ClawRunr/JavaClaw

JavaClaw, a Java version of OpenClaw built on Spring Boot, Spring AI and JobRunr

What it solves

JavaClaw is a personal AI assistant designed to run locally on your own hardware, providing a control plane for an AI agent that can manage tasks, execute system commands, and interact across various communication channels while maintaining data privacy.

How it works

It is built using Java 25 and Spring Boot, utilizing Spring AI to connect to LLM providers like OpenAI, Anthropic, or local instances of Ollama. The agent operates within a dedicated workspace/ directory where it stores its system prompt, environment context, and long-term memory. It uses JobRunr for background task scheduling (one-off, delayed, or recurring) and supports the Model Context Protocol (MCP) for connecting to external tool servers.

Who it’s for

Developers and Java enthusiasts who want a self-hosted, extensible AI agent that can interact with their local machine and third-party messaging apps.

Highlights

  • Multi-Channel Access: Interact with the agent via a built-in Chat UI, Telegram, or Discord.
  • Extensible Skills: Add new capabilities by dropping Markdown files into the workspace or loading packaged JAR files.
  • Local Control: Ability to read/write files and run bash commands on the host machine.
  • Dynamic Tool Discovery: Uses a search-based pattern to surface only the most relevant tools to the LLM, preventing prompt bloat.
  • Task Management: Tracks tasks as human-readable Markdown files with states like todo, in_progress, and completed.

Related

  • Project
    openclaw/openclaw

    OpenClaw is a self-hosted personal AI assistant that connects AI models and tools to your favorite messaging channels and local devices.

  • Project
    ModalityDance/PalmClaw

    A native Android framework for deploying and running private AI agents directly on mobile phones, featuring integrated memory, tools, and multi-channel support.

  • Project
    swarmclawai/swarmclaw

    A self-hosted AI agent runtime and multi-agent framework for creating autonomous agent swarms with durable memory, scheduling, and multi-provider LLM support.

  • Project
    zeroclaw-labs/zeroclaw

    A self-hosted AI agent runtime written in Rust that connects various LLM providers to multiple communication channels and local system tools while prioritizing user data ownership.

  • Project
    fastclaw-ai/fastclaw

    A lightweight AI agent runtime written in Go that allows users to create, manage, and deploy multi-agent systems with isolated sandboxes and multi-channel support.