⚠
High Risk:This skill has significant security concerns. Review the findings below before installing.

ai-sdk-core

⚠Caution·Scanned 2/17/2026

Documents backend AI usage with Vercel AI SDK v5/v6 including providers, tools, MCP, streaming, and migration guidance. Contains instructions that run subprocesses (command: 'npx'), execute shell commands (npm install, npx mcp-to-ai-sdk), and use env vars (OPENAI_API_KEY, ANTHROPIC_API_KEY).

by jezweb·v10a1f16·146.9 KB·484 installs
Scanned from main at 10a1f16 · Transparency log ↗
$ vett add jezweb/claude-skills/ai-sdk-coreReview security findings before installing

AI SDK Core

Backend AI with Vercel AI SDK v5/v6 - text generation, structured output, multi-modal (speech, transcription, image gen), embeddings, tools, and agents.

Auto-Trigger Keywords

This skill should be discovered when working with:

Primary Keywords (High Priority)

  • ai sdk core, vercel ai sdk, ai sdk v5, ai sdk v6
  • generateText, streamText, generate text ai
  • Output.object, Output.array, Output.choice (v6)
  • ai sdk node, ai sdk server, ai sdk backend
  • zod ai schema, zod ai validation
  • ai tools calling, ai agent class, agent with tools
  • openai sdk, anthropic sdk, google gemini sdk
  • multi-provider ai, ai provider switching

v6 Output API Keywords (CRITICAL)

  • Output API, Output.object, Output.array, Output.choice
  • generateObject deprecated, streamObject deprecated
  • ai sdk v6 migration, v6 breaking changes
  • structured output v6, ai sdk structured data

Multi-Modal Keywords

  • ai sdk speech, generateSpeech, text to speech ai
  • ai sdk transcription, transcribe audio, speech to text ai
  • ai sdk image generation, generateImage, dall-e api
  • ai sdk embeddings, embed, embedMany, cosineSimilarity
  • multi-modal prompts, pdf ai, image ai, file ai

Advanced Features Keywords

  • ai sdk reranking, rerank documents, rag relevance
  • ai sdk middleware, wrapLanguageModel, extractReasoningMiddleware
  • ai sdk telemetry, opentelemetry ai, ai tracing
  • ai sdk mcp, model context protocol, mcp tools
  • tool approval, human in the loop ai, needsApproval

Secondary Keywords (Medium Priority)

  • ai streaming backend, stream ai responses
  • ai server actions, nextjs ai server
  • cloudflare workers ai sdk, workers-ai-provider
  • ai sdk migration, v4 to v5 migration, v5 to v6 migration
  • ai chat completion, llm text generation
  • ai sdk typescript, typed ai responses
  • stopWhen ai sdk, multi-step ai execution
  • dynamic tools ai, runtime tools ai

Error Keywords (Discovery on Errors)

  • AI_APICallError, ai api call error
  • AI_NoObjectGeneratedError, no object generated
  • AI_LoadAPIKeyError, ai api key error
  • AI_InvalidArgumentError, invalid argument ai
  • AI_TypeValidationError, zod validation failed
  • AI_RetryError, ai retry failed
  • AI_NoSpeechGeneratedError, speech generation failed
  • AI_NoTranscriptGeneratedError, transcription failed
  • streamText fails silently, stream error swallowed
  • worker startup limit ai sdk, 270ms startup
  • ai rate limit, rate limiting ai
  • maxTokens maxOutputTokens, v5 breaking changes
  • providerMetadata providerOptions, tool inputSchema
  • ToolExecutionError removed, tool-error parts

Framework Keywords

  • nextjs ai sdk, next.js server actions ai
  • cloudflare workers ai integration
  • node.js ai sdk, nodejs llm
  • vercel ai deployment, serverless ai

Provider Keywords

  • openai integration, gpt-5 api, gpt-5.2 api, chatgpt api
  • anthropic claude, claude api integration, claude 4 api
  • google gemini api, gemini 2.5 integration
  • cloudflare llama, workers ai llm
  • elevenlabs ai sdk, deepgram ai sdk

What This Skill Provides

  • v6 Output API: Output.object/array/choice replaces deprecated generateObject/streamObject
  • Multi-Modal: Speech synthesis, transcription, image generation, embeddings
  • Core Functions: generateText, streamText with structured output
  • MCP Integration: Model Context Protocol tools
  • Middleware: Language model middleware patterns
  • Telemetry: OpenTelemetry integration
  • Top Providers: OpenAI, Anthropic, Google, Cloudflare, ElevenLabs, 60+ more
  • v4→v5 Migration: Complete breaking changes guide
  • Top 12 Errors: Common issues with solutions
  • Production Patterns: Best practices for deployment

Quick Links

Installation

npm install ai @ai-sdk/openai @ai-sdk/anthropic @ai-sdk/google zod

Usage

See SKILL.md for comprehensive documentation and examples.

Version

  • Skill Version: 2.0.0
  • AI SDK Version: 6.1.0
  • Last Updated: 2026-01-03

Recent Updates (v2.0.0 - 2026-01-03)

  • BREAKING: Output API: Added v6 Output.object/array/choice documentation (generateObject/streamObject deprecated)
  • Multi-Modal: Added speech synthesis, transcription, image generation sections
  • Embeddings: Added embed(), embedMany(), cosineSimilarity() with full examples
  • MCP Tools: Added Model Context Protocol integration
  • Middleware: Added wrapLanguageModel and extractReasoningMiddleware patterns
  • Telemetry: Added OpenTelemetry integration
  • Models: Added GPT-5.2, o3, o3-mini, o4-mini model references
  • Providers: Updated to 69+ providers (was ~25)
  • Errors: Updated to 31 error types (was 12 documented)

Previous Updates (v1.2.0 - 2025-11-22)

  • v6 Beta: Added AI SDK 6 beta features (Agent abstraction, tool approval, reranking)
  • Updated Models: GPT-5.1, Claude Sonnet 4.5, Gemini 2.5 models

License

MIT