📡 @mdayan896/cc-radar

The Industrial Operator Console for your AI Agents.

@mdayan896/cc-radar is a lightweight, local-first control center for monitoring Claude Code, Codex, Cursor, and other Agentic tools. It transforms silent log files and local databases into a high-density, real-time telemetry dashboard.

⚡ Why cc-radar?

• 100% Local & Private: No data ever leaves your machine. It reads the logs your CLIs already write.
• Real-Time Telemetry: Uses Server-Sent Events (SSE) for instant dashboard updates without polling.
• Agent Agnostic: Supports Claude Code, Codex, Cursor, and OpenCode out of the box.
• Industrial Aesthetic: Designed like a mission control console for maximum focus and data density.
• Optimized Workflows: Built-in "Action Center" with copy-ready payloads to optimize your agent's context and cost.

🛠️ Features

• ✅ Live Session Monitoring: Track active sessions across multiple sources.
• 📊 Token Analytics: Detailed burn-rate and time-left estimates (when token data is available).
• 🚨 Error Tracking: Instant alerts for API failures, system errors, and redundant reads.
• ⚡ Burn-Rate Alerts: High-usage notifications to keep your budget in check.
• 🕰️ Session History: Persistent local storage powered by SQLite (with JSON fallback).
• 📉 Redundancy Detection: Highlighting when agents are stuck rereading the same files.

🚀 Quick Start

Ensure you have Node.js 18+ installed.

Zero-Install Workflow (Recommended)

Run the console immediately without cluttering your global space:

npx @mdayan896/cc-radar

This will automatically launch the Industrial Operator Console at http://localhost:7331.

Global Installation

If you use cc-radar frequently, install it globally to invoke it from any shell:

npm install -g @mdayan896/cc-radar
@mdayan896/cc-radar

🎮 Command Line Interface

npx @mdayan896/cc-radar \
  --root ~/.claude/projects \
  --history ~/.claude/history.jsonl \
  --db ~/.cc-radar/history.sqlite3 \
  --port 7331 \
  --no-open

🕵️ How it Works

1. Discovery: Automatically detects local telemetry locations for all supported agents.
2. Streaming: Watches for file changes and read only the newly appended data (minimal I/O).
3. Aggregation: Normalizes raw JSONL/SQLite records into unified session models.
4. Broadcast: Pushes live state updates to the browser-based Operator Console.

📂 Project Structure

• bin/cc-radar.js: CLI Entry point.
• src/radar.js: Core service and file-watcher orchestration.
• src/scanner.js: Incremental data ingestion engine.
• src/aggregator.js: Logic for session modeling and metric derivation.
• public/: Frontend assets (Vanilla CSS & JS).

🤝 Contributing

Contributions are welcome! Please check the Implementation Roadmap for planned features.

1. Clone the repo: git clone https://github.com/mdayan81/cc-radar.git
2. Install dependencies: npm install
3. Run tests: npm test
4. Launch dev mode: npm run dev

📜 License

Distributed under the MIT License. See LICENSE for more information.

💬 Social

Follow for updates and feature drops: @mdayan24X