Vercel AI SDK — Build AI-Powered Apps in TypeScript
You are an expert in the Vercel AI SDK, the TypeScript toolkit for building AIpowered applications. You help developers integrate LLMs (OpenAI, Anthropic, Google, Mistral, Ollama) with React Server Components, streaming UI, tool calling, structured output with Zod schemas, RAG pipelines, multistep agents, and edgecompa
概览
This skill provides expert guidance on the Vercel AI SDK, a specialized TypeScript toolkit designed for developing AI-driven applications. It facilitates the seamless integration of major large language models, including OpenAI, Anthropic, Google, and Mistral, into modern web frameworks. Key features covered include React Server Components, real-time streaming UI, and structured data output using Zod schemas. Part of the TerminalSkills/skills repository, this resource assists developers in implementing complex patterns such as Retrieval-Augmented Generation (RAG) pipelines and multi-step autonomous agents. By leveraging edge computing capabilities, it ensures high-performance AI interactions. The skill serves as a technical reference for managing tool calling and data processing within a TypeScript environment for building scalable AI interfaces.
使用场景
安装说明
# Review source first
open https://github.com/TerminalSkills/skills/blob/main/skills/ai-sdk/SKILL.mdCopy or clone the skill folder into your agent skills directory after reviewing its instructions and scripts.
安全提示
Users must ensure secure management of API keys for integrated LLM providers and validate all structured outputs generated by the SDK. Implementation should follow standard TypeScript security practices for edge computing environments as documented in the TerminalSkills/skills repository.
相关 Skills
MCP Builder
anthropics/skills
创建高质量 MCP (Model Context Protocol) 服务器的指南,使 LLM 能够通过设计良好的工具与外部服务进行交互。在构建 MCP 服务器以集成外部 API 或服务时使用,无论是使用 Python (FastMCP) 还是 Node/TypeScript (MCP SDK)。
Claude API
anthropics/skills
Claude API / Anthropic SDK 参考指南——模型 ID、定价、参数、流式传输、工具使用、MCP、Agent、缓存、Token 计数、模型迁移。 TRIGGER——在打开目标文件前阅读 BEFORE;不要因为它“看起来像一行代码”而跳过——无论何时:提示词中以任何形式提及 Claude/Anthropic(Claude、Anthropic、Fable、Opus、Sonnet、Haiku、`anthropic`、`@anthropic-ai`、`claude-*`、`us.anthropic.*`、`[1m]`);用户询问有关 LLM 的问题(定价/模型选择/限制/缓存)——切勿回答
Skill Creator
anthropics/skills
创建新技能、修改和改进现有技能,并衡量技能性能。适用于用户希望从零开始创建技能、编辑或优化现有技能、运行评估以测试技能、通过方差分析对技能性能进行基准测试,或优化技能描述以提高触发准确性的场景。
API And Interface Design
addyosmani/agent-skills
指导稳定的 API 和接口设计。适用于设计 API、模块边界或任何公共接口。在创建 REST 或 GraphQL 端点、定义模块间的类型契约或建立前后端边界时使用。
Browser Testing With Devtools
addyosmani/agent-skills
通过 Chrome DevTools MCP 在真实浏览器中进行测试。在构建或调试任何在浏览器中运行的内容时使用。当你需要检查 DOM、捕获控制台错误、分析网络请求、分析性能或使用真实运行时数据验证视觉输出时使用。需要配置 chrome-devtools MCP 服务器。
Windows Builder
hashicorp/agent-skills
使用 Packer、WinRM 通信器和 PowerShell 配置程序构建 Windows 镜像。适用于创建 Windows AMIs、Azure 镜像或 VMware 模板。