LANDSLIDE ⛰️

Ton code est un champ de bataille

LANDSLIDE is an open community for people who like building strange, ambitious things — especially around multi-agent systems and human-machine collaboration. Visual work is welcome too: generative art, interactive experiments, anything that looks as good as it runs.

If you have a crazy idea, or just want to help build someone else's, this is the place.

---

🚪 Who can join?

Anyone. No experience requirements, no interviews, no gatekeeping.

This community is fully open — newcomers, students, hobbyists, researchers, artists, everyone is welcome. You don't need a project of your own to join; you can start by helping with an existing one.

✋ How to join

It's one step:

1. Open an issue in this repo with the title:

   Join request: <your GitHub username>
   

2. (Optional but fun) tell us what you want to build or explore.
3. A maintainer will invite you. That's it 🎉

Your membership is public by default; you can set it to private anytime in your org settings.

🧪 What we do

• Agent infrastructure & governance — signing, registries, approval planes, knowledge that doesn't rot
• Human-machine collaboration tooling and experiments
• Visual / generative works — art that is code, code that is art
• Whatever weird thing you bring to the table

🗿 Projects

Project	What it is
Sphinx · SphinxGo	A human-in-the-loop approval control plane for agents — every dangerous action stops at a human (Python original + Go rewrite)
Tiltmeter · TiltmeterGo	Sign, register, govern and debug agent skills — provenance from author to deployment, in Python and Go
saltglass	Distill agent sessions into human-approved, signed, sovereign knowledge crystals that expire unless reverified
MONOLITH	A digital landslide laboratory — watch a git repository slowly collapse as a geologic body. Observe, don't fix.
yieldpoint · npm	Dynamic power handover for human-machine collaboration — autonomy as a first-class runtime policy (TypeScript)
arena · npm	Framework-agnostic multi-agent simulation & evaluation — same scenario suite, many strategies, comparable quality / safety / autonomy / cost metrics (TypeScript)
SlimeMold	A deterministic agent-based testbed for human + AI organization design — specify roles, reporting lines, institutions and task flow, then simulate how the org performs, coordinates, decides and learns (Python)

🌱 Ground rules

• Be kind. Be curious. Give credit where it's due.
• Everything in this org is open source — that's the point.

🚀 Getting started

1. Join via the issue above.
2. Introduce yourself in an issue or discussion.
3. Find something that sparks you — or open an issue with your own idea.
4. Build. Break. Learn. Repeat.

Let's make something strange.