Context Engineering

v1.0.0Knowledge & Memorystable

🧠 Stop building AI that forgets. Master MCP (Model Context Protocol) with production-ready semantic memory, hybrid RAG, and the WARNERCO Schematica teaching app. FastMCP + LangGraph + Vector/Graph stores. Your AI assistant's long-term memory starts h

ai-agentsai-memoryanthropicazure-ai-searchchromadb
Share:
23
Stars
0
Downloads
0
Weekly
0/5

What is Context Engineering?

Context Engineering is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to 🧠 stop building ai that forgets. master mcp (model context protocol) with production-ready semantic memory, hybrid rag, and the warnerco schematica teaching app. fastmcp + langgraph + vector/graph st...

🧠 Stop building AI that forgets. Master MCP (Model Context Protocol) with production-ready semantic memory, hybrid RAG, and the WARNERCO Schematica teaching app. FastMCP + LangGraph + Vector/Graph stores. Your AI assistant's long-term memory starts h

This server falls under the Knowledge & Memory category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • 🧠 Stop building AI that forgets. Master MCP (Model Context P

Use Cases

Semantic memory and RAG
LangGraph and hybrid search
Long-term AI memory systems
LicenseMIT
Languagepython
Versionv1.0.0
UpdatedMay 3, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

Manual Installation

npx context-engineering

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use Context Engineering

The Context Engineering project is a comprehensive learning platform and production-ready toolkit for building AI systems with persistent, semantic memory using the Model Context Protocol. It combines FastMCP, LangGraph, and vector/graph stores (ChromaDB, Azure AI Search) to implement episodic memory with recency-importance-relevance scoring, and ships with the WARNERCO Schematica teaching application — exposing 28 MCP tools, 11 resources, and 5 prompts across a knowledge graph of 117 entities. Developers use it both as a reference implementation for hybrid RAG architectures and as a launchpad for building agents that remember context across sessions.

Prerequisites

  • Python 3.13 (pinned — check `.python-version`) and the `uv` package manager (`pip install uv` or via astral.sh)
  • Node.js 20+ for the lab exercises and MCP inspector
  • Claude Desktop or Claude Code as your MCP client
  • Optional: Azure AI Search endpoint and key for cloud-based vector search; ChromaDB is used locally by default
1

Clone the repository

Clone the context-engineering repository — it contains multiple labs and the main WARNERCO Schematica application.

git clone https://github.com/timothywarner-org/context-engineering.git
cd context-engineering
2

Try Lab 01 — Hello MCP (Node.js quickstart)

The first lab is the fastest way to see an MCP server running. It uses Node.js and starts in under a minute.

cd labs/lab-01-hello-mcp/starter
npm install
npm start
3

Set up the WARNERCO Schematica backend (Python)

The main application uses the uv package manager. Sync dependencies and start both the FastAPI backend and the MCP server.

cd src/warnerco/backend
uv sync
uv run uvicorn app.main:app --reload
4

Start the WARNERCO MCP server

In a second terminal, start the MCP server process. This is the process your MCP client will connect to.

cd src/warnerco/backend
uv run warnerco-mcp
5

Add the server to your Claude Desktop configuration

Add the WARNERCO MCP server to your Claude Desktop config. The server runs via `uv run warnerco-mcp` in the backend directory.

{
  "mcpServers": {
    "warnerco": {
      "command": "uv",
      "args": ["run", "warnerco-mcp"],
      "cwd": "/absolute/path/to/context-engineering/src/warnerco/backend"
    }
  }
}
6

Inspect the server with the MCP Inspector

Use the official MCP inspector to browse all 28 tools, 11 resources, and 5 prompts the server exposes before using them in Claude.

npx @modelcontextprotocol/inspector uv run warnerco-mcp
# Opens at http://localhost:5173

Context Engineering Examples

Client configuration

Claude Desktop configuration for the WARNERCO Schematica MCP server using uv stdio transport.

{
  "mcpServers": {
    "warnerco": {
      "command": "uv",
      "args": ["run", "warnerco-mcp"],
      "cwd": "/path/to/context-engineering/src/warnerco/backend"
    }
  }
}

Prompts to try

Example prompts that exercise the knowledge graph, episodic memory, and semantic search capabilities.

- "What entities are in the WARNERCO knowledge graph and how are they related?"
- "Store this decision in episodic memory: we chose PostgreSQL over MongoDB for the user service"
- "Search my memories for anything related to the authentication module from last week"
- "What are the most important and recent memories for the current project?"
- "List all MCP tools available and describe what the warn_search_tools meta-tool does"

Troubleshooting Context Engineering

uv command not found when starting the MCP server

Install uv with `curl -LsSf https://astral.sh/uv/install.sh | sh` on Mac/Linux or `pip install uv`. Then run `uv sync` from `src/warnerco/backend/` before starting the server.

Python version mismatch — uv complains about Python 3.13

The project pins Python 3.13 in `.python-version`. Install it via pyenv (`pyenv install 3.13`) or `uv python install 3.13`, then re-run `uv sync`.

Claude Desktop shows the server as disconnected

Ensure the `cwd` in your config is an absolute path to `src/warnerco/backend/`. Relative paths are not supported. Also confirm the uvicorn backend is running before starting the MCP server process.

Frequently Asked Questions about Context Engineering

What is Context Engineering?

Context Engineering is a Model Context Protocol (MCP) server that 🧠 stop building ai that forgets. master mcp (model context protocol) with production-ready semantic memory, hybrid rag, and the warnerco schematica teaching app. fastmcp + langgraph + vector/graph stores. your ai assistant's long-term memory starts h It connects AI assistants to external tools and data sources through a standardized interface.

How do I install Context Engineering?

Follow the installation instructions on the Context Engineering GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.

Which AI clients work with Context Engineering?

Context Engineering works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is Context Engineering free to use?

Yes, Context Engineering is open source and available under the MIT license. You can use it freely in both personal and commercial projects.

Browse More Knowledge & Memory MCP Servers

Explore all knowledge & memory servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Quick Config Preview

{ "mcpServers": { "context-engineering": { "command": "npx", "args": ["-y", "context-engineering"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use Context Engineering?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides