Show HN: OpenHelm – OpenClaw but free (using your Claude Code subscription)

OpenHelm has been introduced as a new autonomous job scheduler and planner, designed to leverage existing Claude Code subscriptions for execution. This macOS-native tool transforms high-level objectives into AI-planned, self-correcting job queues that run entirely locally. It aims to provide persistent, automated task execution without incurring additional AI API costs beyond what users already pay for Claude Code, positioning itself as a "free" extension for existing subscribers.

Operating on macOS 12+ for both Apple Silicon and Intel, OpenHelm's core workflow involves users setting a plain English goal, which the AI then decomposes into a series of one-off or recurring jobs. Users can review and tweak these plans before execution, with jobs running in the background and failures automatically analyzed, adjusted, and retried. A key architectural choice is its fully local operation, ensuring data privacy and eliminating cloud dependencies, complemented by a memory system that compounds project context over time.

This development significantly impacts the OpenClaw ecosystem by providing a cost-effective, persistent orchestration layer directly integrated with Claude Code, a foundational OpenClaw offering. It enables developers to implement sophisticated agentic workflows, such as automated SEO audits or test coverage maintenance, without the overhead of managing separate AI billing. For multi-agent systems, OpenHelm's self-correcting execution and job planning capabilities could serve as a robust local foundation, fostering more complex autonomous applications within the OpenClaw framework and democratizing access to agentic patterns.

Developers already subscribed to Claude Code should pay close attention, as OpenHelm unlocks substantial automation potential for their existing investment without additional API expenses. Researchers will find its local-first, self-correcting architecture and compounding memory system particularly interesting for exploring new paradigms in autonomous agents. Operators can leverage Open