Overview
Last updated:2026-03-11 17:10:06
Use AI coding tools to build Chat integrations. Describe what you want in plain language — the AI scaffolds the project, installs TUIKit, generates test credentials, and runs the app.
Model Context Protocol (MCP)
The Tencent RTC MCP server gives your AI access to TUIKit integration docs, test credential generation, and Chat management tools — all callable directly from your editor.
If you're using an AI coding tool to build a Chat integration, start by connecting it to the MCP server.
Use agent skills
Skills give AI agents task-specific instructions for Chat integration. They improve accuracy by automatically routing your request to the right tools and execution steps — no need to mention specific tool names or framework details in every prompt.
Tencent RTC offers the following Chat skills:
Chat routing skill: Identifies your target platform and integration mode, then routes to the correct MCP tools.
React skill: Step-by-step guidance for React TUIKit integration.
Vue skill: Step-by-step guidance for Vue 3 TUIKit integration.
Install with a single command: the CLI auto-detects your tools (Cursor, Claude Code, Codex, and others):
npx skills add Tencent-RTC/tencent-rtc-skills
Integration guides
AI can scaffold a complete Chat project or add TUIKit to an existing codebase. The integration guide covers all supported platforms:
Web: React 18, Vue3
Android: Java / Kotlin
iOS: Swift / Objective-C
Flutter: Dart (cross-platform)
Other scenarios
Chat window only: You only need one chat window on a page (no conversation list or full TUIKit). Web (React, Vue) only. See integrate Chat window only.
Troubleshooting
If you run into issues with MCP connections, credentials, packages, or platform-specific errors, see Troubleshooting.
Was this page helpful?
You can also Contact Sales or Submit a Ticket for help.
Yes
No
Feedback