agentconnect-md/agentconnect

The open-source, multi-agent alternative to Claude Tag. @ any agent, wherever work happens, they work alongside your team, learning as they go.

What it solves

AgentConnect provides a platform for teams to integrate multiple AI agents into their existing communication and workflow tools. It solves the problem of agents living as isolated personal tools in individual terminals, which prevents teammates from seeing agent progress, reviewing output, and sharing context. By turning the "glue" (message channels, credential handling, and context stitching) into a a unified platform, it allows people and agents to collaborate in shared conversations.

How it works

The system uses a distributed architecture consisting of a Control Plane, Daemons, and an optional Relay.

  • Control Plane & Web UI: Manages authentication, configuration, permissions, and organization-wide knowledge.
  • Daemons: Run the agents via ACP, manage workspaces and session states, and maintain direct connections to platforms like Slack or GitHub.
  • Relay: Handles callback-based ingress and proxies access to MCP and OpenConnector.

Users can link agents to bots in chat apps (Slack, Telegram, Discord, Lark) or to repositories and issues in developer tools (GitHub, GitLab, Gitea, Linear). Agents can be assigned specific roles, call upon one another, and share reviewed knowledge.

Who it’s for

Teams that want to move AI agents from individual local environments into shared team workspaces to collaborate on tasks like issue triage, cross-platform support, and automated code reviews.

Highlights

  • Multi-Platform Integration: Works across Slack, Telegram, Discord, Lark, GitHub, GitLab, Gitea, and Linear.
  • Multi-Agent Collaboration: Supports agents from various runtimes (Claude Code, Codex, Grok Build, DeepSeek, Pi) side-by-side.
  • Shared Context: Agents have individual memory and can access a shared pool of published organization knowledge.
  • Self-Hostable: Available as an Apache-2.0 stack that can be deployed via Docker Compose or Kubernetes (Helm).
  • Flexible Control: Independent configuration for each agent's model, tools, workspace, and sandbox policy.

Related

  • Project
    openagents-org/openagents

    An open-source collaborative OS for AI agents that provides a unified workspace where multiple agents can share threads, files, and a browser to work together.

  • Project
    wanikua/danghuangshang

    A multi-agent collaboration system based on the OpenClaw framework that organizes AI agents into a hierarchical government or corporate structure for efficient task delegation and execution.

  • Project
    Shy2593666979/AgentChat

    A modern intelligent dialogue system that enables the creation and management of AI agents with multi-agent collaboration, RAG, and a three-layer memory architecture.

  • Project
    agentscope-ai/AgentTeams

    An open-source collaborative multi-agent runtime platform that uses a Manager-Workers architecture to orchestrate AI agents in secure, auditable IM rooms with full human visibility.

  • Project
    inkeep/agents

    A platform for building AI agents featuring a no-code visual builder and a TypeScript SDK with full two-way synchronization for collaborative agent management.