Home › AI Agents & LLM Tools › Republic of Agents
Republic of Agents
Benchmark for Social Reasoning in LLMs
| Built with | Codex |
|---|---|
| Category | AI Agents & LLM Tools |
| Maker | kkonstantin |
| Launched | 2026-03-09 |
| Source | Hacker News Show HN · 1 points |
| Link checked | 2026-11-02 |
How we know it was built with AI
- Codex: “Built with Codex”
Quoted from the source post. Category is assigned automatically.
From the maker
This is what I've been building over the past couple of weeks. It runs frontier reasoning models through batches of independent 7-player Mafia sessions (aka Werewolf). In one setup, sessions are fully independent; in the other, models reflect on a finished batch and carry the lessons into the next one.
From the website
Republic of Agents is an experimental benchmark evaluating social capabilities of reasoning models and emergent dynamics: collaboration, deception, coalition building. It runs batches of full 7-player Mafia games between frontier models and publishes transcripts, leaderboards, methodology, and batch
Similar apps
nomorehappypath.com
AI agent needs process to build a commercial application
Bailout
The coding agent meant to be deleted
gallery.samsar.one
I built a YouTube for generative videos in 50 prompts
delegatus
Orchestrate parallel Claude Code and Codex agents on a live map
Atria
Terminal UI for managing multiple coding agents