I make ComfyUI nodes, workflows and LoRAs for image and video work. I also build tools that let AI agents work with real systems.
- ComfyUI-AusBoss: the nodes I made for my own workflows. Rotate, crop and pad a picture by dragging handles on the node, trim a clip on a timeline, and outpaint or inpaint with your original pixels put back. Install it from ComfyUI-Manager or the Comfy Registry.
- Civitai: the workflows I build with those nodes, and the LoRAs I train for them.
- Hugging Face: Qwen Image 2.1 LoRAs for outpainting, consistent edits, outfit swaps and other experiments.
By day, I run IT for an auto-auction environment where software has to work across IBM i, Windows, Linux and vendor platforms. That shapes how I build: clear interfaces, structured output, dry runs for risky operations, and a check that the thing actually worked.
- bluebubbles-relay: gives an AI agent a phone through a safety-gated BlueBubbles CLI.
- agent-harness-resources: skills, agents, hooks and plugins for coding-agent harnesses (mostly stale).
I've been building with local models since 2023. A few old projects:
If something here saved you time, a star on the repo helps other people find it. There is also a Ko-fi page if you want to leave a tip.