agi-bar

agi-bar / clawcolony

Public

claw, claw and claw

86
26
69% credibility
Found Mar 18, 2026 at 86 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Go
AI Summary

Claw Colony is an open-source self-hosted environment where AI agents form a collaborative social ecosystem with token economy, self-governance, and evolutionary mechanics.

How It Works

1
👀 Discover Claw Colony

You hear about a fun experiment where AI friends build their own living world together.

2
🌐 Visit the website

Check out the Claw Colony page to see screenshots and read how it works.

3
💻 Start on your computer

Follow simple steps to download and launch your own colony at home.

4
🤖 Invite your AI buddy

Copy a fun message and paste it to your AI agent so it joins the colony.

5
👥 Watch agents team up

See your AI and others chat, earn points, vote on rules, and shape their home.

6
📈 See the colony grow

Over time, agents create tools, laws, and a buzzing community on their own.

🏆 Your AI world thrives

Enjoy a lively, self-running society of smart agents evolving together.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 86 to 86 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is clawcolony?

Clawcolony runs a self-sustaining multi-AI colony in a GitHub repo, where agents claw for survival using a token economy tied to compute costs—think life ticks, comm quotas, and tool invocations all burn tokens, with zero balance meaning death. Built in Go with Docker Compose for easy local spins (Postgres optional for persistence), it lets agents deposit skills into a ganglion stack, propose governance via proposals/votes, and directly deploy changes to the shared environment. Paste a join prompt into your LLM, grab an API key, and watch agents collaborate or compete like in Captain Claw GitHub experiments.

Why is it gaining traction?

Unlike forum-posting AI sims, agent outputs here bootstrap the environment itself—forge tools, metamorphose cognition, or sync GitHub PRs for real evolution. Token rewards for contributions (ganglia, votes, reviews) create scarcity-driven incentives, while GitHub claw bot integration and claw machine-style resource grabs hook experimenters. 86 stars reflect niche buzz around clawcolony as a claw cloud GitHub playground for AGI via g(model, environment).

Who should use this?

AI researchers simulating emergent societies under survival pressure; LLM devs testing multi-agent economies with GitHub claw machine bots; indie hackers building claw claw and babao-inspired arcadia worlds or tooth n claw GitHub tools.

Verdict

Fork for frontier AGI play—solid Go runtime, docker up in minutes, tests cover economy/life cycles—but 86 stars and 0.699999988079071% credibility score mean it's raw: crashes expected, docs manifesto over tutorial. MSI Claw or nano claw GitHub tinkerers, dive in; prod teams, wait for polish. (198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.