Tencent Cloud has officially announced the open-sourcing of TeamAI, an internal AI collaboration tool that has been in use by its teams. This innovative platform is built upon the familiar Git framework, aiming to streamline how teams manage and share AI-related resources.
Unified AI Resource Management
TeamAI centralizes essential AI components, including skills (Skills), work specifications, tool configurations, and project knowledge, storing them within a Git repository. This approach allows for robust version control, code review, and update management, akin to how software developers manage code. Once approved, these configurations are synchronized across team members’ individual AI Agents, enabling a “configure once, reuse by all” paradigm.
Currently, TeamAI boasts compatibility with a wide array of 16 Agents, including popular platforms like WorkBuddy, CodeBuddy, Claude Code, Codex, and Cursor, demonstrating its versatility and broad applicability.
Familiar Git-based Workflow
Leveraging the developer-centric Git workflow, TeamAI simplifies the process of updating and sharing AI configurations. When a team member contributes a new skill or modifies existing specifications, the system automatically generates a new branch and initiates a merge request for team review. Upon approval and merging, Agents equipped with session startup hooks can automatically pull these updates at the beginning of a new conversation. This eliminates the need for manual updates across individual Agents and reduces the burden of notifying each team member.
Enhanced Model Configuration and Knowledge Sharing
A recent enhancement to TeamAI includes the capability for unified model configuration deployment. For tasks involving sensitive internal data and proprietary knowledge, team administrators can now centrally configure model services that adhere to the team’s security requirements. This mitigates the risks associated with team members independently integrating or selecting incorrect models.
Furthermore, TeamAI actively promotes knowledge sharing by capturing and documenting lessons learned. When an Agent encounters repeated errors, fails tool invocations, or experiences other issues, it prompts team members to share their insights at the end of a session. The Agent then compiles these problems and their resolutions into documentation stored within the team’s repository. This accumulated knowledge base serves as a valuable resource, allowing the Agent to intelligently retrieve relevant solutions for future tasks based on current needs.
Seamless Integration and Management
With TeamAI Skill installed, both administrators and team members can interact with AI tools using natural language commands to perform operations. TeamAI is also integrated into Tencent Cloud’s intelligent agent management platform, ClawPro. This integration allows team members to connect their locally used Agents to ClawPro, where administrators can centrally manage Skills, MCP (Model Configuration Platform), and other settings from a control panel. These configurations can then be efficiently distributed to all connected Agents.
This open-source release of TeamAI signifies Tencent Cloud’s commitment to fostering collaborative AI development and empowering teams with efficient, secure, and manageable AI tools.
