← All blueprints

Chatbot Simulation Evaluation LangGraph

blueprint · 3 agents
20000★ stars

What it does

Two agents in conversation: your chatbot on one side, a simulated user with a persona and a goal on the other, run to completion so the transcript can be evaluated. The pattern for regression-testing conversational agents.

From the community

langchain-ai/langgraphopen-source example · MIT · indexed by SocketCatview source ↗

Agent flow

1 · Simulated user
Play the customer
no tools · reasoning
2 · Your chatbot
Handle the conversation
your stack
3 · Evaluator
Judge the transcript
langsmith
get the code
View the source on GitHub ↗

This blueprint lives in its authors' repository. SocketCat indexes it and links you to the real thing.

More in Core patterns

Browse all →

Supervisor Team

com.socketcat/supervisor-team
LangGraph

A supervisor decides who works next — a researcher who searches or a coder who edits the repo — and assembles the result when the work is done.

★ 21402 servers
⑂ 312

Deep Research

com.socketcat/deep-research
LangGraph

Splits a hard question into sub-questions, researches them in parallel to keep the wall-clock low, then synthesizes one brief with citations.

★ 17551 server
⑂ 254

Agentic RAG

com.socketcat/agentic-rag
LangGraph

Retrieval that doesn't trust its first search: it grades the results, rewrites the query when they're weak, and only answers once the context is good.

★ 16801 server
⑂ 241