Skip to main content
Glama
Zhangeldi123

Product Agent MCP Server

by Zhangeldi123
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Hybrid

MCP + LangGraph Product Agent (Test Task)

This repo implements:

  • MCP server (FastMCP, stdio) with product tools

  • LangGraph agent that connects to the MCP server via stdio subprocess

  • FastAPI endpoint to chat with the agent

  • Dockerfile + docker-compose

  • 3+ tests

Project structure

.
├─ app/
│  ├─ api.py
│  ├─ agent/
│  │  ├─ graph.py
│  │  ├─ mcp_client.py
│  │  ├─ mock_llm.py
│  │  ├─ tools_custom.py
│  │  └─ types.py
│  └─ mcp_server/
│     ├─ products_server.py
│     └─ storage.py
├─ data/products.json
├─ tests/
│  └─ test_api.py
├─ Dockerfile
├─ docker-compose.yml
└─ requirements.txt

Related MCP server: e2e-mcp-test

Run with Docker Compose (recommended)

docker compose up --build

API будет доступен на:

  • http://localhost:8000/docs

  • endpoint: POST http://localhost:8000/api/v1/agent/query

Example request:

curl -X POST "http://localhost:8000/api/v1/agent/query" \
  -H "Content-Type: application/json" \
  -d '{"query":"Покажи все продукты в категории Электроника"}'

Run locally

python -m venv .venv
source .venv/bin/activate  # Windows: .venv\Scripts\activate
pip install -r requirements.txt

export PRODUCTS_DB_PATH=./data/products.json  # Windows: set PRODUCTS_DB_PATH=...
uvicorn app.api:app --reload

Tests

pytest -q

Notes

  • MCP server runs via stdio (python app/mcp_server/products_server.py) and is spawned by the FastMCP Client(...) inside the agent.

  • The agent uses a mock LLM (rule-based) that outputs a JSON plan, then executes the plan by calling MCP tools + custom tools.

This server cannot be deployed

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Maintenance

ActivityInactive
ResponsivenessNo issues

Resources

Related MCP Connectors

Related MCP Servers

  • Products API MCP Server

    Demontie
    A
    license
    B
    quality
    D
    maintenance
    An MCP server that retrieves product data from the DummyJSON API, supporting filtering by various parameters like ID, title, category, brand, price and rating.
    1
    2 npm
    ISC

  • e2e-mcp-test

    rrbanda
    F
    license
    Not graded
    quality
    C
    maintenance
    An end-to-end test MCP server built with FastMCP that exposes a REST API as a set of tools for AI agents. It enables LLMs to perform CRUD operations on an upstream API by mapping HTTP methods to MCP tools.
    -

  • FastMCP Webinar Demo Server

    rabia-s
    F
    license
    Not graded
    quality
    D
    maintenance
    An MCP server with six tools including web search, URL fetching, math calculation, and note management. Designed for a live-coding demo integrating FastMCP with LangGraph ReAct agents.
    8
    -

  • mcp-json-reader

    oovz
    A
    license
    Not graded
    quality
    A
    maintenance
    MCP server for reading, querying, and filtering local JSON files with extended JSONPath syntax, supporting sorting, aggregations, and complex conditions.
    15 npm
    2
    MIT