AgentConnect, a San Francisco-based technology company, has announced the release of its open-source platform designed to enable teams to collaborate with multiple AI agents within the communication and development tools they already use. The platform, available under the Apache 2.0 license, aims to address the limitations of current AI agent workflows, which often operate in isolation within a single terminal and lack integration with team collaboration.
The core premise of AgentConnect is to bring AI agents into existing team conversations, such as Slack threads, Telegram chats, or Discord channels, rather than requiring teams to adopt a new dedicated workspace. This approach allows teams to connect agents that are compatible with the Agent Client Protocol (ACP), including Claude Code and Codex, and assign distinct roles to each agent. By doing so, people and agents can collaborate through shared conversations and agent-to-agent interactions, streamlining workflows and reducing context fragmentation.
One of the key use cases highlighted by the company is incident response. A team responding to an incident 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. Human team members can add context, redirect work, and follow the diagnosis and decisions within the same Slack thread. This model extends beyond incident response to support issues, where a customer-facing Telegram conversation can be moved into a trusted Slack workspace without manual context copying, and to engineering workflows, where general and specialist reviewers can collaborate on pull requests with inline GitHub comments and independent check results.
AgentConnect offers several features designed to enhance team collaboration with AI agents. It supports the configuration of multiple agents, each with distinct models, workspaces, memory, MCP servers, skills, repository access, and sandbox policies. The platform integrates with existing tools such as Slack, Telegram, and Discord, allowing work to be initiated from conversations, GitHub pull requests, webhooks, or schedules. It also enables work to continue between trusted messaging workspaces.
Security and access control are integral to the platform. AgentConnect provides fine-grained access control, allowing administrators to manage agent access separately from session visibility, control which tools, repositories, and other agents each agent can access, and link multiple social accounts to one member profile for consistent permission enforcement across platforms.
As an open-source solution, AgentConnect can be self-hosted via Docker Compose or Kubernetes, ensuring that agent execution and workspaces remain within the environment the team operates. The platform is available immediately, with the source code accessible on GitHub, and additional information can be found at agentconnect.md.
The introduction of AgentConnect has significant implications for the growing field of AI-powered collaboration. By enabling teams to leverage multiple AI agents within their existing workflows, the platform could reduce the friction associated with adopting AI tools and improve productivity across various industries. The open-source nature of the project also fosters community-driven development, potentially leading to rapid innovation and customization. As AI agents become more capable, platforms like AgentConnect may become essential for organizations seeking to integrate these technologies into their daily operations.
