SAME
🏎️ 🏠 🍎 🪟 🐧 - Memory with provenance tracking — records where agent knowledge originated and de
What is SAME?
SAME is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to 🏎️ 🏠 🍎 🪟 🐧 - memory with provenance tracking — records where agent knowledge originated and de
🏎️ 🏠 🍎 🪟 🐧 - Memory with provenance tracking — records where agent knowledge originated and de
This server falls under the Knowledge & Memory category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- MCP protocol support
Use Cases
Maintainer
Works with
Installation
NPM
npx -y @sgx-labs/sameManual Installation
npx -y @sgx-labs/sameConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use SAME
SAME (Stateless Agent Memory Engine) is a persistent, provenance-aware memory system for AI coding agents, exposed as an MCP server. It stores decisions, notes, and session handoffs in a local SQLite-backed vault, records where each piece of knowledge originated, detects stale entries, and surfaces relevant context automatically during multi-turn workflows. SAME exposes 19 MCP tools covering search, knowledge retrieval, memory recording, graph queries, and vault maintenance, and works across Claude Code, Cursor, and any MCP-compatible client on macOS, Windows, and Linux.
Prerequisites
- Node.js 18 or later (for the npm install method) or the platform installer script
- An MCP-enabled client such as Claude Desktop, Claude Code, or Cursor
- Optional: Ollama with the nomic-embed-text model for local semantic embeddings (or an OpenAI API key for cloud embeddings)
- Optional: VAULT_PATH environment variable to set a custom notes directory
Install SAME
Install SAME using the method that fits your platform. The npm method works on all platforms; the shell script installers are faster on macOS and Linux.
# macOS/Linux (shell installer)
curl -fsSL https://statelessagent.com/install.sh | bash
# Windows (PowerShell)
irm https://statelessagent.com/install.ps1 | iex
# Cross-platform via npm
npm install -g @sgx-labs/sameInitialize a project vault
Run 'same init' in your project directory to create the .same directory with a config.toml file. This is where memory files and session handoffs are stored.
same initConfigure the embedding provider (optional but recommended)
Edit .same/config.toml to configure your embedding provider. Ollama with nomic-embed-text is the local-first option; OpenAI works if you have an API key. Without an embedding provider, SAME falls back to keyword search.
[vault]
path = "/path/to/notes"
handoff_dir = "sessions"
[embedding]
provider = "ollama"
model = "nomic-embed-text"
[memory]
max_token_budget = 800
max_results = 2Add SAME to your MCP client configuration
Configure your MCP client to launch SAME via npx. The VAULT_PATH environment variable is optional; SAME defaults to the .same directory in the current working directory.
{
"mcpServers": {
"same": {
"command": "npx",
"args": ["-y", "@sgx-labs/same"]
}
}
}Restart your MCP client and verify tools
Restart your client. SAME exposes 19 tools including save_note, search_notes, get_session_context, save_decision, create_handoff, mem_health, and more.
SAME Examples
Client configuration
claude_desktop_config.json entry for SAME. Set VAULT_PATH if you want the vault stored at a specific location other than the default.
{
"mcpServers": {
"same": {
"command": "npx",
"args": ["-y", "@sgx-labs/same"],
"env": {
"VAULT_PATH": "/Users/you/projects/my-project/.same"
}
}
}
}Prompts to try
Example prompts that use SAME's memory and search capabilities during a development session.
- "Save a note: we decided to use JWT with 15-minute expiry for auth tokens"
- "Search memory for our decision about the database schema"
- "Create a session handoff summarizing what we accomplished today"
- "What do you remember about how we handle error logging in this project?"
- "Run a memory health check and report on stale or conflicting entries"Troubleshooting SAME
Semantic search returns no results or only keyword matches
SAME falls back to keyword search when no embedding provider is configured. Install Ollama, pull the nomic-embed-text model ('ollama pull nomic-embed-text'), and set provider = "ollama" in .same/config.toml. Then run 'same reindex' to rebuild the index.
Memory entries are missing after a new session
SAME stores memory in the vault directory. Ensure VAULT_PATH in your MCP config points to the same directory across sessions. If running from different working directories, set VAULT_PATH to an absolute path in the MCP client env block.
npx @sgx-labs/same fails to install or run
Run 'npm install -g @sgx-labs/same' for a persistent global install, then change the MCP config command to 'same' with no npx wrapper. Ensure Node.js 18+ is installed and npm has write access to the global prefix.
Frequently Asked Questions about SAME
What is SAME?
SAME is a Model Context Protocol (MCP) server that 🏎️ 🏠 🍎 🪟 🐧 - memory with provenance tracking — records where agent knowledge originated and de It connects AI assistants to external tools and data sources through a standardized interface.
How do I install SAME?
Install via npm with the command: npx -y @sgx-labs/same. Then add the server configuration to your AI client's JSON config file (e.g., claude_desktop_config.json or .cursor/mcp.json).
Which AI clients work with SAME?
SAME works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is SAME free to use?
Yes, SAME is open source and available under the NOASSERTION license. You can use it freely in both personal and commercial projects.
SAME Alternatives — Similar Knowledge & Memory Servers
Looking for alternatives to SAME? 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 SAME 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 SAME?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.