# CloudBase AI Toolkit

An integration layer that connects AI coding agents to Tencent CloudBase as a backend, installing the stack through a plugin, steering code generation with skills, and operating databases, functions and deploys through MCP.

CloudBase AI Toolkit connects AI coding tools to Tencent CloudBase as a backend: a plugin installs the stack, Skills steer how the agent writes code against it, and an MCP server operates databases, functions, storage, and deployment directly from chat. It targets both a domestic and an international CloudBase site, with app deployment, database access (including a PostgreSQL/Supabase-compatible path), authentication, and cloud functions.

Recent updates widen an MCP-callable cloud API service allowlist to 57 services with built-in multi-version mapping, add region- and site-aware login and routing for the international site, complete an end-to-end cloud app upload and deploy path, add a PostgreSQL access-pattern best-practices skill covering batching, indexes, and row-level-security role gating, and load skill and API documentation on demand rather than up front, reducing the number of tools always listed to an MCP client.

It is written in TypeScript and released under the MIT license, and supports Claude Code, Cursor, and other MCP- and skill-compatible agents.

Website: https://docs.cloudbase.net/ai/cloudbase-ai-toolkit/
