🔌 MCP Server available — Enhance Claude's memory with persistent storage by setting up the opencontext MCP server. This allows Claude to automatically save, recall, and search your context across conversations. Jump to setup →

opencontext

Save your context, your way

Import chat history from any AI platform · Manage context with MCP · Export to Claude, ChatGPT, or Gemini

Documentation • Quick Start • Usage • Contributing

📖 Overview

opencontext empowers users to manage and maintain a portable memory of AI interactions. With this tool, users can effortlessly transfer their entire conversation history while switching between AI assistants and enable Claude to use persistent memory through the MCP server.

• 🎯 Preferences - Tailored communication style analysis for Claude.
• 🧠 Memory - Relevant context derived from chat history.
• 💬 Conversations - All dialogues stored as easily readable markdown files.
• 🔌 MCP Server - Ensures continuity in conversations with Claude.

Why Use This?

Transitioning between AI assistants often results in the loss of vital context regarding previous interactions, communication styles, and personal nuances. opencontext addresses this challenge by:

1. Importing chat histories from platforms like ChatGPT (Gemini support is on the horizon).
2. Analyzing communication patterns with local AI (Ollama) to create memory and preference profiles.
3. Providing export formats compatible with Claude, ChatGPT, or Gemini.
4. Enabling Claude to save and recall context automatically using an MCP server.

Outcome: Claude retains knowledge about individual preferences and communication styles, tailoring each interaction based on historical context.

✨ Features

🚀 Quick Start

Prerequisites

• Node.js 25+ - Download from Node.js.
• Obtain ChatGPT Export using the guidelines in the README.
• Optional: Install Ollama for enhanced AI analysis capabilities.

Installation Steps

# Clone the repository
git clone https://github.com/adityak74/opencontext.git
cd opencontext

# Install CLI/MCP dependencies
npm install

# Build the project
npm run build

Option A: Run via Docker (Recommended)

The recommended approach is to utilize the official Docker image, which integrates the UI, server API, and MCP server into a single container, ensuring preferences and context are retained without relying on browser storage.

# Pull and run the Docker image (access UI at http://localhost:3000)
docker run -p 3000:3000 \
  -v opencontext-data:/root/.opencontext \
  adityakarnam/opencontext:latest

What gets stored in the volume (/root/.opencontext/):

Usage Example

npm start -- convert path/to/chatgpt-export.zip

This command processes the exported ChatGPT data, producing formatted files for Claude.

📚 Documentation

Getting Your ChatGPT Export: Steps for exporting data can be found in the README. Users can migrate preferences and context into Claude through simple copy-paste actions from the generated markdown files.

Privacy & Security

All processing activities are local, assuring users that no data is shared externally, fostering a secure environment for sensitive information.

🤝 Contributing

Participation and contributions are encouraged; options include reporting issues, suggesting features, improving documentation, or submitting code. See GitHub for guidelines and participation rules.

📊 Performance and Future Features

Performance is based on export size and analysis; users can anticipate planned features such as support for additional AI platforms and improved context management capabilities.