Eion Shared Memory Storage
Shared Memory Storage for Multi-Agent Systems
What is Eion Shared Memory Storage?
Eion Shared Memory Storage is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to shared memory storage for multi-agent systems
Shared Memory Storage for Multi-Agent Systems
This server falls under the Knowledge & Memory category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- Shared Memory Storage for Multi-Agent Systems
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx eionConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Eion Shared Memory Storage
Eion is a shared memory storage backend for multi-agent AI systems, providing persistent conversation history and knowledge graph storage that multiple agents can read from and write to simultaneously. It combines a PostgreSQL database with pgvector for semantic search, Neo4j for graph-based knowledge relationships, and an MCP server interface so any MCP-compatible agent can store and retrieve memories by session, agent, or user ID. This makes it well-suited for building distributed AI workflows where agents need to share context, coordinate tasks, or maintain long-term knowledge across sessions.
Prerequisites
- Docker and Docker Compose installed
- Go 1.20 or later (for building the server from source)
- Python 3.8 or later with pip (for the MCP bridge)
- PostgreSQL with the pgvector extension (handled via Docker Compose)
- An MCP-compatible client or framework
Clone the repository and start dependencies
Clone the Eion repository and use Docker Compose to start PostgreSQL with pgvector and Neo4j.
git clone https://github.com/eiondb/eion.git
cd eion
docker-compose up -dInitialize the database schema
Enable the pgvector extension and run the database setup script to create the required tables.
docker exec eion_postgres psql -U eion -d eion -c "CREATE EXTENSION IF NOT EXISTS vector;"
docker exec -i eion_postgres psql -U eion -d eion < database_setup.sqlInstall Python dependencies and build the Go server
Set up the Python virtual environment for the MCP bridge, then build the Go-based Eion server.
python3 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
go build -o eion-server ./cmd/eion-serverConfigure eion.yaml and start the server
Edit eion.yaml to set your PostgreSQL and Neo4j connection details, MCP port, and other settings. Then start the Eion server.
# Edit eion.yaml with your database credentials, then:
./eion-serverRegister your agent and configure MCP access
Before an agent can use the MCP tools, it must be registered with Eion. Use the REST API to register, then configure the MCP client with the EION_BASE_URL pointing to your running server.
curl -X POST http://localhost:8080/cluster/v1/agents \
-H "Content-Type: application/json" \
-d '{"name": "MyAgent", "capabilities": ["memory", "knowledge"]}'Eion Shared Memory Storage Examples
Client configuration
MCP client configuration pointing to a locally running Eion server. The EION_BASE_URL must match your server's host and port from eion.yaml.
{
"mcpServers": {
"eion": {
"command": "python",
"args": ["-m", "internal.mcp.server"],
"env": {
"EION_BASE_URL": "http://localhost:8080"
}
}
}
}Prompts to try
These prompts demonstrate storing and retrieving shared memory in a multi-agent context.
- "Store this conversation summary in shared memory for session abc123"
- "Retrieve all conversation history for user user_001 from the last 7 days"
- "Search shared knowledge for information about our Q3 product roadmap"
- "Save the following as a knowledge entry: [key findings from analysis]"
- "What does agent ResearchBot know about competitor pricing?"Troubleshooting Eion Shared Memory Storage
MCP tools fail with 'agent not registered' error
Every agent must be pre-registered via the REST API before it can use the MCP tools. Call POST /cluster/v1/agents with the agent's name and capabilities. The returned agent_id must be passed with subsequent MCP requests.
pgvector extension fails to create: 'extension not available'
The Docker Compose setup uses a pgvector-enabled PostgreSQL image. If you are using your own PostgreSQL, install the pgvector extension manually: 'apt install postgresql-15-pgvector' on Debian/Ubuntu, then run 'CREATE EXTENSION vector;' in your database.
Semantic search returns no results even after storing documents
Semantic search requires that documents were stored with embeddings. Check that the embedding service is running and that EION_BASE_URL in your config points to the correct server. Also verify the vector dimension configured in eion.yaml matches the embedding model dimension.
Frequently Asked Questions about Eion Shared Memory Storage
What is Eion Shared Memory Storage?
Eion Shared Memory Storage is a Model Context Protocol (MCP) server that shared memory storage for multi-agent systems It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Eion Shared Memory Storage?
Follow the installation instructions on the Eion Shared Memory Storage GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with Eion Shared Memory Storage?
Eion Shared Memory Storage works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Eion Shared Memory Storage free to use?
Yes, Eion Shared Memory Storage is open source and available under the AGPL-3.0 license. You can use it freely in both personal and commercial projects.
Eion Shared Memory Storage Alternatives — Similar Knowledge & Memory Servers
Looking for alternatives to Eion Shared Memory Storage? Here are other popular knowledge & memory servers you can use with Claude, Cursor, and VS Code.
MemPalace
★ 52.6kA local AI memory system that stores all conversations verbatim and organizes them into navigable structures. It provides 19 MCP tools for AI assistants to search and retrieve past decisions, debugging sessions, and architecture debates automatically
Kratos
★ 25.7k🏛️ Memory System for AI Coding Tools - Never explain your codebase again. MCP server with perfect project isolation, 95.8% context accuracy, and the Four Pillars Framework.
Context Mode
★ 15.4kAn MCP server that preserves LLM context by intercepting large data outputs and returning only concise summaries or relevant sections. It enables efficient sandboxed code execution, file processing, and documentation indexing across multiple programm
Memu
★ 13.7kMemory for 24/7 proactive agents like OpenClaw.
MemOS
★ 9.3kMemOS (Memory Operating System) is a memory management operating system designed for AI applications. Its goal is: to enable your AI system to have long-term memory like a human, not only remembering what users have said but also actively invoking, u
Everos
★ 5.4kBuild, evaluate, and integrate long-term memory for self-evolving agents.
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.
Set Up Eion Shared Memory Storage in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use Eion Shared Memory Storage?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.