Zikkaron
Biologically-inspired persistent memory engine for Claude Code. 26 cognitive subsystems, Hopfield networks, predictive coding, causal discovery, successor representations, all running locally over SQLite.
What is Zikkaron?
Zikkaron is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to biologically-inspired persistent memory engine for claude code. 26 cognitive subsystems, hopfield networks, predictive coding, causal discovery, successor representations, all running locally over sql...
Biologically-inspired persistent memory engine for Claude Code. 26 cognitive subsystems, Hopfield networks, predictive coding, causal discovery, successor representations, all running locally over SQLite.
This server falls under the Knowledge & Memory category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- Biologically-inspired persistent memory engine for Claude Co
Use Cases
Maintainer
Works with
Installation
PIP
pip install zikkaronManual Installation
pip install zikkaronConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Zikkaron
Zikkaron is a biologically-inspired persistent memory engine for Claude Code that implements 26 cognitive subsystems including Hopfield networks, predictive coding, causal discovery, and successor representations—all running locally over SQLite. It gives AI agents human-like memory consolidation, decay, and recall across sessions without relying on any external API or cloud service. Developers use it to build agents that remember project context, learn from past interactions, and reason causally about what they have done before.
Prerequisites
- Python 3.10 or later installed
- pip or uv package manager
- Claude Code or another MCP-compatible client
- Local disk space for the SQLite database (stored in ~/.zikkaron/memory.db by default)
Install Zikkaron via pip
Install the package from PyPI. This also installs the sentence-transformers library which downloads the all-MiniLM-L6-v2 embedding model on first run.
pip install zikkaronVerify the installation
Confirm the zikkaron command is available on your PATH before adding it to your MCP config.
zikkaron --helpOpen your Claude Code MCP configuration
Claude Code reads MCP servers from its settings file. Open ~/.claude/settings.json or the project-level .mcp.json file in your editor.
Add the Zikkaron server block
The minimal config requires no environment variables. Optional ZIKKARON_ prefixed variables let you tune memory decay, embedding model, and database location.
{
"mcpServers": {
"zikkaron": {
"command": "zikkaron"
}
}
}Optionally customise with environment variables
Use ZIKKARON_DB_PATH to store the database inside your project, ZIKKARON_DECAY_FACTOR to control how fast memories fade, and ZIKKARON_EMBEDDING_MODEL to switch the local sentence-transformer model.
{
"mcpServers": {
"zikkaron": {
"command": "zikkaron",
"env": {
"ZIKKARON_DB_PATH": "/home/user/projects/myapp/.zikkaron/memory.db",
"ZIKKARON_DECAY_FACTOR": "0.95",
"ZIKKARON_EMBEDDING_MODEL": "all-MiniLM-L6-v2"
}
}
}
}Restart Claude Code and seed your project
After restarting, ask Claude Code to call the seed_project tool to initialise project-specific memory context. From then on, Zikkaron will record and recall memories automatically across sessions.
Zikkaron Examples
Client configuration
Minimal claude_desktop_config.json entry for Zikkaron with optional database path customisation.
{
"mcpServers": {
"zikkaron": {
"command": "zikkaron",
"env": {
"ZIKKARON_DB_PATH": "~/.zikkaron/memory.db",
"ZIKKARON_DECAY_FACTOR": "0.95",
"ZIKKARON_COGNITIVE_LOAD_LIMIT": "4"
}
}
}
}Prompts to try
Example requests that exercise Zikkaron's memory, recall, causal, and project-context tools.
- "Remember that we decided to use PostgreSQL instead of MySQL for this project"
- "Recall what we discussed about the authentication flow last week"
- "Show me my memory stats and how many memories are currently stored"
- "Get the full project story and summarise the key decisions made so far"
- "Detect any gaps in my knowledge about the payment module"
- "Create a checkpoint of my current memory state before the refactor"Troubleshooting Zikkaron
zikkaron command not found after pip install
The script may not be on your PATH if pip installed it into a user or virtual environment bin directory. Run python -m zikkaron or activate your virtual environment, then retry.
Slow startup or hang while downloading the embedding model
On first run, sentence-transformers downloads all-MiniLM-L6-v2 (~90 MB) from Hugging Face. Ensure internet access and wait for the download to complete. Subsequent starts use the cached model.
Memories not persisting between sessions
Verify ZIKKARON_DB_PATH points to a writable directory. If you run Claude Code under different user contexts (e.g. sudo), the database path may differ. Use an absolute path and confirm file permissions.
Frequently Asked Questions about Zikkaron
What is Zikkaron?
Zikkaron is a Model Context Protocol (MCP) server that biologically-inspired persistent memory engine for claude code. 26 cognitive subsystems, hopfield networks, predictive coding, causal discovery, successor representations, all running locally over sqlite. It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Zikkaron?
Install via pip with: pip install zikkaron. Then configure your AI client to connect to this MCP server.
Which AI clients work with Zikkaron?
Zikkaron works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Zikkaron free to use?
Yes, Zikkaron is open source and available under the MIT license. You can use it freely in both personal and commercial projects.
Zikkaron Alternatives — Similar Knowledge & Memory Servers
Looking for alternatives to Zikkaron? 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 Zikkaron 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 Zikkaron?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.