skills.rest
aicodepro
aicodeproOFFICIAL
Explore all skills by this author →
GitHub

ai-sdk-core

Implement server-side AI text generation, structured output, and tool-calling agents with Vercel AI SDK v5.

Updated Jun 22, 2026
Search Tags:#typescript#llm#tool calling#streaming#structured output#vercel ai sdk#text generation
One-click install
npx skills add https://github.com/aicodepro/ai-agent-nexi --skill ai-sdk-core-aicodepro
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ai-sdk-core
Source: https://github.com/aicodepro/ai-agent-nexi/tree/main/agent/skills/ai-sdk-core
Command: npx skills add https://github.com/aicodepro/ai-agent-nexi --skill ai-sdk-core-aicodepro

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires ai, @ai-sdk/openai, @ai-sdk/anthropic, @ai-sdk/google, workers-ai-provider, zod, and includes scripts (resource) and references (resource) components.

What problem does it solve? Building backend AI features requires juggling multiple provider APIs, inconsistent error types, and breaking changes between SDK versions. This Skill provides production-ready patterns for the Vercel AI SDK v5, covering text generation, structured output, tool calling, and multi-provider configuration in one place. ## Core Features & Use Cases - Four Core Functions: Guidance and templates for generateText, streamText, generateObject, and streamObject with Zod schema validation. - Multi-Provider Setup: Configuration patterns for OpenAI, Anthropic, Google Gemini, and Cloudflare Workers AI through a unified API. - Error Catalog & Migration: Solutions for 12 common errors (AI_APICallError, AI_NoObjectGeneratedError, Worker startup limits) plus a v4-to-v5 migration guide. - Use Case: You are building a Next.js API route that extracts structured contact data from user-submitted text. Use this Skill to generate a generateObject implementation with a Zod schema, proper error handling, and retry logic. ## Quick Start Ask the AI to generate a TypeScript function using the Vercel AI SDK v5 that streams a chat response from OpenAI with error handling.

Frequently Asked Questions about ai-sdk-core

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate structured JSON output with the Vercel AI SDK?▼

Use generateObject with a Zod schema defining your expected shape. The SDK validates the model output against the schema and returns a typed object, throwing AI_NoObjectGeneratedError if the model fails to produce valid output.

What changed in Vercel AI SDK v5 vs v4?▼

Key breaking changes include maxTokens renamed to maxOutputTokens, tool parameters renamed to inputSchema, maxSteps replaced by stopWhen(stepCountIs(n)), and CoreMessage renamed to ModelMessage. Run bunx ai migrate to automate most changes.

Does the AI SDK work with Cloudflare Workers?▼

Yes, via the workers-ai-provider community package. Create the provider inside request handlers rather than at top level to avoid exceeding the Workers startup time limit, and return stream.toTextStream() for streaming responses.

Why does streamText fail silently without throwing errors?▼

Stream errors could be swallowed by createDataStreamResponse in older versions. This was fixed in [email protected]; use the onError callback on streamText to capture and log stream errors reliably.

Which providers does the Vercel AI SDK support?▼

The SDK supports OpenAI, Anthropic, Google Gemini, and Cloudflare Workers AI through official or community provider packages, plus 25+ other providers. All share the same generateText, streamText, and generateObject API.

When should I not use the AI SDK core for backend AI?▼

Avoid it for React chat UIs needing hooks like useChat, which belong to the ai-sdk-ui skill. For Cloudflare-only deployments without multi-provider needs, the native Workers AI binding offers better performance.

Related Skills

Explore More Software Engineering Skills→
juanrejuanre
3

providers

Switch between LLM providers and access raw SDKs via get_provider().

Community
Intermediate
zot24zot24
2

chat-sdk

Scaffold Next.js AI chatbots with generative UI and provider integrations.

Community
Advanced
mhbdevmhbdev

ai-sdk-ui

Develop React chat interfaces using useChat, useCompletion, and useObject hooks.

Community
Advanced
9Ashwin9Ashwin

lark-openapi-explorer

Explore Lark OpenAPI docs and execute API calls via lark-cli.

Community
Advanced