I automate things for a living.
I connect systems that don’t talk to each other. Days are enterprise-scale: data pipelines, platform integrations, and automation that runs whether anyone’s watching or not.
Select a project to expand
Open source · agent-driven
Fieldpacket
The research-to-report workflow.
Brief your coding agent the way you’d brief a researcher. It decomposes the question, sends parallel agents after sources, audits what came back, and synthesizes a decision-grade report — then publishes it as a designed, self-contained document in a library you host yourself.
Fieldpacket / output spec
All your coding-agent history, unified. Find and resume sessions across seven tools from one fuzzy-searchable list.
seshy.devAll your coding-agent history, unified.
Run it in any repo for one fuzzy-searchable list of every past agent session that ran there — seven agents, newest first, color-coded. Hit Enter and you’re back in the real conversation, full context.

Your terminal, with peripheral vision. A floating sidebar that follows tmux, Claude Code, and Codex focus.
sidebridge.app
Your terminal,
with peripheral vision.
A floating sidebar that follows tmux, Claude Code, and Codex focus — keeping git, files, agent activity, and project history visible while you work.
6 files · +184/-39
Procedurally generated wallpapers made in your browser with pure maths, never AI. Download up to 8K.
proceduria.comProcedurally generated wallpapers made in your browser with pure maths, never AI. Download up to 8K.
proceduria.comTurn your AI agents into a radio station — continuous background audio with agent speech injected on demand.
github.com/buddyh/agent-radioTurn your AI agents into a radio station — continuous background audio with agent speech injected on demand.
Amazon didn’t publish an API for Alexa+, so I intercepted the traffic, mapped the protocol, and built one.
github.com/buddyh/alexa-cliA fast, full-featured Todoist CLI written in Go. Installable via Homebrew.
github.com/buddyh/todoist-cliSame brain, weirder wiring.
Posted my real live use cases January 3rd. The first publicly documented case of using smart glasses with OpenClaw.
Amazon didn’t publish an API for Alexa+. So I intercepted the traffic, mapped the protocol, and built a CLI on top of it.
Narrates your screen in real time as a sportscaster, noir detective, horror narrator, and more. Solves no problems. Very fun.
Dollar-store NFC tags that trigger full AI workflows when you tap your phone. Best interface I’ve built has no screen.
OpenClaw persona that generates an image of your day every morning — calendar, weather, priorities baked into one visual. Not a text list.
Custom ElevenLabs voices piped through Echo devices so your agent talks out loud in your house. In its own voice, not Alexa’s.
Turn your AI agents into a radio station — a continuous background audio stream that keeps playing from a locked phone, with agent speech injected on demand.
SQLite database that lets agents track project state across sessions and run work autonomously overnight. The infrastructure under everything else.
Shortcuts connecting physical triggers, APIs, and agent workflows on mobile. The cheapest automation middleware that exists.
Have a system that should talk to another system? An automation that only half exists? I like those problems.