μ©μμ μμ²μ λ°λΌ μΉ κ²μ, λ¬Έμ κ²μ, λ³΄κ³ μ κ³ν μ립, λ³΄κ³ μ μμ±μ μλμΌλ‘ μννλ AI λ©ν° μμ΄μ νΈ μμ€ν μ λλ€. LangGraphμ A2A(Agent-to-Agent) νλ‘ν μ½μ μ¬μ©νμ¬ μ¬λ¬ μ λ¬Έ μμ΄μ νΈλ€μ΄ νλ ₯νμ¬ κ³ νμ§μ λ³΄κ³ μλ₯Ό μμ±ν©λλ€.
μ£Όμ κΈ°λ₯
- π μ€μκ° μΉ κ²μ - Tavily Search APIλ₯Ό ν΅ν μ΅μ μ 보 μμ§
- π λ²‘ν° DB κ²μ - PostgreSQL pgvectorλ₯Ό μ΄μ©ν μ μ¬ λ¬Έμ κ²μ
- π μλ κ³ν μ립 - AI κΈ°λ° μμ κ³ν λ° κ΅¬μ‘° μ€κ³
- βοΈ λ³΄κ³ μ μμ± - μμ§λ μ 보λ₯Ό μ’ ν©ν μ λ¬Έμ μΈ λ³΄κ³ μ μμ±
- π€ MCP ν΅ν© - Model Context Protocolμ ν΅ν λꡬ ν΅ν©
- π A2A ν΅μ - Google ADK κΈ°λ° μμ΄μ νΈ κ° ν΅μ
Tech Stack
- Python 3.13
- LangChain - LLM μ ν리μΌμ΄μ νλ μμν¬
- LangGraph - μν κΈ°λ° AI μν¬νλ‘μ°
- A2A Protocol - μμ΄μ νΈ κ° ν΅μ νμ€
- FastMCP - Model Context Protocol μλ²
- Google ADK - Agent Development Kit
- FastAPI - μΉ μλ²
- PostgreSQL + pgvector - λ²‘ν° λ°μ΄ν°λ² μ΄μ€
- Redis - μΊμ λ° μΈμ κ΄λ¦¬
- Azure OpenAI - LLM μ 곡μ
μμ€ν μν€ν μ²
βββββββββββββββββββββββββββ
β UnifiedResearch Agent β β Google ADK κΈ°λ° μ€μΌμ€νΈλ μ΄ν°
β (ADK Client) β
βββββββββββββ¬ββββββββββββββ
β A2A Protocol
βββββββββ΄ββββββββ¬βββββββββββββββ
βΌ βΌ βΌ
βββββββββββ ββββββββββββ ββββββββββββ
βPlanning β β Research β β Report β
β Agent β β Agent β β Writing β
β(A2A Srv)β β(A2A Srv) β β(A2A Srv) β
ββββββ¬βββββ ββββββ¬ββββββ ββββββββββββ
β β MCP Protocol
β βΌ
β ββββββββββββββββ
β β All-Search β
β β MCP Server β
β βββββ¬βββββββ¬ββββ
β β β
β βΌ βΌ
β [Tavily] [LangConnect]
β β
βββββββββββββββββββββββΌββββββββββββββ
βΌ βΌ
[PostgreSQL+pgvector] [Redis]
μ€μΉ λ°©λ²
1. μ¬μ μꡬμ¬ν
- Python 3.13+
- Docker & Docker Compose
- uv (Python ν¨ν€μ§ λ§€λμ )
2. νλ‘μ νΈ ν΄λ‘
git clone https://github.com/yourusername/ttimes_guide_coding.git
cd ttimes_guide_coding
3. νκ²½ μ€μ
# νκ²½ λ³μ νμΌ μμ±
cp .env.example .env
# .env νμΌμ νΈμ§νμ¬ νμν API ν€ μ
λ ₯:
# - AZURE_OPENAI_API_KEY
# - AZURE_OPENAI_ENDPOINT
# - AZURE_OPENAI_DEPLOYMENT_NAME
# - TAVILY_API_KEY
# - GOOGLE_API_KEY (ADKμ©)
# - κΈ°ν μ€μ κ°
4. μμ‘΄μ± μ€μΉ
# uvλ₯Ό μ¬μ©ν μμ‘΄μ± μ€μΉ
uv sync --dev
# λλ κ°λ° μμ‘΄μ± μ μΈ
uv sync
μ€ν λ°©λ²
1. LangConnect μμ (λ²‘ν° DB)
cd langconnect
make up # Docker Composeλ‘ PostgreSQL + API μλ² μμ
cd ..
2. μ 체 μμ€ν μ€ν
# μ€ν κΆν λΆμ¬ (μ΅μ΄ 1ν)
chmod +x scripts/run_all_agents.sh
# μμ€ν
μμ
./scripts/run_all_agents.sh
3. LangGraph μμ΄μ νΈ κ°λ³ ν μ€νΈ
# LangGraph Studioλ‘ ν
μ€νΈ
./scripts/test_langgraph_agents.sh
4. μμ€ν μ¬μ©
# CLI λͺ¨λ
python a2a_client/unified_research_agent.py
# λλ API νΈμΆ
curl -X POST http://localhost:8000/research \
-H "Content-Type: application/json" \
-d '{"topic": "AI νΈλ λ λΆμ", "depth": "comprehensive"}'
5. κ°λ³ A2A μλ² μ€ν
# Planning Agent A2A Server
python agents/a2a_servers/planning_a2a_server.py
# Research Agent A2A Server
python agents/a2a_servers/research_a2a_server.py
# Report Writing Agent A2A Server
python agents/a2a_servers/report_writing_a2a_server.py
# MCP Server
python all-search-mcp/run_server.py --transport http
API μλν¬μΈνΈ
κ° μμ΄μ νΈλ A2A νλ‘ν μ½μ λ°λ₯΄λ μλν¬μΈνΈλ₯Ό μ 곡ν©λλ€:
- UnifiedResearch Agent: http://localhost:8000
- Research Agent A2A: http://localhost:8001
- Planning Agent A2A: http://localhost:8003
- Report Writing Agent A2A: http://localhost:8004
- MCP Server: http://localhost:8090/mcp
- LangConnect API: http://localhost:8080
- LangGraph Studio: http://localhost:8123
μ¬μ© μμ
Python SDK μ¬μ©
from a2a_client import UnifiedResearchAgent, ResearchRequest
# μμ΄μ νΈ μμ±
agent = UnifiedResearchAgent()
# μ°κ΅¬ μμ²
request = ResearchRequest(
topic="Python μΉ νλ μμν¬ λΉκ΅ λΆμ",
depth="comprehensive",
output_format="markdown"
)
# μ°κ΅¬ μν
result = await agent.conduct_research(request)
# κ²°κ³Ό μΆλ ₯
print(result.report)
REST API μ¬μ©
curl -X POST http://localhost:8000/research \
-H "Content-Type: application/json" \
-d '{
"topic": "Python μΉ νλ μμν¬ λΉκ΅ λΆμ",
"depth": "comprehensive"
}'
Google ADK Web UI
# ADK μΉ UI μμ
adk web a2a_client/
κ°λ° κ°μ΄λ
νλ‘μ νΈ κ΅¬μ‘°
ttimes_guide_coding/
βββ agents/ # LangGraph μμ΄μ νΈ
β βββ agent/ # μμ΄μ νΈ κ΅¬ν체
β βββ a2a_servers/ # A2A μλ² λνΌ
β βββ base/ # κΈ°λ³Έ ν΄λμ€
β βββ tools/ # MCP ν΄λΌμ΄μΈνΈ
βββ all-search-mcp/ # MCP μλ²
βββ a2a_client/ # Google ADK ν΄λΌμ΄μΈνΈ
βββ langconnect/ # λ²‘ν° DB API
βββ docker/ # Docker μ€μ
βββ scripts/ # μ€ν μ€ν¬λ¦½νΈ
βββ docs/ # μ°Έκ³ λ¬Έμ
μλ‘μ΄ LangGraph μμ΄μ νΈ μΆκ°νκΈ°
/agents/agent/λλ ν 리μ μ μμ΄μ νΈ νμΌ μμ±BaseAgentν΄λμ€ μμ λ° λ Έλ/μ£μ§ ꡬν/agents/graph_builders.pyμ κ·Έλν λΉλ ν¨μ μΆκ°langgraph.jsonμ μμ΄μ νΈ λ±λ‘- A2A μλ² λνΌ κ΅¬ν (optional)
MCP λꡬ μΆκ°νκΈ°
/all-search-mcp/server.pyμ μ λꡬ ν¨μ μΆκ°@mcp.tool()λ°μ½λ μ΄ν°λ‘ λꡬ μ μ- μμ΄μ νΈμμ MCP ν΄λΌμ΄μΈνΈλ₯Ό ν΅ν΄ μ¬μ©
MCP μλ² κ°λ° κ°μ΄λ
Describing your server
Once you've provided the documentation, clearly describe to Claude what kind of server you want to build. Be specific about:
- What resources your server will expose
- What tools it will provide
- Any prompts it should offer
- What external systems it needs to interact with
For example:
Build an MCP server that:
- Connects to my company's PostgreSQL database
- Exposes table schemas as resources
- Provides tools for running read-only SQL queries
- Includes prompts for common data analysis tasks
Working with Claude
When working with Claude on MCP servers:
- Start with the core functionality first, then iterate to add more features
- Ask Claude to explain any parts of the code you don't understand
- Request modifications or improvements as needed
- Have Claude help you test the server and handle edge cases
Claude can help implement all the key MCP features:
- Resource management and exposure
- Tool definitions and implementations
- Prompt templates and handlers
- Error handling and logging
- Connection and transport setup
Best practices
When building MCP servers with Claude:
- Break down complex servers into smaller pieces
- Test each component thoroughly before moving on
- Keep security in mind - validate inputs and limit access appropriately
- Document your code well for future maintenance
- Follow MCP protocol specifications carefully
Next steps
After Claude helps you build your server:
- Review the generated code carefully
- Test the server with the MCP Inspector tool
- Connect it to Claude.app or other MCP clients
- Iterate based on real usage and feedback
A2A κ°λ° κ°μ΄λ
Reference
LangChain
LangConnect-Client
LangGraph
LangChain-MCP-Adapter(client)
ModelContextProtocol
MCP-Python-SDK
FastMCP
FastMCP-llms.txt
A2A-SDK
A2A-Directory
