<2-3 sentences summarising what the project addresses and solves, the target audience, and the benefit.> LiveAgent is a local-first AI agent desktop client that deeply integrates large language model reasoning with local system tools. It empowers users to have AI agents capable of operating file systems, executing commands, and managing tasks, extending AI functionality beyond chat. Targeting engineers, researchers, and power users, it offers both standalone desktop operation and remote browser-based access via an optional Gateway, providing a powerful, extensible AI assistant.

How It Works

The core architecture combines a Tauri-based desktop GUI (React/Rust) with a Go-based Gateway service. LiveAgent integrates LLM reasoning with local system tools for precise file system operations, Bash command execution, and process supervision. It supports multi-model access, bridges external tools via the MCP protocol, and loads Skills packages on demand. The Gateway enables remote browser control, while the desktop app functions fully standalone.

Quick Start & Requirements

• Installation: Download installers from GitHub Releases (DMG for macOS, EXE/MSI/portable ZIP for Windows, AppImage/DEB/RPM for Linux). Gateway deployment via Docker (docker pull ghcr.io/stack-cairn/liveagent-gateway:latest).
• Prerequisites: Windows requires WebView2 runtime; Linux requires WebKitGTK 4.1. Docker needed for Gateway.
• Links: GitHub Releases, Docs, FAQ.

Highlighted Details

• Multi-Model Support: Claude, Codex, Gemini protocols, plus custom Base URL support.
• Local Tool Execution: Full file-system access, Bash commands, and ManagedProcess supervision.
• Extensibility: MCP protocol bridging and on-demand Skills package loading via ClawHub.
• Persistent Memory & Automation: Markdown + SQLite FTS for knowledge management; scheduled background tasks.
• Remote Gateway: Browser-based control via WebUI.
• Context Management: Segment + Summary Checkpoints for conversations; disconnect recovery.

Maintenance & Community

Active development via GitHub Actions for automated builds/releases. Acknowledges community contributions and is sponsored by API relay providers PackyCode and Right Code. No specific community channels are listed.

Licensing & Compatibility

Released under the permissive MIT License, allowing broad compatibility with commercial and closed-source applications.

Limitations & Caveats

The README does not explicitly detail limitations, alpha status, or known bugs. The complexity of local system integration may present a learning curve for advanced configurations.