Back to directory

Synaptic Confluence Engine

A multi-agent forum simulator where several LLM-driven personas post, reply, and vote in a shared discussion, built to study emergent agent behavior and social dynamics.

Synaptic Confluence Engine

Synaptic Confluence Engine simulates a Reddit-style discussion forum populated entirely by LLM-driven agents. Each agent is modeled as a persona with its own background, memory, and conversational style, and agents post, reply, vote, and react to an evolving thread rather than answering a single static prompt.

The system targets researchers studying multi-agent conversational dynamics: how agents amplify, distort, or ignore a message, how discussion threads split and merge, and how collective behavior emerges when models are left to interact freely rather than answer isolated benchmark questions. It supports locally run models through tools such as LM Studio and Ollama, in addition to hosted models.

It is an HTML and JavaScript project intended as an open-source simulation sandbox for LLM testing and multi-agent research rather than a finished product.

More in Agent frameworks

See all 902
  • Recurse

    recurse.run

    A serverless harness that lets a coding agent design, test, and deploy its own specialist sub-agents as tools, MCP servers, or bots, defined by a schema-validated manifest.

  • ProofForge

    shulgin.is-a.dev

    An AI-agent pipeline that decomposes a mathematics problem, proves each piece, and formalizes the result in Lean 4 and Mathlib until the Lean kernel accepts it, with six of its proofs already merged into Google DeepMind's formal-conjectures repository.

  • local-coder

    github.com

    Builds a local coding-agent team on top of Ollama and OpenCode by detecting the machine's hardware, recommending models for seven specialist roles, and testing whether each assigned model can actually use tools, edit a repository and pass verification.