@tanstack/ai-anthropic 0.18.10 → 0.18.11

package/README.md

@@ -0,0 +1,213 @@
+<div align="center">
+  <picture>
+    <source
+      media="(prefers-color-scheme: dark)"
+      srcset="https://tanstack.com/api/readme/ai.png?theme=dark"
+    />
+    <source
+      media="(prefers-color-scheme: light)"
+      srcset="https://tanstack.com/api/readme/ai.png"
+    />
+    <img
+      src="https://tanstack.com/api/readme/ai.png"
+      alt="TanStack AI"
+      width="900"
+    />
+  </picture>
+</div>
+
+<br />
+
+# @tanstack/ai-anthropic
+
+Anthropic Claude adapter for TanStack AI chat, tool calling, thinking, and structured outputs
+
+## Installation
+
+```bash
+npm install @tanstack/ai-anthropic
+# or
+pnpm add @tanstack/ai-anthropic
+# or
+yarn add @tanstack/ai-anthropic
+```
+
+## Setup
+
+Set your API key as an environment variable:
+
+```bash
+export ANTHROPIC_API_KEY="sk-ant-..."
+```
+
+## Usage
+
+### Text/Chat Adapter
+
+```typescript
+import { chat } from '@tanstack/ai'
+import { anthropicText } from '@tanstack/ai-anthropic'
+
+const stream = chat({
+  adapter: anthropicText('claude-sonnet-4-6'),
+  messages: [{ role: 'user', content: 'Hello!' }],
+})
+```
+
+### With Explicit API Key
+
+```typescript
+import { createAnthropicChat } from '@tanstack/ai-anthropic'
+
+const adapter = createAnthropicChat(
+  'claude-sonnet-4-6',
+  process.env.ANTHROPIC_API_KEY!,
+  {
+    baseURL: 'https://api.anthropic.com', // optional, for custom endpoints
+  },
+)
+```
+
+### With Tools
+
+```typescript
+import { chat, toServerSentEventsResponse, toolDefinition } from '@tanstack/ai'
+import { anthropicText } from '@tanstack/ai-anthropic'
+import { z } from 'zod'
+
+const searchDatabase = toolDefinition({
+  name: 'search_database',
+  description: 'Search the database',
+  inputSchema: z.object({ query: z.string() }),
+}).server(async ({ query }) => {
+  return { results: [] }
+})
+
+export async function POST(request: Request) {
+  const { messages } = await request.json()
+
+  const stream = chat({
+    adapter: anthropicText('claude-sonnet-4-6'),
+    messages,
+    tools: [searchDatabase],
+  })
+
+  return toServerSentEventsResponse(stream)
+}
+```
+
+### Thinking
+
+Newer Claude models use adaptive thinking — the model decides when and how much to think, and depth is tuned with `output_config.effort`:
+
+```typescript
+const stream = chat({
+  adapter: anthropicText('claude-sonnet-5'),
+  messages: [{ role: 'user', content: 'Plan a database migration.' }],
+  modelOptions: {
+    thinking: { type: 'adaptive', display: 'summarized' },
+    output_config: { effort: 'xhigh' },
+    max_tokens: 64_000,
+  },
+})
+```
+
+Claude 4.6 models also accept the manual `{ type: 'enabled', budget_tokens }` shape; `budget_tokens` must be less than `max_tokens`. Which shapes and sampling parameters each model accepts is enforced by the adapter's types — see the docs for the per-model rules.
+
+### Prompt Caching
+
+Mark a content part with `cache_control` to cache it:
+
+```typescript
+const stream = chat({
+  adapter: anthropicText('claude-sonnet-4-6'),
+  messages: [
+    {
+      role: 'user',
+      content: [
+        {
+          type: 'text',
+          content: 'What is the capital of France?',
+          metadata: { cache_control: { type: 'ephemeral' } },
+        },
+      ],
+    },
+  ],
+})
+```
+
+### Summarization
+
+```typescript
+import { summarize } from '@tanstack/ai'
+import { anthropicSummarize } from '@tanstack/ai-anthropic'
+
+const result = await summarize({
+  adapter: anthropicSummarize('claude-sonnet-4-6'),
+  text: 'Your long text to summarize...',
+  maxLength: 100,
+  style: 'concise', // "concise" | "bullet-points" | "paragraph"
+})
+
+console.log(result.summary)
+```
+
+### Provider Tools
+
+Anthropic's native tools live on `@tanstack/ai-anthropic/tools` and go into `chat({ tools })` alongside your own: `webSearchTool`, `webFetchTool`, `codeExecutionTool`, `computerUseTool`, `bashTool`, `textEditorTool`, `memoryTool`, and `customTool`.
+
+```typescript
+import { chat } from '@tanstack/ai'
+import { anthropicText } from '@tanstack/ai-anthropic'
+import { webSearchTool } from '@tanstack/ai-anthropic/tools'
+
+const stream = chat({
+  adapter: anthropicText('claude-opus-4-7'),
+  messages: [{ role: 'user', content: "What's new in AI this week?" }],
+  tools: [
+    webSearchTool({
+      name: 'web_search',
+      type: 'web_search_20250305',
+      max_uses: 2,
+    }),
+  ],
+})
+```
+
+### Claude on Vertex AI
+
+Install `@anthropic-ai/vertex-sdk` next to this package and import from `@tanstack/ai-anthropic/vertex`:
+
+```typescript
+import { chat } from '@tanstack/ai'
+import { anthropicVertexText } from '@tanstack/ai-anthropic/vertex'
+
+const stream = chat({
+  adapter: anthropicVertexText('claude-sonnet-5', {
+    project: 'my-project',
+    location: 'europe-west1',
+  }),
+  messages: [{ role: 'user', content: 'Hello!' }],
+})
+```
+
+`location` is required (on the factory or via `GOOGLE_CLOUD_LOCATION`, `GOOGLE_VERTEX_LOCATION`, or `CLOUD_ML_REGION`); `project` can come from Application Default Credentials. Only the Claude models in the Vertex catalog are accepted.
+
+## Features
+
+- ✅ Streaming chat
+- ✅ Function/tool calling and Anthropic's native provider tools
+- ✅ Extended and adaptive thinking
+- ✅ Structured output
+- ✅ Prompt caching
+- ✅ Summarization (`anthropicSummarize`)
+- ✅ Claude on Vertex AI (`@tanstack/ai-anthropic/vertex`)
+
+## Documentation
+
+- [Anthropic adapter](https://tanstack.com/ai/latest/docs/adapters/anthropic): configuration, per-model thinking rules, every provider tool, and the API reference
+- [Provider Tools](https://tanstack.com/ai/latest/docs/tools/provider-tools): the concept, comparison matrix, and which models support which tools
+
+## License
+
+MIT

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@tanstack/ai-anthropic",
-  "version": "0.18.10",
+  "version": "0.18.11",
   "description": "Anthropic Claude adapter for TanStack AI chat, tool calling, thinking, and structured outputs.",
   "author": "Tanner Linsley",
   "license": "MIT",