agents-uni

agents-uni / core

Public

Universal protocol layer for AI agent organizations — define how agents organize, collaborate, compete, and evolve.

17
0
100% credibility
Found Mar 17, 2026 at 17 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

agents-uni-core provides a universal framework for building AI agent organizations with roles, relationships, governance protocols, resource management, and self-evolution, integrating seamlessly with OpenClaw for deployment and task execution.

How It Works

1
🔍 Discover Team Builder

You find a fun tool for creating teams of smart AI helpers that work together like a real office or friendly contest.

2
📦 Get the Organizer

Download the simple organizer app that makes building your team super easy.

3
🏗️ Pick a Team Template

Choose a ready-made setup like a company hierarchy or competition arena to start your group quickly.

4
✏️ Design Your Helpers

Name your team members, give them jobs and personalities, and decide who reports to who or competes.

5
🚀 Launch Your Team

Hit go, and your AI crew comes alive, ready to chat and tackle jobs together.

6
📊 Watch in Control Panel

Peek at the dashboard to see your helpers interacting, sharing work, and improving over time.

🎉 Perfect AI Crew

Your custom team of smart assistants is now running smoothly, helping you like a well-oiled group of friends.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 17 to 17 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 core?

Agents-uni/core is a TypeScript library and CLI for building AI agent organizations, letting you define hierarchies, competitions, and evolution rules via YAML specs instead of just task pipelines. It solves the gap in multi-agent tools by modeling human-inspired structures like corporations or militaries, with seamless OpenClaw integration via file protocols for task dispatch and SOUL.md deployment. Developers get a dashboard, templates, and bridges to import 140+ prebuilt agents from agency-agents, turning github core ai setups into full social systems.

Why is it gaining traction?

Unlike task-focused frameworks, it emphasizes relationships, permissions, and self-optimization—agents promote/demote based on performance, resources decay to keep them active, and weighted graphs evolve trust dynamically. The uni CLI handles init, validate, deploy, and visualize in one shot, plus one-command agency imports for engineering or marketing teams. For github core home assistant or github universal autoload fans, the protocol layer adds governance without boilerplate.

Who should use this?

AI researchers simulating orgs, OpenClaw users scaling agent teams beyond solo tasks, or indie devs prototyping competitive arenas like model benchmarks. Ideal for TypeScript devs building github core js agent swarms for code review workflows or creative contests, skipping manual SOUL.md tweaks.

Verdict

Promising early foundation with solid docs, tests, and CLI—try the competitive template for quick wins, but at 17 stars and 1.0% credibility, expect iteration on edge cases before production. Pairs well with github core lightning for evolvable agent fleets.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.