Owl-Listener

Owl 1 — AI skill for Claude Code

AI community

OWL-1 — a workstation for directing Designpowers AI design agents.

How to install Owl 1

This entry records only its repository, not the path inside it, so there is no exact command to give. Open Owl-Listener/OWL-1 and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Owl 1 does

OWL-1 — a workstation for directing Designpowers AI design agents. Describe what you want, watch the team work, approve each handoff, steer along the way

Alternatives in AI

AIHawk — Open-source AI browser agent: describe any task in plain language and it autonomously browses, clicks, types a 30.3k ★
Context Engineering Intro — GitHub Repo stars Context engineering - the new way of AI coding assistants Context engineering guide 12.8k ★
Skill Seekers Roadmap — Transform Skill Seekers into the easiest way to create Claude AI skills from any knowledge source - documentat 11.1k ★

README

OWL-1

An **experimental** prototype exploring how directing a team of AI design agents could feel. Built by MC Dean.

⚠️ **Experimental personal project.** This is a research prototype / design exploration — not a product, not a service, and not affiliated with, endorsed by, or representing any employer. It's provided as-is, with no support or warranty. It explores ideas about human-in-the-loop agent orchestration; it is not a commercial or competitive offering.

![OWL-1 directing the Designpowers team — the Design Strategist track paused for your approval](docs/images/owl-1-arrangement.png)

OWL-1 borrows from the language of digital audio workstations: your design agents are tracks, the pipeline is a transport, and the whole system runs on a shared clock. The result is a UI where you direct a team of ten [Designpowers](https://github.com/Owl-Listener/designpowers) agents — watch them work, intervene when they need you, and stay oriented in a complex creative process. It's an experiment in *how agent orchestration should feel*, not a finished tool.

Run it

OWL-1 drives a real team of 10 Designpowers agents (vendored in `vendor/designpowers/`) through the Claude Agent SDK. Describe what you want, watch the team work in real time, approve each handoff, and steer any agent along the way.

npm install
export ANTHROPIC_API_KEY=sk-ant-...
npm start                    # open http://localhost:4318/?source=live

**→ Full walkthrough: [QUICKSTART.md](QUICKSTART.md)** (prerequisites, how to direct, troubleshooting).

Want to look around without a key (or any spend)? `npm run demo` runs the same UI on a scripted offline mock — same lanes, babble, and approval gates, no agents, no cost.

How it works

OWL-1 (front end) and Designpowers (agents) talk over the **OWL Agent Protocol** ([`docs/owl-agent-protocol.md`](docs/owl-agent-protocol.md)). The backend runs Designpowers headless via the Claude Agent SDK and translates the live run into OAP events