Skip to main content

Tencent Cloud Goes Open Source with TeamAI, an Internal Tool for Cross-Agent Skill Sharing

Tencent Cloud Opensources TeamAI, a Git-Based Hub for AI Agent Skills

Tencent Cloud has taken a significant step toward streamlining AI collaboration by open-sourcing its internal tool, TeamAI. Built on Git, this tool aims to unify a team's skills, work standards, tool configurations, and project knowledge into a single repository. The idea is simple: manage AI agent skills like code, with version control and review processes, then sync everything to each member's agent. No more chasing colleagues for the latest config—it's "one configuration, all reuse."

Image

Works with 16 Major AI Agents

TeamAI currently plays nice with 16 popular agents, including WorkBuddy, CodeBuddy, Claude Code, Codex, and Cursor. That means teams aren't locked into one ecosystem; they can share skills across different tools. When someone submits a new skill or tweaks a standard, the system automatically creates a branch and opens a merge request for team review. Once approved, agents that support session start hooks will pull the latest updates at the beginning of a new conversation. No more manual replacements or mass notifications—it just works.

Security and Model Configs Under Control

For administrators, TeamAI offers a way to distribute model configurations that meet enterprise security requirements. This reduces the risk of team members independently selecting incorrect or unsafe models. The tool also prompts members to share experiences based on correction records and tool failures at the end of a session. These insights are compiled by the agent and stored in the repository, ready for reference in future tasks. It's like a collective memory for your team's AI interactions.

Integrated with ClawPro for Centralized Management

TeamAI is also integrated into Tencent Cloud's intelligent agent management platform, ClawPro. Members can connect their local agents to ClawPro, allowing administrators to centrally manage Skills, MCP, and other configurations through a control panel. These can then be efficiently distributed to all connected agents, further simplifying workflow collaboration and standard implementation in the era of large models.

Key Points

  • Git-based collaboration: Manage AI skills, standards, and configs in a Git repo with version control and review.
  • Broad compatibility: Works with 16 major agents, including WorkBuddy, Claude Code, and Cursor.
  • Automated sync: Agents pull updates at session start, eliminating manual updates.
  • Security controls: Admins can enforce model configurations that meet enterprise security requirements.
  • ClawPro integration: Centralized management of Skills, MCP, and more through Tencent Cloud's platform.

With TeamAI, Tencent Cloud is betting that the future of AI collaboration is open, standardized, and Git-native. And by open-sourcing it, they're inviting the wider community to build on that vision.