M
MCP Directory CURATED DIRECTORY
911 MCP servers indexed
中文
← Back to directory
C

CloudBase AI Toolkit

Official
AI writes the code. CloudBase runs the backend.
How to install ↓ Compare GitHub source repository ↗
Category
Database #20 of 58
Stars
★ 1.1k Very popular
Transport
stdio (local process) · Streamable HTTP
Runtime
Node.js 18.15.0+ · Hosted (no local runtime)
Credentials
API key / credential required
License
MIT
Last commit
59FMRS · C

CloudBase AI Toolkit is Tencent Cloud's official MCP server, deeply integrated with CloudBase backend services. It is suitable for AI-driven development requiring database, functions, hosting, and authentication. Offers rich features and multiple setup options, but users must be cautious about AI operations on data.

Strongest · Maintenance 15/20 Weakest · Reliability 8/20

Reliability
8/20
Security and permissions
13/20
Maintenance
15/20
Documentation
13/20
Setup experience
10/20
Why each score
Reliability 8/20
The repository provides a tool manifest (scripts/tools.json) and a CLI entry (npx @cloudbase/cloudbase-mcp), but no test files or CI configuration to verify the MCP handshake and tool behavior. README claims support for multiple modes (local, hosted, cloud), but lacks reproducible verification evidence. Reliability is capped at 8 because key paths are plausible but untested.
Security and permissions 13/20
The repository provides a cloud-mode environment variable (CLOUDBASE_MCP_CLOUD_MODE) to disable local file/process tools and recommends confirmation of sensitive actions, but no code-level evidence (e.g., permission controls, confirmation mechanisms) is present. No malicious behavior or credential leakage found. Since least privilege and confirmation mechanisms are not verifiable from source, and hosted mode requires user-supplied keys, security scores 13.
Maintenance 15/20
The repository is active, with recent releases (v2.25.x) and a clear changelog, and an explicit MIT license. npm version and download badges indicate ongoing maintenance. However, no security response channel or dependency update policy is evidenced. Score 15.
Documentation 13/20
README provides detailed install instructions, multiple IDE integration config examples, and FAQs, but tool parameters, limits, and cost information are mainly in external docs, not verifiable from the repo. Troubleshooting steps are limited. Score 13.
Setup experience 10/20
README provides clear setup steps and multiple client config examples (e.g., MCP config JSON), but depends on external services (CloudBase environment) and npm package, and lacks local testing or verification steps. Setup score limited to 10 due to missing reproducible install verification evidence.

Static review · not runListed 2026-08-07

Read the FMRS scoring method →

Fit and risk

What it can accessReads local filesWrites / deletes local filesRuns commands or codeUses the networkChanges third-party account dataConnects to a database

Best for

  • Developers using AI coding assistants (like Cursor, Claude Code) who want full-stack development on Tencent Cloud
  • Teams that need rapid setup of apps with login, database, and permissions
  • Developers who want to perform environment configuration, database operations, and deployment from chat

Not for

  • Users who only need static hosting or frontend deployment without backend services
  • Users who cannot review AI-proposed sensitive actions (as the AI can directly modify data and permissions)
  • Users who do not want to use Tencent Cloud services

Required permissions

  • Requires login and access to a CloudBase environment
  • Access to database, cloud functions, storage, and other resources
  • May need elevated permissions to create databases, modify security rules, etc.

Risks and side effects

  • AI might execute destructive operations (e.g., deleting data) – review critical steps
  • Local mode runs on your machine; code doesn't leave until you deploy, but cloud traffic uses HTTPS
  • Hosted mode transmits SecretId/SecretKey over HTTP – ensure trusted environment

Setup

Before you start

Runtime：Node.js 18.15.0+ · Hosted (no local runtime)

Other optional settings (2)
CLOUDBASE_MCP_CLOUD_MODE optional Set to true to disable local file/process tools; required for self-hosted remote deployments. Nothing to obtain externally.
MCP_CLOUD_MODE optional Alternative name for CLOUDBASE_MCP_CLOUD_MODE; set to true to enable cloud mode. Nothing to obtain externally.
  1. Ensure Node.js v18.15.0+ and a CloudBase environment.
  2. Recommended plugin install: npx plugins add TencentCloudBase/cloudbase-plugin.
  3. Or use the CloudBase AI CLI: npm i -g @cloudbase/cli && tcb ai.
  4. Manual MCP configuration: add MCP server entry to your IDE's config file (e.g., .cursor/mcp.json).
.mcp.json 
{
  "mcpServers": {
    "cloudbase": {
      "command": "npx",
      "args": [
        "@cloudbase/cloudbase-mcp@latest"
      ]
    }
  }
}

Shown for Claude Code. Other clients may use a different file or key (VS Code uses "servers") — the configurator below converts it.

.vscode/mcp.json
{
  "servers": {
    "cloudbase": {
      "command": "npx",
      "args": [
        "@cloudbase/cloudbase-mcp@latest"
      ]
    }
  }
}

Goes in your project's .vscode/mcp.json (VS Code uses a "servers" key).

Terminal
claude mcp add cloudbase -- npx @cloudbase/cloudbase-mcp@latest

Run it in a terminal; replace any <…> placeholders with your own values first.

Generate and validate client config

Check that it works

After connecting, send the prompt "Login to CloudBase"; if login succeeds and your CloudBase environment is listed, the server is working. You can also confirm cloudbase tools appear in the client's tool list.

Troubleshooting

  1. If login says environment not found, verify the environment exists and is healthy in the console, then re-login
  2. Ensure Node.js version >= 18.15.0
  3. For hosted mode, check enable_plugins/disable_plugins environment variable configuration
  4. Verify MCP config command and args (e.g., 'npx @cloudbase/cloudbase-mcp@latest')

Things to try

Once connected, you can ask your AI assistant things like:

  • Login to CloudBase
  • Use CloudBase Skills to build a todo app with login, database and permissions, then deploy
  • Create a cloud function, deploy it, and invoke it
  • Show recent logs for my cloud functions

Use cases

Operate CloudBase databases (PostgreSQL and document DB) from chat
Author, deploy, invoke, and debug cloud functions / Cloud Run
Manage object storage and login methods
Publish static hosting or Mini Programs
Inspect logs and redeploy environments

Supported clients

Claude Code
Cursor
Codex
Gemini CLI
OpenCode
Windsurf
Trae
Tongyi Lingma
Roo Code
Cline
GitHub Copilot
Baidu Comate
Augment Code
OpenClaw
WorkBuddy
ZCode
CodeBuddy

Listed from the project's documentation, not tested by this site.

Overview

CloudBase AI Toolkit is Tencent Cloud's integration layer for AI coding tools. It enables AI IDEs to interact with CloudBase backend services including database, storage, authentication, cloud functions, and hosting through plugins, agent skills, and an MCP server. The MCP server provides tools for login, querying and modifying data, managing functions and hosting, and reading logs. The package '@cloudbase/cloudbase-mcp' along with skills and plugins are provided in this repo.

Similar servers

Compare with other MCP servers

Source revision 6b9bcd1d74ec Data synced 2026-10-12 Read the FMRS scoring method