JobRunr, a popular scheduling library, has expanded its ecosystem with ClawRunr (formerly JavaClaw), an open-source Java AI agent designed for background tasks. ClawRunr runs directly on users’ hardware and brings together conversational AI, persistent task execution, Browser Automation, MCP tools, and integration with multiple channels. By leveraging JobRunr for scheduling, retries, and monitoring, it offers a complete solution for developers who need to automate complex workflows. Below we explore the most common questions about this new tool.

1. What is ClawRunr and how does it relate to JobRunr?

ClawRunr is an open-source Java AI agent intended for scheduled, recurring, and one-off background tasks. It was originally known as JavaClaw and is now part of the JobRunr ecosystem. While JobRunr itself is a library for managing background jobs (scheduling, retries, monitoring), ClawRunr extends that concept by adding an AI-powered conversational layer. It uses JobRunr under the hood to handle the heavy lifting of job orchestration, but adds features like natural language interaction, browser automation, and multi-channel support. This means developers can combine AI-driven decisions with reliable background execution without buying into a cloud platform.

2. What are the key features of ClawRunr?

ClawRunr bundles several capabilities into a single open-source agent:

• Conversational interaction – users can give instructions in plain English (or other languages) to define or control tasks.
• Persistent task execution – tasks survive restarts and are managed with JobRunr’s proven scheduling engine.
• MCP tools (Model Context Protocol) for connecting with external AI models and data sources.
• Browser Automation – ClawRunr can open web pages, fill forms, and scrape data just like a human user.
• Multi-channel support – it integrates with web, Telegram, and Discord, so users can send commands from their preferred platform.
• On-premises execution – runs on your own hardware, giving you full control over data and privacy.

3. How does ClawRunr handle task scheduling and execution?

ClawRunr relies on JobRunr for all scheduling and execution logic. This means tasks can be scheduled with cron expressions, delays, or one-shot triggers. JobRunr guarantees that jobs are persisted to a database and will be retried automatically if they fail (configurable number of retries). ClawRunr builds on top of that by allowing users to describe a task in natural language, which the AI agent translates into a concrete job definition. Once defined, JobRunr handles the rest: it puts the job in a queue, executes it at the right time, and logs the result. If a task requires browser automation or external API calls, ClawRunr orchestrates those steps as a chain of background jobs.

4. What communication channels does ClawRunr support?

ClawRunr currently supports three channels: web (a built-in UI), Telegram, and Discord. Each channel allows you to send messages to the AI agent – for example, instructing it to “check the weather every morning at 8 AM and send a summary to our team channel”. The agent processes the command, sets up the corresponding scheduled job, and confirms. Because everything runs on your own hardware, the messages and results stay within your infrastructure. This multi-channel design makes it easy to integrate ClawRunr into existing workflows: developers can use the web dashboard, while non-technical teams can use Telegram or Discord.

5. How does ClawRunr combine AI conversation with background tasks?

The core innovation of ClawRunr is its AI-powered conversational layer. Instead of writing code to schedule a job, you can simply type something like: “Run a price check on our competitor website every hour and alert me if the price drops below $50.” The agent uses an LLM (integrated via MCP tools) to interpret the request, break it down into actionable steps (e.g., scrape URL, compare price, trigger alert), and then create one or more persistent background jobs using JobRunr. The same conversational interface can be used to monitor, pause, or cancel existing tasks. This makes task management accessible even to users with little programming experience, while still giving full power to developers.

6. Why was ClawRunr made open-source?

JobRunr decided to open-source ClawRunr to encourage community contribution and transparency. Being open-source means anyone can inspect the code, contribute features, and customize the agent for their own needs. It also aligns with JobRunr’s philosophy: JobRunr itself is open-source, so extending that to ClawRunr creates a consistent, trustable ecosystem. Open-sourcing also allows the project to benefit from external ideas, faster bug fixes, and broader adoption. Users who need enterprise support can still get commercial licenses from JobRunr, but the core agent remains free to use and modify.