AgentConnect today announced the release of its open-source platform designed to enable teams and multiple AI agents to work together across the communication and development tools they already use. The platform aims to address a growing challenge in the AI agent ecosystem: while individual agents have become increasingly capable, their work often remains isolated within a single terminal or session, limiting collaboration and context sharing.
According to the AgentConnect team, most agent workflows begin and end with one person, leading to fragmented context and duplicated efforts. Teams often build separate bots and integrations for each new workflow, and existing team-layer products typically require moving work into a dedicated workspace or tie users to a single agent provider. AgentConnect seeks to change this by bringing agents directly into existing team conversations.
With AgentConnect, teams can connect ACP-compatible agents such as Claude Code and Codex, assign each a distinct role, and enable collaboration through shared conversations and agent-to-agent calls. This allows human teammates to participate in the same threads where agents are working, providing context, redirecting efforts, and following decisions in real time.
For example, during an incident response, a triage agent can inspect context, a code specialist can trace a suspected regression, and a verification agent can check proposed results—all within a Slack thread where human teammates can contribute. The platform extends beyond incident response to support scenarios like customer support, where issues can move from a Telegram conversation to a trusted Slack workspace without manual context copying. Engineering teams can also combine general and specialist reviewers around pull requests, with inline GitHub comments and independent check results.
Key features of AgentConnect include support for multiple agents with distinct configurations for models, workspaces, memory, MCP servers, skills, repository access, and sandbox policies. The platform integrates with Slack, Telegram, and Discord, and can start work from conversations, GitHub pull requests, webhooks, or schedules. It also offers fine-grained access control, allowing administrators to manage agent access separately from session visibility, and control which tools, repositories, and other agents each agent can reach.
AgentConnect is open source under the Apache 2.0 license and can be self-hosted with Docker Compose or Kubernetes, ensuring that agent execution and workspaces remain within the team’s environment. The AgentConnect OSS is available today, with the source code hosted on GitHub and documentation available at agentconnect.md.
The launch of AgentConnect comes at a time when organizations are increasingly exploring ways to integrate AI agents into their workflows. By enabling multi-agent collaboration within existing tools, AgentConnect addresses a critical need for scalability and context sharing, potentially making AI agents more useful in team settings. The platform’s open-source nature and self-hosting options may appeal to organizations with strict data governance requirements.
