AgentConnect today announced the launch of its open-source platform designed to enable teams and multiple AI agents to work together seamlessly within the communication and development tools they already use. The platform addresses a critical limitation in current AI agent deployments, where agent work is often isolated to individual terminals, leading to fragmented context and duplicated integration efforts.
“AI agents become much more useful when they can work with the rest of the team, not just one person in one terminal,” the AgentConnect team stated. “AgentConnect brings agents with different roles into the conversations where work is already happening.” This approach allows teams to integrate ACP-compatible agents like Claude Code and Codex into shared conversations, assigning distinct roles to each agent and facilitating collaboration between humans and agents.
The platform's utility is demonstrated in scenarios such as incident response. A team can deploy a triage agent to inspect available context, a code specialist to trace a suspected regression, and a verification agent to check proposed results—all within a single Slack thread. Human teammates can interject with additional context, redirect tasks, and monitor the diagnosis and decision-making process in real time.
Beyond incident response, AgentConnect supports a wide range of applications. For instance, customer support issues can transition from a Telegram conversation to a trusted Slack workspace without manual context copying. Engineering teams can combine general and specialist reviewers to assess pull request risks, with inline GitHub comments and independent check results. Additionally, scheduled and webhook-driven tasks can continue in shared conversations when human input becomes necessary.
AgentConnect offers several key features: multiple agents working as a team with distinct models, workspaces, memory, MCP servers, skills, repository access, and sandbox policies; integration with existing tools like Slack, Telegram, and Discord; the ability to initiate work from conversations, GitHub pull requests, webhooks, or schedules; and fine-grained access control that separates agent access from session visibility, allowing teams to manage which tools, repositories, and other agents each agent can reach. It also supports linking multiple social accounts to one member profile for consistent permission application.
As an open-source solution licensed under Apache 2.0, AgentConnect can be self-hosted using Docker Compose or Kubernetes. This ensures that agent execution and workspaces remain within the team's own environment, providing full control and security. The platform is available immediately, and interested parties can explore the source code on GitHub, visit agentconnect.md for more information, or read the documentation.
The launch of AgentConnect marks a significant step forward in making AI agents collaborative team members rather than isolated tools. By embedding agent interactions into existing communication channels, it reduces friction and enhances productivity, potentially transforming how organizations leverage AI for complex tasks.
