Vidurai Memory Manager for VS Code

🧠 Intelligent AI memory for your coding workflow

Vidurai automatically tracks your context—files, commands, errors—and makes it instantly available when you need to prompt Claude Code or other AI assistants. No manual context gathering. No token waste.

36.6% average token reduction. Privacy-first. Works offline.

Why Vidurai?

Traditional AI workflows:

• ❌ Manually copy-paste context
• ❌ Forget what you did yesterday
• ❌ Repeat information across prompts
• ❌ Waste tokens on irrelevant details

With Vidurai:

• ✅ Automatic context tracking
• ✅ Intelligent forgetting (keeps what matters)
• ✅ One-click context export
• ✅ 36.6% fewer tokens

Features

🧠 Smart Memory

• Tracks files, commands, errors automatically
• Salience-based importance (CRITICAL → NOISE)
• Gist extraction (meaning, not words)

🔒 Privacy-First

• Local-only by default (no cloud)
• Automatic secrets detection
• Your code never leaves your machine

⚡ Zero-Config

• Auto-detects Python
• Auto-installs Vidurai SDK
• Works offline

📊 Research-Backed

• Based on Fuzzy-Trace Theory
• Neuroscience-inspired architecture
• Proven 36.6% token reduction

Quick Start

1. Install extension from VS Code Marketplace
2. Reload VS Code
3. Click "Quick Setup" when prompted (auto-installs dependencies)
4. Done! Start coding.

Using Vidurai

When you need context for AI:

1. Press Ctrl+Shift+P (or Cmd+Shift+P on Mac)
2. Run "Vidurai: Copy Relevant Context"
3. Paste into Claude Code or any AI assistant
4. AI gets full context automatically

Commands

• Vidurai: Copy Relevant Context - Copy tracked memories to clipboard
• Vidurai: Restart Python Bridge - Restart the memory bridge
• Vidurai: Show Logs - View extension logs for debugging

Settings

• vidurai.enabled - Enable/disable context tracking (default: true)
• vidurai.debounceMs - File edit debounce time in ms (default: 2000)
• vidurai.pythonPath - Custom Python path (auto-detected if empty)
• vidurai.maxFileSize - Maximum file size to track in bytes (default: 51200)
• vidurai.trackTerminal - Track terminal commands (default: true)
• vidurai.trackDiagnostics - Track errors/warnings (default: true)
• vidurai.ignoredPaths - Paths to ignore (default: ["node_modules", ".git", "dist", "build"])
• vidurai.logLevel - Logging level (default: info)

Requirements

• VS Code 1.80.0 or higher
• Python 3.8 or higher

The extension will automatically detect Python and install the Vidurai SDK if needed.

Philosophy

विस्मृति भी विद्या है "Forgetting too is knowledge"

Vidurai isn't just a cache—it's an intelligent memory system that knows what to remember and what to forget, just like your brain does.

Based on the Three-Kosha Architecture from Vedantic philosophy, Vidurai manages memory in layers:

• Annamaya (Raw) - Verbatim content
• Pranamaya (Active) - Frequently accessed
• Manomaya (Wisdom) - Compressed gists

Privacy & Security

• ✅ All data stored locally in ~/.vidurai/sessions/
• ✅ Secrets automatically detected and excluded (API keys, passwords, tokens)
• ✅ No telemetry, no data sent to external servers
• ✅ Works completely offline

Troubleshooting

Extension won't activate

1. Check Python is installed: python --version or python3 --version
2. Install Vidurai SDK manually: pip install vidurai==1.6.1
3. Check logs: Run "Vidurai: Show Logs"

Bridge crashes

1. Run "Vidurai: Restart Python Bridge"
2. If it continues crashing, check logs for errors
3. Report bugs at GitHub Issues

"Python not found" error

• Install Python 3.8+ from python.org
• Or set custom path in settings: vidurai.pythonPath

Support

• 📖 Documentation
• 🐛 Report Bug
• 💬 Discord Community

License

MIT

Built with 🧠 by the Vidurai team