context-engineering

Overview

This is a comprehensive MCP server implementing the WARNERCO Schematica application, which demonstrates production-grade semantic memory systems. It utilizes FastMCP, FastAPI, and LangGraph to implement all four CoALA memory tiers (Working, Episodic, Semantic, Procedural) in a 9-node pipeline. The server provides 28 MCP tools, 11 resources, and 5 prompts for building AI systems with persistent memory. It features progressive tool loading capabilities and episodic recall using Park et al.'s scoring formula.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

• warn_search_tools

  Keyword discovery of MCP tools with detail levels

  Try:Search for tools related to memory managementTry:Find tools with 'episodic' in the name
• warn_describe_tool

  Get full schema for a specific MCP tool by name

  Try:Describe the retrieve toolTry:Show me the details of log_episode tool
• parse_intent

  Parse user intent for the LangGraph pipeline

  Try:I need to find information about vector storesTry:Help me understand CoALA memory tiers
• query_graph

  Query the knowledge graph for relationships and entities

  Try:What entities are related to vector stores?Try:Show relationships between memory models
• inject_scratchpad

  Inject working memory into the current session

  Try:Add this observation to my working memoryTry:Record that I'm focusing on memory consolidation
• recall_episodes

  Recall episodic memories based on relevance and importance

  Try:What did we discuss about semantic memory last week?Try:Find important episodes about RAG systems
• retrieve

  Retrieve information from semantic memory vector store

  Try:Retrieve facts about long-term memory in AITry:Find information about context engineering
• compress_context

  Compress retrieved context for efficient processing

  Try:Compress this information about vector databasesTry:Summarize the key points about memory consolidation
• reason

  Reason over compressed context to generate insights

  Try:Reason about the trade-offs between different memory systemsTry:Analyze the impact of context length on AI performance
• respond

  Generate a response based on the processed context

  Try:Explain how CoALA memory tiers workTry:Summarize what we've learned about MCP tools
• log_episode

  Log the current interaction as an episodic memory

  Try:Log this conversation as an important episodeTry:Record this interaction with high importance
• warn_episodic_recall

  Recall episodic memories with detailed scoring breakdown

  Try:Show me the scoring breakdown for memory recallTry:Why did certain memories surface higher than others?

Note: Tool names inferred from the LangGraph pipeline description and CoALA memory tiers documentation. Descriptions are based on their roles in the pipeline and memory system. Signatures are not explicitly provided in the README.

Comparable tools

Installation

git clone https://github.com/timothywarner-org/context-engineering.git
cd context-engineering/labs/lab-01-hello-mcp/starter
npm install && npm start

cd src/warnerco/backend
uv sync
uv run uvicorn app.main:app --reload    # HTTP server
uv run warnerco-mcp                      # MCP stdio server

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

FAQ

What is the CoALA memory system?

CoALA (Context-Aware Long-term Architecture) is a four-tier memory system consisting of Working, Episodic, Semantic, and Procedural memory tiers implemented in this MCP server.

How do I test this MCP server?

You can test the server using the MCP Inspector with the command: `npx @modelcontextprotocol/inspector uv run warnerco-mcp`, which opens http://localhost:5173.

Compare context-engineering with