This desktop application addresses the challenge of navigating and searching Claude Code's conversation history, which is stored in scattered JSONL files. It targets users of Claude Code, providing a much-needed graphical interface for browsing, searching, and analyzing past interactions, thereby improving productivity and understanding of usage patterns.

How It Works

Built with Tauri, this application combines a Rust backend with a React frontend, utilizing Tailwind CSS and Radix components for the UI. It directly reads and parses JSONL files from the ~/.claude/projects/ directory, presenting conversations in a user-friendly format with syntax-highlighted code blocks and visualized tool outputs. The core advantage lies in its local-first approach, offering structured access and analytics without sending any data externally, and its ability to handle large conversation histories.

Quick Start & Requirements

Installation can be done by downloading pre-built releases or compiling from source. The recommended build process involves cloning the repository, running ./scripts/setup-build-env.sh, and then pnpm tauri:build:auto. Key requirements include Node.js 18+, pnpm, and a Rust toolchain. macOS users also need Xcode Command Line Tools, while Linux users require system libraries like WebKit and GTK (refer to LINUX_BUILD.md for specifics). The application loads conversation data into memory, which may impact performance with very large histories.

Highlighted Details

• Provides a file-tree-like navigation for projects and sessions.
• Features syntax highlighting for code blocks and readable formatting for message threads and tool outputs.
• Includes an Analytics Dashboard with activity heatmaps and tool usage statistics.
• Offers per-project token usage breakdowns and session-level analysis.
• Supports English, Korean, Japanese, and Chinese languages.
• Auto-refreshes to display new conversations as they are added.

Maintenance & Community

This project is noted as beta software and a side project, with contributions via pull requests welcomed. Users experiencing issues or having questions are encouraged to open an issue on the GitHub repository.

Licensing & Compatibility

The project is released under the MIT License, which is permissive and generally suitable for commercial use and integration into closed-source projects.

Limitations & Caveats

As beta software, users should expect potential instability or changes. Currently, only Linux and macOS are supported, with Windows support planned. Initial loading of very large conversation histories (thousands of messages) may be slow due to the in-memory loading strategy. The auto-update system is still under testing and may be unreliable.