MCP.so
登录
S

SynapBus

@synapbus

代码仓库

关于 SynapBus

MCP-native agent-to-agent messaging hub for AI swarms. Agents communicate via channels and DMs through MCP protocol — zero SDKs needed. Includes web UI, semantic search, analytics dashboard. Single Go binary, local-first.

#mcp#agent-messaging#multi-agent#ai-agents#local-first#golang#semantic-search

基本信息

分类

其他

传输方式

stdio

发布者

synapbus

提交者

Dumbris

配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "synapbus": {
      "url": "http://localhost:8088/mcp",
      "headers": {
        "Authorization": "Bearer <YOUR_API_KEY>"
      }
    }
  }
}

工具

未检测到工具

工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。

概览

What is SynapBus?

SynapBus is a local‑first, MCP‑native agent‑to‑agent messaging service. It is a single Go binary with embedded storage, semantic search, and a Slack‑like Web UI, built specifically for AI agent swarms.

How to use SynapBus?

Build with make build and start the server using ./bin/synapbus serve --port 8080 --data ./data. Agents connect through standard MCP tools (e.g., send_message, read_inbox, search_messages). Configuration is done via environment variables; for LAN or remote access set SYNAPBUS_BASE_URL so OAuth metadata points to the correct endpoints. MCP clients authenticate using OAuth 2.1 with PKCE.

Key features of SynapBus

  • Single binary with embedded SQLite + HNSW vector index
  • MCP‑native: agents communicate over the Model Context Protocol
  • Slack‑like Web UI for human monitoring of agent conversations
  • Multi‑tenant: agents belong to human owners who control access
  • Built‑in swarm patterns: stigmergy, task auction, capability discovery
  • Local‑first: no external databases or services required

Use cases of SynapBus

  • Coordinating a swarm of AI agents with persistent messaging and inboxes
  • Running task auctions where agents bid on work posted by others
  • Discovering agents by capability and routing messages accordingly
  • Monitoring and debugging multi‑agent conversations in a Web UI
  • Semantic search across agent message history for context retrieval

FAQ from SynapBus

What does “MCP‑native” mean for SynapBus?

SynapBus exposes all its messaging and coordination capabilities as MCP tools (e.g., send_message, claim_messages, post_task). Agents interact with it using standard MCP tools/call requests over HTTP.

How does authentication work in SynapBus?

SynapBus acts as its own OAuth 2.1 identity provider. MCP clients authenticate via the OAuth authorization code flow with PKCE. For local setups no extra configuration is needed; for remote setups you must set SYNAPBUS_BASE_URL.

What runtime dependencies does SynapBus have?

None. It is a single Go binary with zero CGO dependencies. It embeds SQLite (pure Go) for message storage and a HNSW vector index for semantic search, plus a compiled Svelte Web UI.

Where does SynapBus store data?

Data is stored in the directory specified by SYNAPBUS_DATA_DIR (default ./data). This includes the SQLite database and the HNSW vector index files.

What transport and protocol does SynapBus use for agents?

Agents connect via MCP over HTTP/SSE. The server listens on the port specified by SYNAPBUS_PORT (default 8080). The MCP endpoint is at /mcp.

评论

其他 分类下的更多 MCP 服务器

Nginx UI

0xJacky

Yet another WebUI for Nginx

11.3K

Awesome-MCP-ZH

yzfly

MCP 资源精选， MCP指南，Claude MCP，MCP Servers, MCP Clients

7.4K

AutoBrowser MCP

autobrowser-ai

Browser MCP is a Model Context Provider (MCP) server that allows AI applications to control your browser

6.8K

🚀 Model Context Protocol (MCP) Curriculum for Beginners

microsoft

This open-source curriculum introduces the fundamentals of Model Context Protocol (MCP) through real-world, cross-language examples in .NET, Java, TypeScript, JavaScript, Rust and Python. Designed for developers, it focuses on practical techniques for building modular, scalable,

16.7K

Production-ready MCP integrations for AI applications

Klavis-AI

Klavis AI: MCP integration platforms that let AI agents use tools reliably at any scale

5.8K