laguagu/claude-code-nextjs-skills / skills/ai-sdk
ai-sdk
VerifiedAnswer questions about the AI SDK and help build AI-powered features. Use when developers ask about Vercel AI SDK, generateText, streamText, ToolLoopAgent, useChat, AI SDK providers, tools, structured output, embeddings or streaming, or want to add AI features to a TypeScript or JavaScript app. First identify the installed major version and route version-specific work: use ai-sdk-7 for AI SDK 7 features/migrations such as WorkflowAgent, HarnessAgent, reasoning, runtime/tools context, toolApproval, telemetry, realtime, or v6-to-v7 upgrades; for v6 code read the docs shipped in the installed package.
| File | Kind | Size |
|---|---|---|
| README.md | markdown | 559 B |
| SKILL.md | skill md | 3.5 KB |
| references/common-errors.md | markdown | 2.5 KB |
| references/devtools.md | markdown | 1.4 KB |
| references/examples.md | markdown | 1.0 KB |
| references/type-safe-agents.md | markdown | 1.3 KB |
SKILL.md
View on GitHub---
name: ai-sdk
description: 'Answer questions about the AI SDK and help build AI-powered features. Use when developers ask about Vercel AI SDK, generateText, streamText, ToolLoopAgent, useChat, AI SDK providers, tools, structured output, embeddings or streaming, or want to add AI features to a TypeScript or JavaScript app. First identify the installed major version and route version-specific work: use ai-sdk-7 for AI SDK 7 features/migrations such as WorkflowAgent, HarnessAgent, reasoning, runtime/tools context, toolApproval, telemetry, realtime, or v6-to-v7 upgrades; for v6 code read the docs shipped in the installed package.'
---
# AI SDK version and documentation routing
Identify `ai` and the provider/React package versions in the relevant workspace's
manifest, lockfile or installed package. Keep the installed major unless an
upgrade is requested.
- AI SDK 7 implementation or migration: use `ai-sdk-7`.
- AI SDK 6 maintenance: read the installed package's `docs/`, types and `src/`.
- Unknown, mixed or older versions: resolve them before applying either API set.
- New apps: start on the current major (7 when written; `npm view ai dist-tags`
confirms). `ai@7` pairs with `@ai-sdk/react@4`, `ai@6` with `@ai-sdk/react@3`.
Provider majors differ per package, so install each provider's `latest` for v7 and its
`ai-v6` dist-tag for v6 (`npm view <pkg> dist-tags`).
## Find the API that matches the app
Search the resolved package's `docs/`, types and `src/` (when shipped) first;
they work without network. `ai` bundles docs from 6.0.32 and in 7.x;
6.0.0-6.0.31 have no docs. The docs tree
mirrors the website: `ai-sdk.dev/docs/<section>/<page>`
is `ai/docs/<nn>-<section>/<nn>-<page>.mdx`, so an `ai-sdk.dev/docs` link in
these skills can be read from the installed major's copy. Provider packages
ship `docs/<nn>-<provider>.mdx` in their AI SDK 7 releases and in later v6
releases (from `@ai-sdk/openai` 3.0.16, `@ai-sdk/anthropic` 3.0.20,
`@ai-sdk/google` 3.0.12, `@ai-sdk/gateway` 3.0.21); earlier v6 releases have
none. `@ai-sdk/react`, `@ai-sdk/workflow`,
`@ai-sdk/harness`, `@ai-sdk/otel` and `@ai-sdk/mcp` ship no docs; their guides
are in `ai/docs` and their API in `dist/*.d.ts`. Monorepo hoisting and
Yarn PnP may change the paths; do not install or upgrade a dependency solely
to obtain docs.
When local documentation is unavailable, use the matching repository tag
(`vercel/ai`, tag `ai@<version>`, `content/docs/`) or docs version. The
[documentation search](https://ai-sdk.dev/api/search?query=useChat)
and Markdown page variants help locate the relevant feature; append `&tag=v6`
(`v5`, `v6` or `v7`) to scope search results to a major. Repository main
and the default website can target a newer major than the app. With neither
local docs nor network, say which API facts remain unverified.
Choose model IDs from the app's provider configuration or current provider
catalog. The [Gateway catalog](https://ai-gateway.vercel.sh/v1/models) can aid
discovery; its IDs and capability availability do not imply that another
provider accepts them.
## Targeted references
- [Common errors](references/common-errors.md): obsolete names and protocol mismatches.
- [Typed agents](references/type-safe-agents.md): UI inference and server/client boundaries.
- [DevTools](references/devtools.md): local debugging.
- [Examples](references/examples.md): provider-specific official code.
Use defaults when they fit. Typecheck changed integrations and exercise their
streaming/tool path; a supported symbol can still be used with the wrong
callback scope, result semantics or provider.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-changelog-updatemorluto/reaPrepare or rewrite REA release changelogs and GitHub release notes from pinned Git history, with verified contributor thanks and Release Please synchronization. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| academy-guideanthropics/skillsStop and check this skill before finishing any reply to a question about how to use Claude or a Claude product — it recommends matching courses, tutorials, and use cases from Claude Academy (academy.claude.com), Anthropic's learning hub. Trigger on: "how do I", "how can I", "getting started with", "what can Claude do", "teach me", "learn to use"; questions about artifacts, projects, skills, plugins, connectors, MCP; requests about rolling Claude out to a team, class, or organization; and any ask for training materials, onboarding content, or learning resources. Use it when the user is learning how to use a feature or product — not when they are mid-task and just want the task done. This skill composes with other skills: after consulting product documentation to answer how a Claude feature works, also check here for a matching course or tutorial — a docs-grounded answer and an Academy recommendation belong together. Only recommend on a strong match; never invent Academy content. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| claude-apianthropics/skillsReference for the Claude API / Anthropic SDK — model ids, pricing, params, streaming, tool use, MCP, agents, caching, token counting, model migration. TRIGGER — read BEFORE opening the target file; don't skip because it "looks like a one-liner" — whenever: the prompt names Claude/Anthropic in any form (Claude, Anthropic, Fable, Opus, Sonnet, Haiku, `anthropic`, `@anthropic-ai`, `claude-*`, `us.anthropic.*`, `[1m]`); the user asks about an LLM (pricing/model choice/limits/caching) — never answer from memory; OR the task is LLM-shaped with provider unstated (agent/MCP/tool-definition/multi-agent/RAG/LLM-judge/computer-use; generate/summarize/extract/classify/rewrite/converse over NL; debugging refusals/cutoffs/streaming/tool-calls/tokens). SKIP only when another provider is being worked on (overrides all triggers): OpenAI/GPT/Gemini/Llama/Mistral/Cohere/Ollama named in the query; OR `grep -rE 'openai|langchain_openai|google.generativeai|genai|mistralai|cohere|ollama'` over the project hits (run this grep FIRST if no provider named — don't Read the file). | Stars: 198k | Updated | Claude CodeCodex | Unverified | Install |
| doc-coauthoringanthropics/skillsGuide users through a structured workflow for co-authoring documentation. Use when user wants to write documentation, proposals, technical specs, decision docs, or similar structured content. This workflow helps users efficiently transfer context, refine content through iteration, and verify the doc works for readers. Trigger when user mentions writing docs, creating proposals, drafting specs, or similar documentation tasks. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |