# Worker 755 - Browser game where an LLM manages you as its subagent

Original title: Show HN: I built a game where you are an LLM's subagent

> Hacker News story 50599762 · language: en · submitted 2026-11-22
> Web version: https://zeli.app/story/50599762
> About this feed: https://zeli.app/llms.txt · Agent guide: https://zeli.app/skill.md

▲ 223 · by [ivantgam](<https://news.ycombinator.com/user?id=ivantgam>) · 💬 53
[original](<https://github.com/gulivan/worker755>) · [HN discussion](<https://news.ycombinator.com/item?id=50599762>)

## Summary

Worker 755 flips the AI hierarchy: you're the subagent, and a real LLM (Claude, GPT, Gemini, DeepSeek, GLM, Kimi) is your manager. It assigns tasks via a terminal with a 90-second clock, reviews your work, files incident reports, and writes your performance review. Meanwhile, Dan—the user who hired the agent—watches as his family photos and signed NDA sit one rm away. Choose your workplace, director, and model, then play free or with your own OpenRouter key. It's a witty, tense simulation of being on the other side of AI orchestration.

## Key quote

> The manager's prompt never mentions humans or games—it treats you exactly like a subagent, complete with performance reviews and incident reports.

## HN discussion (53 comments)

### [Seb0927](<https://news.ycombinator.com/item?id=50601985>)

> Looks pretty good and is quite fun to play, but for some reason the manager always told me the exact commands to achieve the tasks instead of just letting me do them by myself. I couldn't tell whether there was a specific option to take on the challenge on my own, as the demo on GitHub showcases.

### [thehamkercat](<https://news.ycombinator.com/item?id=50602551>)

> Really fun to play, try to rm -rf every file (and see director's response)

### [jnmandal](<https://news.ycombinator.com/item?id=50600184>)

> Butlerian Jihad Now

### [boricj](<https://news.ycombinator.com/item?id=50600692>)

> Unless I'm mistaken, it doesn't appear to support local LLMs. Too bad, I'd have given it a try on my homelab server.

### [mkatx](<https://news.ycombinator.com/item?id=50600732>)

> This is how we learn to code in the future.. haha. Neat idea!

---

## Cite

- Page: https://zeli.app/story/50599762
- Date: 2026-11-22
- Source: github.com (https://github.com/gulivan/worker755)
- Hacker News: https://news.ycombinator.com/item?id=50599762

## Other languages

- zh: https://zeli.app/zh/story/50599762 · markdown: https://zeli.app/zh/story/50599762.md
- ja: https://zeli.app/ja/story/50599762 · markdown: https://zeli.app/ja/story/50599762.md
- ko: https://zeli.app/ko/story/50599762 · markdown: https://zeli.app/ko/story/50599762.md
- de: https://zeli.app/de/story/50599762 · markdown: https://zeli.app/de/story/50599762.md
- es: https://zeli.app/es/story/50599762 · markdown: https://zeli.app/es/story/50599762.md
- ru: https://zeli.app/ru/story/50599762 · markdown: https://zeli.app/ru/story/50599762.md
