Octopoda OS

v1.0.0Knowledge & Memorystable

The open-source memory operating system for AI agents. Persistent memory, semantic search, loop detection, agent messaging, crash recovery, and real-time observability.

agent-frameworkai-agentsai-memoryautogencrash-recovery
Share:
333
Stars
0
Downloads
0
Weekly
0/5

What is Octopoda OS?

Octopoda OS is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to open-source memory operating system for ai agents. persistent memory, semantic search, loop detection, agent messaging, crash recovery, and real-time observability.

The open-source memory operating system for AI agents. Persistent memory, semantic search, loop detection, agent messaging, crash recovery, and real-time observability.

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

Features

  • The open-source memory operating system for AI agents. Persi

Use Cases

Persistent agentic memory system
Semantic search and loop detection
Crash recovery and agent messaging
LicenseNOASSERTION
Languagepython
Versionv1.0.0
UpdatedMay 21, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

Manual Installation

npx octopoda-os

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 Octopoda OS

Octopoda OS is an open-source memory operating system for AI agents that provides persistent, cross-session memory, semantic search, inter-agent messaging, goal tracking, crash recovery via snapshots, and loop detection. It integrates with LangChain, CrewAI, AutoGen, and OpenAI Agents SDK, and exposes 28 MCP tools for direct use with Claude and other MCP-compatible clients. Agents can store structured facts, recall them semantically, share memory between agents in named spaces, and recover their state after unexpected failures.

Prerequisites

  • Python 3.10 or later (3.9+ for core features without MCP)
  • An Octopoda API key from octopodas.com (free tier available); or set OCTOPODA_LOCAL_MODE=1 for fully offline use
  • pip or uv for package installation
  • An MCP-compatible client such as Claude Desktop or Claude Code
1

Install Octopoda with MCP support

Install the octopoda package with the mcp extra to include the MCP server module. Python 3.10+ is required for the MCP extra.

pip install "octopoda[mcp]"
2

Get your API key

Sign up for a free account at octopodas.com to obtain an OCTOPODA_API_KEY (format: sk-octopoda-...). Alternatively, skip this step and set OCTOPODA_LOCAL_MODE=1 to run fully offline.

export OCTOPODA_API_KEY="sk-octopoda-YOUR_KEY_HERE"
3

Register with Claude Code

Add the Octopoda MCP server to Claude Code using the mcp add command. The server module is synrix_runtime.api.mcp_server.

claude mcp add octopoda -s user -e OCTOPODA_API_KEY=sk-octopoda-YOUR_KEY \
  -- python -m synrix_runtime.api.mcp_server
4

Or configure Claude Desktop

Add the server to claude_desktop_config.json for Claude Desktop. Use the python binary from your virtual environment for isolated dependencies.

{
  "mcpServers": {
    "octopoda": {
      "command": "python",
      "args": ["-m", "synrix_runtime.api.mcp_server"],
      "env": {
        "OCTOPODA_API_KEY": "sk-octopoda-YOUR_KEY"
      }
    }
  }
}
5

Optional: configure local embedding model

Install the ai extra for local semantic search without sending data to a cloud provider. The default embedding model is BAAI/bge-small-en-v1.5.

pip install "octopoda[ai]"
export OCTOPODA_EMBEDDING_MODEL="BAAI/bge-small-en-v1.5"

Octopoda OS Examples

Client configuration

Claude Desktop configuration for Octopoda OS. For fully local operation without cloud sync, replace the API key with OCTOPODA_LOCAL_MODE=1.

{
  "mcpServers": {
    "octopoda": {
      "command": "python",
      "args": ["-m", "synrix_runtime.api.mcp_server"],
      "env": {
        "OCTOPODA_API_KEY": "sk-octopoda-YOUR_KEY"
      }
    }
  }
}

Prompts to try

Example prompts that use Octopoda's 28 MCP tools for persistent memory, semantic search, goal tracking, and agent messaging.

- "Remember that my preferred database is PostgreSQL and I always use async Python"
- "What do you remember about my coding preferences?"
- "Search my memory for anything related to authentication decisions I've made"
- "Set a goal to migrate the API to GraphQL with milestones: schema design, resolver implementation, testing"
- "Take a memory snapshot called before-refactor so I can restore it if needed"
- "Share the API design decisions with agent-team-b in the shared knowledge space"

Troubleshooting Octopoda OS

synrix_runtime module not found when starting the MCP server

Ensure octopoda was installed with the mcp extra: `pip install "octopoda[mcp]"`. The synrix_runtime package is only included in the mcp and all extras. Verify with `python -c "import synrix_runtime"`.

Memory is not persisting between sessions

If running in cloud mode, verify OCTOPODA_API_KEY is valid and you are connected to octopodas.com. If in local mode (OCTOPODA_LOCAL_MODE=1), check that SYNRIX_DATA_DIR points to a writable directory that persists between restarts.

Semantic search returns irrelevant results

For best results install the ai extra to enable local embeddings: `pip install "octopoda[ai]"`. Without it, search falls back to keyword matching. Set OCTOPODA_EMBEDDING_MODEL to a model suited to your data domain.

Frequently Asked Questions about Octopoda OS

What is Octopoda OS?

Octopoda OS is a Model Context Protocol (MCP) server that open-source memory operating system for ai agents. persistent memory, semantic search, loop detection, agent messaging, crash recovery, and real-time observability. It connects AI assistants to external tools and data sources through a standardized interface.

How do I install Octopoda OS?

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

Which AI clients work with Octopoda OS?

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

Is Octopoda OS free to use?

Yes, Octopoda OS is open source and available under the NOASSERTION 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": { "octopoda-os": { "command": "npx", "args": ["-y", "octopoda-os"] } } }

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

Read the full setup guide →

Ready to use Octopoda OS?

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