← All projects

Project 02

Multi-Agent Research Assistant

A LangGraph supervisor routing to specialist agents, exposed over MCP

LangGraphMCPMulti-agentAgentic workflowsHuman-in-the-loopContext engineering

A supervisor agent decomposes a research question and routes to specialists: a retriever that queries the knowledge base, a summariser, a fact-checker that re-grounds every claim, and a SQL analyst for structured questions. State, retries and an approval gate are explicit graph nodes rather than prompt instructions. The same tools are published as an MCP server, so Claude Desktop or any MCP client can query the knowledge base directly.

Try it

How it works

  1. 01
    Plan

    Supervisor node decomposes the question into a task list held in graph state.

  2. 02
    Route

    Conditional edges dispatch each task to the specialist that can answer it.

  3. 03
    Verify

    Fact-checker re-retrieves evidence for each claim; unsupported claims are dropped, not softened.

  4. 04
    Approve

    Interrupt node pauses the graph for human approval before any write action.

  5. 05
    Expose

    The same tool definitions served over MCP for external clients.

Infrastructure & running cost

ServiceRoleCost
Lambda (container)Graph execution, same image as project 1Free tier
DynamoDBGraph checkpointer — survives Lambda cold startsFree tier
Lambda Function URLStreaming SSE responses, no API Gateway needed$0

Closes the largest gap between the CV and current agent-engineering job specs.