# Technical Documentation: anthropics/skills

> ℹ️ **Provenance:** Multi-Tier Ingestion · [CodeWiki Reference](https://codewiki.google/github.com/anthropics/skills)

### Overview (`README.md`)

> **Note:** This repository contains Anthropic's implementation of skills for Claude. For information about the Agent Skills standard, see [agentskills.io](http://agentskills.io).

[](https://skills.sh/anthropics/skills)

# Skills
Skills are folders of instructions, scripts, and resources that Claude loads dynamically to improve performance on specialized tasks. Skills teach Claude how to complete specific tasks in a repeatable way, whether that's creating documents with your company's brand guidelines, analyzing data using your organization's specific workflows, or automating personal tasks.

For more information, check out:
- [What are skills?](https://support.claude.com/en/articles/12512176-what-are-skills)
- [Using skills in Claude](https://support.claude.com/en/articles/12512180-using-skills-in-claude)
- [How to create custom skills](https://support.claude.com/en/articles/12512198-creating-custom-skills)
- [Equipping agents for the real world with Agent Skills](https://anthropic.com/engineering/equipping-agents-for-the-real-world-with-agent-skills)

# About This Repository

This repository contains skills that demonstrate what's possible with Claude's skills system. These skills range from creative applications (art, music, design) to technical tasks (testing web apps, MCP server generation) to enterprise workflows (communications, branding, etc.).

Each skill is self-contained in its own folder with a `SKILL.md` file containing the instructions and metadata that Claude uses. Browse through these skills to get inspiration for your own skills or to understand different patterns and approaches.

Many skills in this repo are open source (Apache 2.0). We've also included the document creation & editing skills that power [Claude's document capabilities](https://www.anthropic.com/news/create-files) under the hood in the [`skills/docx`](./skills/docx), [`skills/pdf`](./skills/pdf), [`skills/pptx`](./skills/pptx), and [`skills/xlsx`](./skills/xlsx) subfolders. These are source-available, not open source, but we wanted to share these with developers as a reference for more complex skills that are actively used in a production AI application.

## Disclaimer

**These skills are provided for demonstration and educational purposes only.** While some of these capabilities may be available in Claude, the implementations and behaviors you receive from Claude may differ from what is shown in these skills. These skills are meant to illustrate patterns and possibilities. Always test skills thoroughly in your own environment before relying on them for critical tasks.

# Skill Sets
- [./skills](./skills): Skill examples for Creative & Design, Development & Technical, Enterprise & Communication, and Document Skills
- [./spec](./spec): The Agent Skills specification
- [./template](./template): Skill template

# Try in Claude Code, Claude.ai, and the API

## Claude Code
You can register this repository as a Claude Code Plugin marketplace by running the following command in Claude Code:
```
/plugin marketplace add anthropics/skills
```

Then, to install a specific set of skills:
1. Select `Browse and install plugins`
2. Select `anthropic-agent-skills`
3. Select `document-skills` or `example-skills`
4. Select `Install now`

Alternatively, directly install either Plugin via:
```
/plugin install document-skills@anthropic-agent-skills
/plugin install example-skills@anthropic-agent-skills
```

After installing the plugin, you can use the skill by just mentioning it. For instance, if you install the `document-skills` plugin from the marketplace, you can ask Claude Code to do something like: "Use the PDF skill to extract the form fields from `path/to/some-file.pdf`"

## Claude.ai

These example skills are all already available to paid plans in Claude.ai. 

To use any skill from this repository or upload custom skills, follow the instructions in [Using skills in Claude](https://support.claude.com/en/articles/12512180-using-skills-in-claude#h_a4222fa77b).

## Claude API

You can use Anthropic's pre-built skills, and upload custom skills, via the Claude API. See the [Skills API Quickstart](https://docs.claude.com/en/api/skills-guide#creating-a-skill) for more.

# Creating a Basic Skill

Skills are simple to create - just a folder with a `SKILL.md` file containing YAML frontmatter and instructions. You can use the **template-skill** in this repository as a starting point:

```markdown
---
name: my-skill-name
description: A clear description of what this skill does and when to use it
---

# My Skill Name

[Add your instructions here that Claude will follow when this skill is active]

## Examples
- Example usage 1
- Example usage 2

## Guidelines
- Guideline 1
- Guideline 2
```

The frontmatter requires only two fields:
- `name` - A unique identifier for your skill (lowercase, hyphens for spaces)
- `description` - A complete description of what the skill does and when to use it

The markdown content below contains the instructions, examples, and guidelines that Claude will follow. For more details, see [How to create custom skills](https://support.claude.com/en/articles/12512198-creating-custom-skills).

# Partner Skills

Skills are a great way to teach Claude how to get better at using specific pieces of software. As we see awesome example skills from partners, we may highlight some of them here:

- **Notion** - [Notion Skills for Claude](https://www.notion.so/notiondevs/Notion-Skills-for-Claude-28da4445d27180c7af1df7d8615723d0)

---

### Overview (`skills/claude-api/csharp/claude-api/README.md`)

# Claude API - C#

> **Note:** The C# SDK is the official Anthropic SDK for C#. Tool use is supported via the Messages API with a beta `BetaToolRunner` for automatic tool execution loops. The SDK also supports Microsoft.Extensions.AI IChatClient integration with function invocation and Managed Agents (beta).

## Namespace Reference

Types are organized by namespace. If a type you need isn't shown in an example below, locate it via this table first - don't block on fetching SDK source over the network.

| `using` | Contains |
|---|---|
| `Anthropic` | `AnthropicClient`, top-level options |
| `Anthropic.Models.Messages` | non-beta request/response types - `MessageCreateParams`, `Model`, `Role`, `ContentBlock`, `TextBlock`, `ToolUseBlock`, `ToolResultBlockParam`, `Tool*` (tool definition classes) |
| `Anthropic.Models.Beta.Messages` | beta-endpoint equivalents - `MessageCreateParams`, `BetaMessage`, `BetaTool*`, `Speed`, `BetaRequestMcpServerUrlDefinition`, context-editing/compaction configs |
| `Anthropic.Models.Beta` | shared beta constants |
| `Anthropic.Models.Beta.Files` | Files API types |
| `Anthropic.Models.Messages.Batches` | Batch API types |
| `Anthropic.Helpers.Beta` | `BetaToolRunner`, beta helper utilities |
| `Anthropic.Exceptions` | `AnthropicApiException`, `AnthropicRateLimitException`, `Anthropic5xxException`, etc. - see `shared/error-codes.md` |
| `Anthropic.Bedrock` / `Anthropic.Vertex` / `Anthropic.Foundry` / `Anthropic.Aws` | platform clients (separate NuGet packages): `AnthropicBedrockMantleClient`, `AnthropicFoundryClient`, `AnthropicAwsClient` |

`client.Messages.*` uses non-beta types; `client.Beta.Messages.*` uses the `Anthropic.Models.Beta.Messages` types. Both namespaces define a `MessageCreateParams` - pick the one matching the client path you call.

### Key types per feature

Write from this table instead of reflecting the SDK assembly. Endpoint column tells you whether to use `client.Messages.*` or `client.Beta.Messages.*`.

| Feature | Endpoint | Key C# types (namespace per table above) |
|---|---|---|
| User profiles | beta | `client.Beta.UserProfiles.Create(...)` / `.Retrieve(id)` / `.List()`. Pass the returned profile id on the beta messages call. Requires a beta header - check the SDK's beta-headers reference for the current flag. |
| Agent Skills | beta | `BetaContainerParams` (with `Skills = [new BetaSkillParams { ... }]`), `BetaCodeExecutionTool20250825`. `Betas = ["code-execution-2025-08-25"]` (Skills is out of beta - no `skills-2025-10-02`). Download the output via `client.Beta.Files.Download(fileId)`. |
| Advisor tool | beta | `BetaAdvisorTool20260301` - may not be in all SDK releases yet |
| Cache diagnostics | beta | `Diagnostics = new() { PreviousMessageID = ... }`, `BetaCacheControlEphemeral`, `BetaContentBlockParam` |
| Context editing | beta | `ContextManagement = new BetaContextManagementConfig { Edits = [new BetaClearToolUses20250919Edit()] }`. `Betas = ["context-management-2025-06-27"]` (not `compact-2026-01-12` - that's for `BetaCompact20260112Edit`). |
| Memory tool | non-beta | `Tools = [new ToolUnion(new MemoryTool20250818())]` |
| Programmatic tool calling | non-beta | `CodeExecutionTool20260120`, `ToolResultBlockParam`, `ContentBlockParam` |
| Task budgets | beta | `BetaOutputConfig` with `TaskBudget = new BetaTokenTaskBudget { ... }` |
| Tool search | non-beta | `new ToolUnion(new ToolSearchToolRegex20251119 { Type = ToolSearchToolRegex20251119Type.ToolSearchToolRegex20251119 })` - `Type` must be set explicitly. |
| Web search | non-beta | `new ToolUnion(new WebSearchTool20260209())` - the latest variant with dynamic filtering (Claude Fable 5.1 + Claude Opus 5.5 + Claude Opus 5 + Opus 4.8/4.7/4.6 + Claude Sonnet 5.5 + Claude Sonnet 5 + Sonnet 4.6). For older models or Vertex, use `WebSearchTool20250305()` |

### Discovering type and member names

If a type or member you need isn't in the tables above, `strings ~/.nuget/packages/anthropic/*/lib/*/Anthropic.dll | grep -i <term>` is fast and sufficient for locating class and property names. **Do not escalate to a `dotnet run` reflection probe** to dump members precisely - the first compile is slow enough to be backgrounded in many environments, trapping you in a polling loop. Instead, write `Program.cs` using the names `strings | grep` found; if a member name is wrong the compiler error (`error CS1061: 'X' does not contain a definition for 'Y'`) points at it in a few seconds, faster than any reflection probe.

Note that `strings` will not surface wire-format snake_case field names (`output_tokens`, `stop_reason`) - those are stored in the DLL differently. **C# properties are the PascalCase equivalent of the wire field** (`response.Usage.OutputTokens`, `response.StopReason`). If you know the wire field name from the docs, write the PascalCase property and compile; do not probe for the snake_case string.

### Minimal working skeleton

**Write a plain `Program.cs` body** - `using` statements followed by top-level statements, as below. Do **not** add a `#!/usr/bin/env dotnet` shebang or `#:package Anthropic@*` directive: those are .NET file-based-app syntax and fail with `CS1024: Preprocessor directive expected` when the file is compiled via an existing `.csproj`. The standard project setup (per the [C# quickstart](https://platform.claude.com/docs/en/get-started): `dotnet new console` -> `dotnet add package Anthropic` -> edit `Program.cs` -> `dotnet run`) provides the `.csproj` and package reference.

Start from this - it compiles as-is. Fill in the feature-specific fields; do not spend turns running reflection or XML-doc inspection to discover type names first.

```csharp
using System;
using Anthropic;
using Anthropic.Models.Messages;       // or Anthropic.Models.Beta.Messages for beta endpoints

AnthropicClient client = new();

var message = await client.Messages.Create(new MessageCreateParams
{
    Model = "claude-opus-5-5",
    MaxTokens = 1024,
    Messages = [ new() { Role = Role.User, Content = "Hello, Claude" } ],
});

Console.WriteLine(message);
```

For beta features (anything behind an `anthropic-beta` header), use the beta client path and namespace - same overall shape:

```csharp
using System;
using Anthropic;
using Anthropic.Models.Beta.Messages;

AnthropicClient client = new();

var response = await client.Beta.Messages.Create(new MessageCreateParams
{
    Model = "claude-opus-5-5",
    MaxTokens = 4096,
    Betas = ["<beta-flag>"],
    Messages = [ new() { Role = Role.User, Content = "..." } ],
    // Tools = new BetaToolUnion[] { new BetaSomeTool { ... } },   // for tool features
});

Console.WriteLine(response);
```

If a type name the feature needs isn't in this file, write it following the naming pattern in the Namespace Reference above and fix from compiler output - producing a `Program.cs` and iterating beats researching.

### Common C# compile errors

- **CS8803 (top-level statements must precede type declarations):** put any `record`/`class`/`struct` definitions **after** the last top-level statement, at the end of the file. A record defined above `var client = new AnthropicClient()` will not compile.
- **`await foreach` on a `Task<...Page>`:** `client.Models.List()` returns a `Task<ModelListPage>`, which is not directly async-enumerable. Await it first, then iterate: `var page = await client.Models.List(); foreach (var m in page.Items) {...}`. For auto-pagination, check whether the page type exposes `AutoPagingEachAsync()` or similar before reaching for `await foreach`.

## Installation

```bash
dotnet add package Anthropic
```

## Client Initialization

```csharp
using Anthropic;

// Default (uses ANTHROPIC_API_KEY env var)
AnthropicClient client = new();

// Explicit API key (use environment variables - never hardcode keys)
AnthropicClient client = new() {
    ApiKey = Environment.GetEnvironmentVariable("ANTHROPIC_API_KEY")
};
```

---

## Basic Message Request

```csharp
using Anthropic.Models.Messages;

var parameters = new MessageCreateParams
{
    Model = "claude-opus-5-5",
    MaxTokens = 16000,
    Messages = [new() { Role = Role.User, Content = "What is the capital of France?" }]
};
var response = await client.Messages.Create(parameters);

// ContentBlock is a union wrapper. .Value unwraps to the variant object,
// then OfType<T> filters to the type you want. Or use the TryPick* idiom
// shown in the Thinking section below.
foreach (var text in response.Content.Select(b => b.Value).OfType<TextBlock>())
{
    Console.WriteLine(text.Text);
}
```

---

## Thinking

**Adaptive thinking is the recommended mode for Claude 4.6+ models.** Claude decides dynamically when and how much to think.

> **Fable 5, Claude Opus 5.5, Claude Opus 5, Opus 4.8, Opus 4.7, Opus 4.6, and Sonnet 4.6:** Use adaptive thinking (below). `new ThinkingConfigEnabled { BudgetTokens = N }` is removed on Fable 5, Claude Opus 5.5, Claude Opus 5, Opus 4.8, and 4.7 (400 if sent); deprecated on Opus 4.6 and Sonnet 4.6.
> **Claude Opus 5.5:** thinking is always on - omit `Thinking` (or send `ThinkingConfigAdaptive`, which is equivalent); `ThinkingConfigDisabled` returns a 400 at every effort, as does a thinking budget. Control depth with `OutputConfig.Effort` instead - the default is `medium` on this model, where Claude Opus 5 defaults to `high`.
> **Claude Opus 5:** thinking is on by default - omitting `Thinking` runs adaptive (`ThinkingConfigAdaptive` is equivalent), unlike Opus 4.8/4.7 where omitting it meant no thinking. `ThinkingConfigDisabled` is accepted only at effort `high` or lower; pairing it with `xhigh`/`max` returns a 400.
> **Older models:** Use `new ThinkingConfigEnabled { BudgetTokens = N }` (budget must be < `MaxTokens`, min 1024).

```csharp
using Anthropic.Models.Messages;

var response = await client.Messages.Create(new MessageCreateParams
{
    Model = "claude-opus-5-5",
    MaxTokens = 16000,
    // ThinkingConfigParam? implicitly converts from the concrete variant classes -
    // no wrapper needed.
    // display opt-in: default is omitted (empty thinking text) on Fable 5/5.1, Mythos 5/5.1, Claude Opus 5.5, Claude Opus 5, Opus 4.8/4.7, Claude Sonnet 5.5, and Claude Sonnet 5
    Thinking = new ThinkingConfigAdaptive { Display = Display.Summarized },
    Messages =
    [
        new() { Role = Role.User, Content = "Solve: 27 * 453" },
    ],
});

// ThinkingBlock(s) precede TextBlock in Content. TryPick* narrows the union.
foreach (var block in response.Content)
{
    if (block.TryPickThinking(out ThinkingBlock? t))
    {
        Console.WriteLine($"[thinking] {t.Thinking}");
    }
    else if (block.TryPickText(out TextBlock? text))
    {
        Console.WriteLine(text.Text);
    }
}
```

Alternative to `TryPick*`: `.Select(b => b.Value).OfType<ThinkingBlock>()` (same LINQ pattern as the Basic Message example).

---

## Context Editing / Compaction (Beta)

**Beta-namespace prefix is inconsistent** (source-verified against `src/Anthropic/Models/Beta/Messages/*.cs` @ 12.9.0). No prefix: `MessageCreateParams`, `MessageCountTokensParams`, `Role`, `Speed`. **Everything else has the `Beta` prefix**: `BetaMessageParam`, `BetaMessage`, `BetaContentBlock`, `BetaToolUseBlock`, all block param types. The unprefixed `Role` WILL collide with `Anthropic.Models.Messages.Role` if you import both namespaces (CS0104). Safest: import only Beta; if mixing, alias the beta `Role`:

```csharp
using Anthropic.Models.Beta.Messages;
using NonBeta = Anthropic.Models.Messages;  // only if you also need non-beta types
// Now: MessageCreateParams, BetaMessageParam, Role (beta's), NonBeta.Role (if needed)
```


`BetaMessage.Content` is `IReadOnlyList<BetaContentBlock>` - a 15-variant discriminated union. Narrow with `TryPick*`. **Response `BetaContentBlock` is NOT assignable to param `BetaContentBlockParam`** - there's no `.ToParam()` in C#. Round-trip by converting each block:

```
/* Detailed source-code truncated for AI context efficiency. */
```

All 15 `BetaContentBlock.TryPick*` variants: `Text`, `Thinking`, `RedactedThinking`, `ToolUse`, `ServerToolUse`, `WebSearchToolResult`, `WebFetchToolResult`, `CodeExecutionToolResult`, `BashCodeExecutionToolResult`, `TextEditorCodeExecutionToolResult`, `ToolSearchToolResult`, `McpToolUse`, `McpToolResult`, `ContainerUpload`, `Compaction`.

**`BetaToolUseBlock.Input` is `IReadOnlyDictionary<string, JsonElement>`** - index by key then call the `JsonElement` extractor:

```csharp
if (block.TryPickToolUse(out BetaToolUseBlock? tu))
{
    int a = tu.Input["a"].GetInt32();
    string s = tu.Input["name"].GetString()!;
}
```

---

## Effort Parameter

Effort is nested under `OutputConfig`, NOT a top-level property. `ApiEnum<string, Effort>` has an implicit conversion from the enum, so assign `Effort.High` directly.

```csharp
OutputConfig = new OutputConfig { Effort = Effort.High },
```

Values: `Effort.Low`, `Effort.Medium`, `Effort.High`, `Effort.Max`. Combine with `Thinking = new ThinkingConfigAdaptive()` for cost-quality control.

---

## Prompt Caching

`System` takes `MessageCreateParamsSystem?` - a union of `string` or `List<TextBlockParam>`. There is no `SystemTextBlockParam`; use plain `TextBlockParam`. The implicit conversion needs the concrete `List<TextBlockParam>` type (array literals won't convert). For placement patterns and the silent-invalidator audit checklist, see `shared/prompt-caching.md`.

```csharp
System = new List<TextBlockParam> {
    new() {
        Text = longSystemPrompt,
        CacheControl = new CacheControlEphemeral(),  // auto-sets Type = "ephemeral"
    },
},
```

Optional `Ttl` on `CacheControlEphemeral`: `new() { Ttl = Ttl.Ttl1h }` or `Ttl.Ttl5m`. `CacheControl` also exists on `Tool.CacheControl` and top-level `MessageCreateParams.CacheControl`.

Verify hits via `response.Usage.CacheCreationInputTokens` / `response.Usage.CacheReadInputTokens`.

---

## Token Counting

```csharp
MessageTokensCount result = await client.Messages.CountTokens(new MessageCountTokensParams {
    Model = "claude-opus-5-5",
    Messages = [new() { Role = Role.User, Content = "Hello" }],
});
long tokens = result.InputTokens;
```

`MessageCountTokensParams.Tools` uses a different union type (`MessageCountTokensTool`) than `MessageCreateParams.Tools` (`ToolUnion`) - if you're passing tools, the compiler will tell you when it matters.

---

## PDF / Document Input

`DocumentBlockParam` takes a `DocumentBlockParamSource` union: `Base64PdfSource` / `UrlPdfSource` / `PlainTextSource` / `ContentBlockSource`. `Base64PdfSource` auto-sets `MediaType = "application/pdf"` and `Type = "base64"`.

```csharp
new MessageParam {
    Role = Role.User,
    Content = new List<ContentBlockParam> {
        new DocumentBlockParam { Source = new Base64PdfSource { Data = base64String } },
        new TextBlockParam { Text = "Summarize this PDF" },
    },
}
```

---

## Fast Mode (Beta)

```csharp
var response = await client.Beta.Messages.Create(new MessageCreateParams {
    Model = "claude-opus-5-5", MaxTokens = 4096,
    Speed = Speed.Fast,
    Betas = ["fast-mode-2026-02-01"],
    Messages = [...],
});
```

## Models API

```csharp
var page = await client.Models.List();
foreach (var m in page.Items) { Console.WriteLine(m.ID); }
var one = await client.Models.Retrieve("claude-opus-5-5");
```

## Long Output (128k) + Prefill

Set `MaxTokens = 128000` on `client.Messages` and use the streaming path (see `streaming.md`). On Claude 4+ models, 128k output is native - no `output-128k-*` beta header or beta namespace is needed.

**Prefilling the assistant message** (putting a trailing `Role.Assistant` message in the input) is **not supported** on Claude Fable 5.1, Claude Opus 5.5, Claude Opus 5, Opus 4.8, Opus 4.7, Opus 4.6, Sonnet 4.6, and Claude Haiku 5.5 - requests return a 400. Use `OutputConfig.Format` (structured outputs) instead.

## Stop Details

When `StopReason` is `"refusal"`, the response includes structured `StopDetails`:

```csharp
if (response.StopReason == "refusal" && response.StopDetails is { } details)
{
    Console.WriteLine($"Category: {details.Category}");
    Console.WriteLine($"Explanation: {details.Explanation}");
}
```

**Refusal fallbacks (Claude Fable 5.1) - opt in by default.** Fallbacks are opt-in: without them a refused request simply stops. In `claude-fable-5-1` code, include `Fallbacks = [new(Model.ClaudeOpus4_8)]` plus the `AnthropicBeta.ServerSideFallback2026_06_01` beta on the beta messages call by default - a policy decline is re-served by the fallback model inside the same call. Not available on Amazon Bedrock, Vertex AI, or Microsoft Foundry - use the client-side handler there: `new AnthropicClient { Handlers = [new BetaRefusalFallbackHandler { Fallbacks = [new(Model.ClaudeOpus4_8)] }] }` (namespace `Anthropic.Helpers`), with per-conversation state via `BetaFallbackState.Create()` scoped with `using (fallbackState.Use()) { ... }`. Full semantics (billing, sticky routing, streaming) and a runnable example: `shared/model-migration.md` -> Migrating to Claude Fable 5.1 -> `refusal` stop reason, and the C# SDK repo's `examples/` (WebFetch via `shared/live-sources.md`).

---

## Managed Agents (Beta)

The C# SDK supports Managed Agents via `client.Beta.Agents`, `client.Beta.Sessions`, `client.Beta.Environments`, and related namespaces. See `shared/managed-agents-overview.md` for the architecture and `curl/managed-agents.md` for the wire-level reference.

---

### Overview (`skills/claude-api/go/claude-api/README.md`)

# Claude API - Go

> **Note:** The Go SDK supports the Claude API and beta tool use with `BetaToolRunner`. Agent SDK is not yet available for Go.

## Installation

```bash
go get github.com/anthropics/anthropic-sdk-go
```

## Client Initialization

```go
import (
    "github.com/anthropics/anthropic-sdk-go"
    "github.com/anthropics/anthropic-sdk-go/option"
)

// Default (uses ANTHROPIC_API_KEY env var)
client := anthropic.NewClient()

// Explicit API key
client := anthropic.NewClient(
    option.WithAPIKey("your-api-key"),
)
```

---

## Model IDs

`anthropic.Model` is an alias for `string`, so pass the model as its plain id: `Model: "claude-opus-5-5"`. Default to Claude Opus 5.5 unless the user specifies otherwise; if they ask for Fable or the most powerful model, use `"claude-fable-5-1"`; if they ask for a cheaper tier, use the current generation - `"claude-sonnet-5-5"` or `"claude-haiku-5-5"` (see `shared/models.md` for the full resolution table).

The SDK also ships typed `anthropic.ModelClaude*` constants, but they lag model launches - a given SDK release may have constants only for previous-generation models. Do not pick a model because it has a typed constant; the string id works for every model on every SDK version. Check the SDK release notes before assuming a typed constant exists for a current model.

---

## Basic Message Request

```go
response, err := client.Messages.New(context.Background(), anthropic.MessageNewParams{
    Model:     "claude-opus-5-5",
    MaxTokens: 16000,
    Messages: []anthropic.MessageParam{
        anthropic.NewUserMessage(anthropic.NewTextBlock("What is the capital of France?")),
    },
})
if err != nil {
    log.Fatal(err)
}
for _, block := range response.Content {
    switch variant := block.AsAny().(type) {
    case anthropic.TextBlock:
        fmt.Println(variant.Text)
    }
}
```

---

## Thinking

Enable Claude's internal reasoning by setting `Thinking` in `MessageNewParams`. The response will contain `ThinkingBlock` content before the final `TextBlock`.

**Adaptive thinking is the recommended mode for Claude 4.6+ models.** Claude decides dynamically when and how much to think. Combine with the `effort` parameter for cost-quality control.

Derived from `anthropic-sdk-go/message.go` (`ThinkingConfigParamUnion`, `ThinkingConfigAdaptiveParam`).

```go
// There is no ThinkingConfigParamOfAdaptive helper - construct the union
// struct-literal directly and take the address of the variant.
// display opt-in: default is omitted (empty thinking text) on Fable 5/5.1, Mythos 5/5.1, Claude Opus 5.5, Claude Opus 5, Opus 4.8/4.7, Claude Sonnet 5.5, and Claude Sonnet 5
adaptive := anthropic.ThinkingConfigAdaptiveParam{Display: anthropic.ThinkingConfigAdaptiveDisplaySummarized}
params := anthropic.MessageNewParams{
    Model:     "claude-opus-5-5",
    MaxTokens: 16000,
    Thinking:  anthropic.ThinkingConfigParamUnion{OfAdaptive: &adaptive},
    Messages: []anthropic.MessageParam{
        anthropic.NewUserMessage(anthropic.NewTextBlock("How many r's in strawberry?")),
    },
}

resp, err := client.Messages.New(context.Background(), params)
if err != nil {
    log.Fatal(err)
}

// ThinkingBlock(s) precede TextBlock in content
for _, block := range resp.Content {
    switch b := block.AsAny().(type) {
    case anthropic.ThinkingBlock:
        fmt.Println("[thinking]", b.Thinking)
    case anthropic.TextBlock:
        fmt.Println(b.Text)
    }
}
```

> **Fable 5, Claude Opus 5.5, Claude Opus 5, Opus 4.8, Opus 4.7, Opus 4.6, and Sonnet 4.6:** Use adaptive thinking (above). `ThinkingConfigParamOfEnabled(budgetTokens)` is removed on Fable 5, Claude Opus 5.5, Claude Opus 5, Opus 4.8, and 4.7 (400 if sent); deprecated on Opus 4.6 and Sonnet 4.6.
> **Claude Opus 5.5:** thinking is always on - leave `Thinking` unset (or send the adaptive union, which is equivalent); `OfDisabled` returns a 400 at every effort, as does `ThinkingConfigParamOfEnabled`. Control depth with `Effort` under `OutputConfig` instead - the default is `medium` on this model, where Claude Opus 5 defaults to `high`.
> **Claude Opus 5:** thinking is on by default - leaving `Thinking` unset runs adaptive (the adaptive union is equivalent), unlike Opus 4.8/4.7 where leaving it unset meant no thinking.
> **Older models:** Use `anthropic.ThinkingConfigParamOfEnabled(N)` (budget must be < `MaxTokens`, min 1024).

To disable: `anthropic.ThinkingConfigParamUnion{OfDisabled: &anthropic.ThinkingConfigDisabledParam{}}`. On Claude Opus 5 that is accepted only at effort `high` or lower - pairing it with `xhigh`/`max` returns a 400; on Claude Opus 5.5 it returns a 400 at every effort (lower `Effort` instead).

---

## Prompt Caching

`System` is `[]TextBlockParam`; set `CacheControl` on the last block to cache tools + system together. For placement patterns and the silent-invalidator audit checklist, see `shared/prompt-caching.md`.

```go
System: []anthropic.TextBlockParam{{
    Text:         longSystemPrompt,
    CacheControl: anthropic.NewCacheControlEphemeralParam(), // default 5m TTL
}},
```

For 1-hour TTL: `anthropic.CacheControlEphemeralParam{TTL: anthropic.CacheControlEphemeralTTLTTL1h}`. There's also a top-level `CacheControl` on `MessageNewParams` that auto-places on the last cacheable block.

Verify hits via `resp.Usage.CacheCreationInputTokens` / `resp.Usage.CacheReadInputTokens`.

---

## Stop Details

When `StopReason` is `anthropic.StopReasonRefusal`, the response includes structured `StopDetails`:

```go
if resp.StopReason == anthropic.StopReasonRefusal {
    fmt.Println("Category:", resp.StopDetails.Category)     // e.g. "cyber", "bio", "reasoning_extraction", "frontier_llm", or "" - see docs for the full set
    fmt.Println("Explanation:", resp.StopDetails.Explanation)
}
```

**Refusal fallbacks (Claude Fable 5.1) - opt in by default.** Fallbacks are opt-in: without them a refused request simply stops. In `claude-fable-5-1` code, include `Fallbacks: []anthropic.BetaFallbackParam{{Model: "claude-opus-4-8"}}` plus the `anthropic.AnthropicBetaServerSideFallback2026_06_01` beta on `client.Beta.Messages.New` by default - a policy decline is re-served by the fallback model inside the same call. Not available on Amazon Bedrock, Vertex AI, or Microsoft Foundry - register the client-side middleware there: `option.WithMiddleware(betafallback.BetaRefusalFallbackMiddleware(...))` from `lib/betafallback`, with per-conversation state via `betafallback.WithBetaFallbackState(&betafallback.BetaFallbackState{})`. Full semantics (billing, sticky routing, streaming) and a runnable example: `shared/model-migration.md` -> Migrating to Claude Fable 5.1 -> `refusal` stop reason, and the Go SDK repo's `examples/` (WebFetch via `shared/live-sources.md`).

---

## PDF / Document Input

`NewDocumentBlock` generic helper accepts any source type. `MediaType`/`Type` are auto-set.

```go
b64 := base64.StdEncoding.EncodeToString(pdfBytes)

msg := anthropic.NewUserMessage(
    anthropic.NewDocumentBlock(anthropic.Base64PDFSourceParam{Data: b64}),
    anthropic.NewTextBlock("Summarize this document"),
)
```

Other sources: `URLPDFSourceParam{URL: "https://..."}`, `PlainTextSourceParam{Data: "..."}`.

---

## Context Editing / Compaction (Beta)

Use `Beta.Messages.New` with `ContextManagement` on `BetaMessageNewParams`. There is no `NewBetaAssistantMessage` - use `.ToParam()` for the round-trip.

```go
params := anthropic.BetaMessageNewParams{
    Model:     "claude-opus-5-5",
    MaxTokens: 16000,
    Betas:     []anthropic.AnthropicBeta{"compact-2026-01-12"},
    ContextManagement: anthropic.BetaContextManagementConfigParam{
        Edits: []anthropic.BetaContextManagementConfigEditUnionParam{
            {OfCompact20260112: &anthropic.BetaCompact20260112EditParam{}},
        },
    },
    Messages: []anthropic.BetaMessageParam{ /* ... */ },
}

resp, err := client.Beta.Messages.New(ctx, params)
if err != nil {
    log.Fatal(err)
}

// Round-trip: append response to history via .ToParam()
params.Messages = append(params.Messages, resp.ToParam())

// Read compaction blocks from the response
for _, block := range resp.Content {
    if c, ok := block.AsAny().(anthropic.BetaCompactionBlock); ok {
        fmt.Println("compaction summary:", c.Content)
    }
}
```

Other edit types: `BetaClearToolUses20250919EditParam`, `BetaClearThinking20251015EditParam` - these need `Betas: []anthropic.AnthropicBeta{"context-management-2025-06-27"}`, not `compact-2026-01-12`.

---

### Overview (`skills/claude-api/go/managed-agents/README.md`)

# Managed Agents - Go

> **Bindings not shown here:** This README covers the most common managed-agents flows for Go. If you need a class, method, namespace, field, or behavior that isn't shown, WebFetch the Go SDK repo **or the relevant docs page** from `shared/live-sources.md` rather than guess. Do not extrapolate from cURL shapes or another language's SDK.

> **Agents are persistent - create once, reference by ID.** Store the agent ID returned by `agents.New` and pass it to every subsequent `sessions.New`; do not call `agents.New` in the request path. **Recommended:** define agents and environments as version-controlled files synced with `ant apply` - see `shared/anthropic-cli.md` (its live-docs URL is in `shared/live-sources.md`). The CLI owns the control plane (create/update); your code owns the data plane (sessions with the stored ID). The examples below show in-code creation for when you must provision programmatically; in production the create call belongs in setup, not in the request path.

## Installation

```bash
go get github.com/anthropics/anthropic-sdk-go
```

## Client Initialization

```go
import (
    "context"

    "github.com/anthropics/anthropic-sdk-go"
    "github.com/anthropics/anthropic-sdk-go/option"
)

// Default (uses ANTHROPIC_API_KEY env var)
client := anthropic.NewClient()

// Explicit API key
client := anthropic.NewClient(
    option.WithAPIKey("your-api-key"),
)

ctx := context.Background()
```

---

## Create an Environment

```go
environment, err := client.Beta.Environments.New(ctx, anthropic.BetaEnvironmentNewParams{
    Name: "my-dev-env",
    Config: anthropic.BetaEnvironmentNewParamsConfigUnion{
        OfCloud: &anthropic.BetaCloudConfigParams{
            Networking: anthropic.BetaCloudConfigParamsNetworkingUnion{
                OfLimited: &anthropic.BetaLimitedNetworkParams{
                    AllowPackageManagers: anthropic.Bool(true),
                    AllowMCPServers:      anthropic.Bool(true),
                },
            },
        },
    },
})
if err != nil {
    panic(err)
}
fmt.Println(environment.ID) // env_...
```

---

## Create an Agent (required first step)

> Warning: **There is no inline agent config.** `Model`/`System`/`Tools` live on the agent object, not the session. Always start with `Beta.Agents.New()` - the session only takes `Agent: anthropic.BetaSessionNewParamsAgentUnion{OfString: anthropic.String(agent.ID)}` (or the typed `OfBetaManagedAgentsAgents` variant when you need a specific version).

### Minimal

The examples on this page turn both web tools off. Set `enabled` to true on `web_fetch` / `web_search` only when the job as described needs the web (a general-purpose or open-ended job stays off; tell the user how to switch it on) - see `shared/managed-agents-tools.md` § Agent Toolset. They also set the `auto` permission policy (Go SDK 1.72.0+), under which a call can pause for your approval - the event loop under Stream Events answers it. When nobody is watching the run, answer `deny`; never answer `allow` to every paused call.

```
/* Detailed source-code truncated for AI context efficiency. */
```

### Updating an Agent

Updates create new versions; the agent object is immutable per version.

```go
updatedAgent, err := client.Beta.Agents.Update(ctx, agent.ID, anthropic.BetaAgentUpdateParams{
    Version: agent.Version,
    System:  anthropic.String("You are a helpful coding agent. Always write tests."),
})
if err != nil {
    panic(err)
}
fmt.Printf("New version: %d\n", updatedAgent.Version)

// List all versions
iter := client.Beta.Agents.Versions.ListAutoPaging(ctx, agent.ID, anthropic.BetaAgentVersionListParams{})
for iter.Next() {
    version := iter.Current()
    fmt.Printf("Version %d: %s\n", version.Version, version.UpdatedAt.Format(time.RFC3339))
}
if err := iter.Err(); err != nil {
    panic(err)
}

// Archive the agent
_, err = client.Beta.Agents.Archive(ctx, agent.ID, anthropic.BetaAgentArchiveParams{})
if err != nil {
    panic(err)
}
```

---

## Send a User Message

```go
_, err = client.Beta.Sessions.Events.Send(ctx, session.ID, anthropic.BetaSessionEventSendParams{
    Events: []anthropic.BetaManagedAgentsEventParamsUnion{{
        OfUserMessage: &anthropic.BetaManagedAgentsUserMessageEventParams{
            Type: anthropic.BetaManagedAgentsUserMessageEventParamsTypeUserMessage,
            Content: []anthropic.BetaManagedAgentsUserMessageEventParamsContentUnion{{
                OfText: &anthropic.BetaManagedAgentsTextBlockParam{
                    Type: anthropic.BetaManagedAgentsTextBlockTypeText,
                    Text: "Review the auth module",
                },
            }},
        },
    }},
})
if err != nil {
    panic(err)
}
```

> Tip: **Stream-first:** Open the stream *before* (or concurrently with) sending the message. The stream only delivers events that occur after it opens - stream-after-send means early events arrive buffered in one batch. See [Steering Patterns](../../shared/managed-agents-events.md#steering-patterns).

---

## Stream Events (SSE)

```
/* Detailed source-code truncated for AI context efficiency. */
```

### Reconnecting and Tailing

When reconnecting mid-session, list past events first to dedupe, then tail live events. Answer paused calls as in the loop above, including an `ask` in the history that no `user.tool_confirmation` follows.

```go
stream := client.Beta.Sessions.Events.StreamEvents(ctx, session.ID, anthropic.BetaSessionEventStreamParams{})
defer stream.Close()

// Stream is open and buffering. List history before tailing live.
seenEventIDs := map[string]struct{}{}
history := client.Beta.Sessions.Events.ListAutoPaging(ctx, session.ID, anthropic.BetaSessionEventListParams{})
for history.Next() {
    seenEventIDs[history.Current().ID] = struct{}{}
}
if err := history.Err(); err != nil {
    panic(err)
}

// Tail live events, skipping anything already seen
tail:
for stream.Next() {
    event := stream.Current()
    if _, seen := seenEventIDs[event.ID]; seen {
        continue
    }
    seenEventIDs[event.ID] = struct{}{}
    switch event := event.AsAny().(type) {
    case anthropic.BetaManagedAgentsAgentMessageEvent:
        for _, block := range event.Content {
            fmt.Print(block.Text)
        }
    case anthropic.BetaManagedAgentsSessionStatusIdleEvent:
        if event.StopReason.Type != "requires_action" { // requires_action: answer the paused call as under Stream Events, then keep streaming
            break tail
        }
    }
}
if err := stream.Err(); err != nil {
    panic(err)
}
```

---

## Provide Custom Tool Result

> Note: The Go managed-agents bindings for `user.custom_tool_result` are not yet documented in this skill or in the apps source examples. Refer to `shared/managed-agents-events.md` for the wire format and the `github.com/anthropics/anthropic-sdk-go` repository for the corresponding Go params types.

---

## Poll Events

```go
// Auto-paginating iterator
iter := client.Beta.Sessions.Events.ListAutoPaging(ctx, session.ID, anthropic.BetaSessionEventListParams{})
for iter.Next() {
    event := iter.Current()
    fmt.Printf("%s: %s\n", event.Type, event.ID)
}
if err := iter.Err(); err != nil {
    panic(err)
}
```

---

## Upload a File

```go
csvFile, err := os.Open("data.csv")
if err != nil {
    panic(err)
}
defer csvFile.Close()

file, err := client.Beta.Files.Upload(ctx, anthropic.BetaFileUploadParams{
    File: csvFile,
})
if err != nil {
    panic(err)
}
fmt.Printf("File ID: %s\n", file.ID)

// Mount in a session
session, err := client.Beta.Sessions.New(ctx, anthropic.BetaSessionNewParams{
    Agent: anthropic.BetaSessionNewParamsAgentUnion{
        OfString: anthropic.String(agent.ID),
    },
    EnvironmentID: environment.ID,
    Resources: []anthropic.BetaSessionNewParamsResourceUnion{{
        OfFile: &anthropic.BetaManagedAgentsFileResourceParams{
            Type:      anthropic.BetaManagedAgentsFileResourceParamsTypeFile,
            FileID:    file.ID,
            MountPath: anthropic.String("/data.csv"),
        },
    }},
})
if err != nil {
    panic(err)
}
```

### Add and Manage Resources on an Existing Session

```go
// Attach an additional file to an open session
resource, err := client.Beta.Sessions.Resources.Add(ctx, session.ID, anthropic.BetaSessionResourceAddParams{
    BetaManagedAgentsFileResourceParams: anthropic.BetaManagedAgentsFileResourceParams{
        Type:   anthropic.BetaManagedAgentsFileResourceParamsTypeFile,
        FileID: file.ID,
    },
})
if err != nil {
    panic(err)
}
fmt.Println(resource.ID) // "sesrsc_01ABC..."

// List resources on the session
listed, err := client.Beta.Sessions.Resources.List(ctx, session.ID, anthropic.BetaSessionResourceListParams{})
if err != nil {
    panic(err)
}
for _, entry := range listed.Data {
    fmt.Println(entry.ID, entry.Type)
}

// Detach a resource
if _, err := client.Beta.Sessions.Resources.Delete(ctx, resource.ID, anthropic.BetaSessionResourceDeleteParams{
    SessionID: session.ID,
}); err != nil {
    panic(err)
}
```

---

## List and Download Session Files

> Note: Listing and downloading files an agent wrote during a session is not yet documented for Go in this skill or in the apps source examples. See `shared/managed-agents-events.md` and the `github.com/anthropics/anthropic-sdk-go` repository for the `Beta.Files.List` and `Beta.Files.Download` Go params types.

---

## Session Management

```go
// List environments
environments, err := client.Beta.Environments.List(ctx, anthropic.BetaEnvironmentListParams{})
if err != nil {
    panic(err)
}

// Retrieve a specific environment
env, err := client.Beta.Environments.Get(ctx, environment.ID, anthropic.BetaEnvironmentGetParams{})
if err != nil {
    panic(err)
}

// Archive an environment (read-only, existing sessions continue)
_, err = client.Beta.Environments.Archive(ctx, environment.ID, anthropic.BetaEnvironmentArchiveParams{})
if err != nil {
    panic(err)
}

// Delete an environment (only if no sessions reference it)
_, err = client.Beta.Environments.Delete(ctx, environment.ID, anthropic.BetaEnvironmentDeleteParams{})
if err != nil {
    panic(err)
}

// Delete a session
_, err = client.Beta.Sessions.Delete(ctx, session.ID, anthropic.BetaSessionDeleteParams{})
if err != nil {
    panic(err)
}
```

---

## MCP Server Integration

```
/* Detailed source-code truncated for AI context efficiency. */
```

See `shared/managed-agents-tools.md` §Vaults for creating vaults and adding credentials.

---

## Vaults

```
/* Detailed source-code truncated for AI context efficiency. */
```

---

## GitHub Repository Integration

Mount a GitHub repository as a session resource (a vault holds the GitHub MCP credential):

```go
session, err := client.Beta.Sessions.New(ctx, anthropic.BetaSessionNewParams{
    Agent:         anthropic.BetaSessionNewParamsAgentUnion{OfString: anthropic.String(agent.ID)},
    EnvironmentID: environment.ID,
    VaultIDs:      []string{vault.ID},
    Resources: []anthropic.BetaSessionNewParamsResourceUnion{
        {
            OfGitHubRepository: &anthropic.BetaManagedAgentsGitHubRepositoryResourceParams{
                Type:               anthropic.BetaManagedAgentsGitHubRepositoryResourceParamsTypeGitHubRepository,
                URL:                "https://github.com/org/repo",
                MountPath:          anthropic.String("/workspace/repo"),
                AuthorizationToken: "ghp_your_github_token",
            },
        },
    },
})
if err != nil {
    panic(err)
}
```

Multiple repositories on the same session:

```go
resources := []anthropic.BetaSessionNewParamsResourceUnion{
    {
        OfGitHubRepository: &anthropic.BetaManagedAgentsGitHubRepositoryResourceParams{
            Type:               anthropic.BetaManagedAgentsGitHubRepositoryResourceParamsTypeGitHubRepository,
            URL:                "https://github.com/org/frontend",
            MountPath:          anthropic.String("/workspace/frontend"),
            AuthorizationToken: "ghp_your_github_token",
        },
    },
    {
        OfGitHubRepository: &anthropic.BetaManagedAgentsGitHubRepositoryResourceParams{
            Type:               anthropic.BetaManagedAgentsGitHubRepositoryResourceParamsTypeGitHubRepository,
            URL:                "https://github.com/org/backend",
            MountPath:          anthropic.String("/workspace/backend"),
            AuthorizationToken: "ghp_your_github_token",
        },
    },
}
```

Rotating a repository's authorization token:

```go
listed, err := client.Beta.Sessions.Resources.List(ctx, session.ID, anthropic.BetaSessionResourceListParams{})
if err != nil {
    panic(err)
}
repoResourceID := listed.Data[0].ID

_, err = client.Beta.Sessions.Resources.Update(ctx, repoResourceID, anthropic.BetaSessionResourceUpdateParams{
    SessionID:          session.ID,
    AuthorizationToken: "ghp_your_new_github_token",
})
if err != nil {
    panic(err)
}
```

---

### Overview (`skills/claude-api/java/claude-api/README.md`)

# Claude API - Java

> **Note:** The Java SDK supports the Claude API and beta tool use with annotated classes. Agent SDK is not yet available for Java.

## Package Reference

Types are organized by package. If a class you need isn't shown in an example below, locate it via this table first - don't block on fetching SDK source over the network.

| `import` prefix | Contains |
|---|---|
| `com.anthropic.client` / `com.anthropic.client.okhttp` | `AnthropicClient`, `AnthropicOkHttpClient` |
| `com.anthropic.models.messages` | non-beta request/response types - `MessageCreateParams`, `Model`, `Message`, `TextBlockParam`, `ContentBlockParam`, `ToolUseBlockParam`, `ToolResultBlockParam`, `CacheControlEphemeral`, `Tool*` (e.g. `ToolBash20250124`, `ToolTextEditor20250728`), `StopReason`, `StructuredMessage*` |
| `com.anthropic.models.messages.batches` | Batch API - `BatchResultsParams`, `MessageBatchIndividualResponse` |
| `com.anthropic.models.beta` | `AnthropicBeta` (beta-flag constants) |
| `com.anthropic.models.beta.messages` | beta-endpoint types - `MessageCreateParams`, `BetaMessage`, `BetaStopReason`, `BetaContextManagementConfig`, `BetaMcpToolset`, `BetaRequestMcpServerUrlDefinition`, `BetaTool*` |
| `com.anthropic.core` | `JsonValue`, `JsonField`, `JsonSchemaLocalValidation`, `com.anthropic.core.http.StreamResponse` |
| `com.anthropic.errors` | typed exceptions - `AnthropicServiceException`, `RateLimitException`, `NotFoundException`, etc. (see `shared/error-codes.md`) |

`client.messages()` uses `com.anthropic.models.messages.*`; `client.beta().messages()` uses `com.anthropic.models.beta.messages.*`. Both packages define a `MessageCreateParams` - import the one matching the client path you call.

### Key types per feature

Write from this table instead of `javap`/jar inspection. Endpoint column tells you whether to use `client.messages()` or `client.beta().messages()`.

| Feature | Endpoint | Key Java types / builder calls |
|---|---|---|
| User profiles | beta | `client.beta().userProfiles().create(...)` / `.retrieve(id)` / `.list()`. Pass the returned profile id on the beta `MessageCreateParams`. Requires a beta header - check the SDK's beta-headers reference for the current flag. |
| Agent Skills | beta | `BetaContainerParams`, `BetaSkillParams`, `BetaCodeExecutionTool20250825`. `.addBeta("code-execution-2025-08-25")` (Skills is out of beta - no `skills-2025-10-02`). Download the output via `client.beta().files().download(fileId)`. |
| Cache diagnostics | beta | `BetaDiagnosticsParam`, `BetaCacheControlEphemeral` |
| Context editing | beta | `.contextManagement(BetaContextManagementConfig.builder()...)`. The edit strategy is a `BetaClearToolUses20250919Edit` (or `BetaClearThinking20251015Edit`); its trigger is a `BetaInputTokensTrigger` built separately and passed to the edit's builder - there is no direct `.inputTokensTrigger(N)` shortcut on the edit builder. `javap` the edit and trigger classes for the exact setter names. |
| Memory tool | non-beta | `.addTool(MemoryTool20250818.builder().build())` from `com.anthropic.models.messages` |
| Programmatic tool calling | non-beta | `CodeExecutionTool20260120`, `Tool`, `ContentBlockParam` |
| Strict tool use | non-beta | `Tool`, `Tool.InputSchema` |
| Task budgets | beta | `.outputConfig(BetaOutputConfig.builder().taskBudget(BetaTokenTaskBudget.builder()...))` |
| Tool search | non-beta | `.addTool(ToolSearchToolRegex20251119.builder()...)` from `com.anthropic.models.messages` |
| Web search | non-beta | `WebSearchTool20260209` from `com.anthropic.models.messages` - the latest variant with dynamic filtering (Claude Fable 5.1 + Claude Opus 5.5 + Claude Opus 5 + Opus 4.8/4.7/4.6 + Claude Sonnet 5.5 + Claude Sonnet 5 + Sonnet 4.6). For older models or Vertex, use `WebSearchTool20250305` |

### Discovering type and member names

If a class or builder method you need isn't in the tables above, `jar tf <anthropic-java-core jar> | grep -i <term>` or `javap -classpath <jar> com.anthropic.models....` is fast enough to locate names. **Do not compile and run a separate reflection program** to enumerate members - the first build is slow enough to be backgrounded in many environments, trapping you in a polling loop. Write the script with the names you found and let the compiler error (`cannot find symbol`) point at any wrong member.

## Installation

Maven:

```xml
<dependency>
    <groupId>com.anthropic</groupId>
    <artifactId>anthropic-java</artifactId>
    <version>2.34.0</version>
</dependency>
```

Gradle:

```groovy
implementation("com.anthropic:anthropic-java:2.34.0")
```

## Client Initialization

```java
import com.anthropic.client.AnthropicClient;
import com.anthropic.client.okhttp.AnthropicOkHttpClient;

// Default (reads ANTHROPIC_API_KEY from environment)
AnthropicClient client = AnthropicOkHttpClient.fromEnv();

// Explicit API key
AnthropicClient client = AnthropicOkHttpClient.builder()
    .apiKey("your-api-key")
    .build();
```

---

## Basic Message Request

```java
import com.anthropic.models.messages.MessageCreateParams;
import com.anthropic.models.messages.Message;

MessageCreateParams params = MessageCreateParams.builder()
    .model("claude-opus-5-5")  // .model(String) overload - works for every model id; typed Model.* constants lag model launches
    .maxTokens(16000L)
    .addUserMessage("What is the capital of France?")
    .build();

Message response = client.messages().create(params);
response.content().stream()
    .flatMap(block -> block.text().stream())
    .forEach(textBlock -> System.out.println(textBlock.text()));
```

---

## Thinking

**Adaptive thinking is the recommended mode for Claude 4.6+ models.** Claude decides dynamically when and how much to think. The builder has a direct `.thinking(ThinkingConfigAdaptive)` overload - no manual union wrapping.

> **Fable 5, Claude Opus 5.5, Claude Opus 5, Opus 4.8, Opus 4.7, Opus 4.6, and Sonnet 4.6:** Use adaptive thinking (below). `ThinkingConfigEnabled.builder().budgetTokens(N)` is removed on Fable 5, Claude Opus 5.5, Claude Opus 5, Opus 4.8, and 4.7 (400 if sent); deprecated on Opus 4.6 and Sonnet 4.6.
> **Claude Opus 5.5:** thinking is always on - omit `.thinking(...)` (or send `ThinkingConfigAdaptive`, which is equivalent); `ThinkingConfigDisabled` returns a 400 at every effort, as does a thinking budget. Control depth with `.outputConfig(OutputConfig.builder().effort(...))` instead - the default is `medium` on this model, where Claude Opus 5 defaults to `high`.
> **Claude Opus 5:** thinking is on by default - omitting `.thinking(...)` runs adaptive (`ThinkingConfigAdaptive` is equivalent), unlike Opus 4.8/4.7 where omitting it meant no thinking. `ThinkingConfigDisabled` is accepted only at effort `HIGH` or lower; pairing it with `XHIGH`/`MAX` returns a 400.
> **Older models:** Use `.thinking(ThinkingConfigEnabled.builder().budgetTokens(N).build())` (budget must be < `maxTokens`, min 1024).

```java
import com.anthropic.models.messages.ContentBlock;
import com.anthropic.models.messages.MessageCreateParams;
import com.anthropic.models.messages.ThinkingConfigAdaptive;

MessageCreateParams params = MessageCreateParams.builder()
    .model("claude-opus-5-5")
    .maxTokens(16000L)
    // display opt-in: default is omitted (empty thinking text) on Fable 5/5.1, Mythos 5/5.1, Claude Opus 5.5, Claude Opus 5, Opus 4.8/4.7, Claude Sonnet 5.5, and Claude Sonnet 5
    .thinking(ThinkingConfigAdaptive.builder().display(ThinkingConfigAdaptive.Display.SUMMARIZED).build())
    .addUserMessage("Solve this step by step: 27 * 453")
    .build();

for (ContentBlock block : client.messages().create(params).content()) {
    block.thinking().ifPresent(t -> System.out.println("[thinking] " + t.thinking()));
    block.text().ifPresent(t -> System.out.println(t.text()));
}
```

`ContentBlock` narrowing: `.thinking()` / `.text()` return `Optional<T>` - use `.ifPresent(...)` or `.stream().flatMap(...)`. Alternative: `isThinking()` / `asThinking()` boolean+unwrap pairs (throws on wrong variant).

---

## Effort Parameter

Effort is nested inside `OutputConfig` - there is NO `.effort()` directly on `MessageCreateParams.Builder`.

```java
import com.anthropic.models.messages.OutputConfig;

.outputConfig(OutputConfig.builder()
    .effort(OutputConfig.Effort.HIGH)  // or LOW, MEDIUM, XHIGH, MAX
    .build())
```

Combine with `Thinking = ThinkingConfigAdaptive` for cost-quality control.

---

## Prompt Caching

System message as a list of `TextBlockParam` with `CacheControlEphemeral`. Use `.systemOfTextBlockParams(...)` - the plain `.system(String)` overload can't carry cache control. For placement patterns and the silent-invalidator audit checklist, see `shared/prompt-caching.md`.

```java
import com.anthropic.models.messages.TextBlockParam;
import com.anthropic.models.messages.CacheControlEphemeral;

.systemOfTextBlockParams(List.of(
    TextBlockParam.builder()
        .text(longSystemPrompt)
        .cacheControl(CacheControlEphemeral.builder()
            .ttl(CacheControlEphemeral.Ttl.TTL_1H)  // optional; also TTL_5M
            .build())
        .build()))
```

There's also a top-level `.cacheControl(CacheControlEphemeral)` on `MessageCreateParams.Builder` and on `Tool.builder()`.

Verify hits via `response.usage().cacheCreationInputTokens()` / `response.usage().cacheReadInputTokens()`.

---

## Token Counting

```java
import com.anthropic.models.messages.MessageCountTokensParams;

long tokens = client.messages().countTokens(
    MessageCountTokensParams.builder()
        .model("claude-opus-5-5")
        .addUserMessage("Hello")
        .build()
).inputTokens();
```

---

## PDF / Document Input

`DocumentBlockParam` builder has source shortcuts. Wrap in `ContentBlockParam.ofDocument()` and pass via `.addUserMessageOfBlockParams()`.

```java
import com.anthropic.models.messages.DocumentBlockParam;
import com.anthropic.models.messages.ContentBlockParam;
import com.anthropic.models.messages.TextBlockParam;

DocumentBlockParam doc = DocumentBlockParam.builder()
    .source(Base64PdfSource.builder().data(base64String).build())
    // or .source(UrlPdfSource.builder().url("https://...").build())
    .title("My Document")        // optional
    .build();
```

For **Files API** document references, use the beta path and beta types - see `files-api.md`: `BetaRequestDocumentBlock.builder().source(BetaFileDocumentSource.builder().fileId(id).build())`.

```java
.addUserMessageOfBlockParams(List.of(
    ContentBlockParam.ofDocument(doc),
    ContentBlockParam.ofText(TextBlockParam.builder().text("Summarize this").build())))
```

---

## Stop Details

When `stopReason()` is `"refusal"`, the response includes structured `stopDetails()`:

```java
response.stopDetails().ifPresent(details -> {
    System.out.println("Category: " + details.category());
    System.out.println("Explanation: " + details.explanation());
});
```

**Refusal fallbacks (Claude Fable 5.1) - opt in by default.** Fallbacks are opt-in: without them a refused request simply stops. New `claude-fable-5-1` code should include the server-side `fallbacks` parameter (beta header `server-side-fallback-2026-06-01`, fallback model `claude-opus-4-8`, on the beta messages call) by default. The exact Java builder methods (and the client-side middleware for providers without server-side support) are not documented here - WebFetch the Java SDK repo's `examples/` from `shared/live-sources.md`; full semantics in `shared/model-migration.md` -> Migrating to Claude Fable 5.1 -> `refusal` stop reason.

---

## Error Type

`AnthropicServiceException` exposes `.errorType()` returning `Optional<ErrorType>` for programmatic error classification:

```java
try {
    client.messages().create(params);
} catch (AnthropicServiceException e) {
    e.errorType().ifPresent(type ->
        System.out.println("Error type: " + type)  // RATE_LIMIT_ERROR, OVERLOADED_ERROR, etc.
    );
}
```

---

---

### Overview (`skills/claude-api/java/managed-agents/README.md`)

# Managed Agents - Java

> **Bindings not shown here:** This README covers the most common managed-agents flows for Java. If you need a class, method, namespace, field, or behavior that isn't shown, WebFetch the Java SDK repo **or the relevant docs page** from `shared/live-sources.md` rather than guess. Do not extrapolate from cURL shapes or another language's SDK.

> **Agents are persistent - create once, reference by ID.** Store the agent ID returned by `client.beta().agents().create` and pass it to every subsequent `client.beta().sessions().create`; do not call `agents().create` in the request path. **Recommended:** define agents and environments as version-controlled files synced with `ant apply` - see `shared/anthropic-cli.md` (its live-docs URL is in `shared/live-sources.md`). The CLI owns the control plane (create/update); your code owns the data plane (sessions with the stored ID). The examples below show in-code creation for when you must provision programmatically; in production the create call belongs in setup, not in the request path.

## Installation

```xml
<dependency>
    <groupId>com.anthropic</groupId>
    <artifactId>anthropic-java</artifactId>
</dependency>
```

## Client Initialization

```java
import com.anthropic.client.okhttp.AnthropicOkHttpClient;

// Default (uses ANTHROPIC_API_KEY env var)
var client = AnthropicOkHttpClient.fromEnv();
```

---

## Create an Environment

```java
import com.anthropic.models.beta.environments.BetaCloudConfigParams;
import com.anthropic.models.beta.environments.BetaLimitedNetworkParams;
import com.anthropic.models.beta.environments.EnvironmentCreateParams;

var environment = client.beta().environments().create(EnvironmentCreateParams.builder()
    .name("my-dev-env")
    .config(BetaCloudConfigParams.builder()
        .networking(BetaLimitedNetworkParams.builder()
            .allowPackageManagers(true)
            .allowMcpServers(true)
            .build())
        .build())
    .build());
System.out.println("Environment ID: " + environment.id()); // env_...
```

---

## Create an Agent (required first step)

> Warning: **There is no inline agent config.** Model, system, and tools live on the agent object, not the session. Always start with `client.beta().agents().create()` - the session takes either `.agent(agent.id())` or the typed `BetaManagedAgentsAgentParams.builder()...build()`.

### Minimal

The examples on this page turn both web tools off. Set `enabled` to true on `web_fetch` / `web_search` only when the job as described needs the web (a general-purpose or open-ended job stays off; tell the user how to switch it on) - see `shared/managed-agents-tools.md` § Agent Toolset. They also set the `auto` permission policy (Java SDK 2.63.0+), under which a call can pause for your approval - the event loop under Stream Events answers it. When nobody is watching the run, answer `deny`; never answer `allow` to every paused call.

```
/* Detailed source-code truncated for AI context efficiency. */
```

### Updating an Agent

Updates create new versions; the agent object is immutable per version.

```java
import com.anthropic.models.beta.agents.AgentUpdateParams;

var updatedAgent = client.beta().agents().update(agent.id(), AgentUpdateParams.builder()
    .version(agent.version())
    .system("You are a helpful coding agent. Always write tests.")
    .build());
System.out.println("New version: " + updatedAgent.version());

// List all versions
for (var version : client.beta().agents().versions().list(agent.id()).autoPager()) {
    System.out.println("Version " + version.version() + ": " + version.updatedAt());
}

// Archive the agent
var archived = client.beta().agents().archive(agent.id());
System.out.println("Archived at: " + archived.archivedAt().orElseThrow());
```

---

## Send a User Message

```java
import com.anthropic.models.beta.sessions.events.BetaManagedAgentsUserMessageEventParams;
import com.anthropic.models.beta.sessions.events.EventSendParams;

client.beta().sessions().events().send(session.id(), EventSendParams.builder()
    .addEvent(BetaManagedAgentsUserMessageEventParams.builder()
        .type(BetaManagedAgentsUserMessageEventParams.Type.USER_MESSAGE)
        .addTextContent("Review the auth module")
        .build())
    .build());
```

> Tip: **Stream-first:** Open the stream *before* (or concurrently with) sending the message. The stream only delivers events that occur after it opens - stream-after-send means early events arrive buffered in one batch. See [Steering Patterns](../../shared/managed-agents-events.md#steering-patterns).

---

## Stream Events (SSE)

```
/* Detailed source-code truncated for AI context efficiency. */
```

### Reconnecting and Tailing

When reconnecting mid-session, list past events first to dedupe, then tail live events. Answer paused calls as in the loop above, including an `ask` in the history that no `user.tool_confirmation` follows. The cross-variant `id` field is read from the raw `_json()` value:

```java
import com.anthropic.core.JsonValue;
import java.util.HashSet;
import java.util.Map;
import java.util.Optional;

try (var stream = client.beta().sessions().events().streamStreaming(session.id())) {
    // Stream is open and buffering. List history before tailing live.
    var seenEventIds = new HashSet<String>();
    for (var past : client.beta().sessions().events().list(session.id()).autoPager()) {
        Optional<Map<String, JsonValue>> obj = past._json().orElseThrow().asObject();
        seenEventIds.add(obj.orElseThrow().get("id").asStringOrThrow());
    }

    // Tail live events, skipping anything already seen
    for (var event : (Iterable<StreamEvents>) stream.stream()::iterator) {
        Optional<Map<String, JsonValue>> obj = event._json().orElseThrow().asObject();
        if (!seenEventIds.add(obj.orElseThrow().get("id").asStringOrThrow())) continue;
        if (event.isAgentMessage()) {
            event.asAgentMessage().content().forEach(block -> System.out.print(block.text()));
        } else if (event.isSessionStatusIdle()) {
            // requires_action: answer the paused call as under Stream Events, then keep streaming
            if (!event.asSessionStatusIdle().stopReason().isRequiresAction()) break;
        }
    }
}
```

---

## Provide Custom Tool Result

> Note: The Java managed-agents bindings for `user.custom_tool_result` are not yet documented in this skill or in the apps source examples. Refer to `shared/managed-agents-events.md` for the wire format and the `anthropic-java` repository for the corresponding params types.

---

## Poll Events

```java
for (var event : client.beta().sessions().events().list(session.id()).autoPager()) {
    System.out.println(event.type() + ": " + event);
}
```

---

## Upload a File

```java
import com.anthropic.models.beta.files.FileUploadParams;
import com.anthropic.models.beta.sessions.BetaManagedAgentsFileResourceParams;
import java.nio.file.Path;

var dataCsv = Path.of("data.csv");

var file = client.beta().files().upload(FileUploadParams.builder()
    .file(dataCsv)
    .build());
System.out.println("File ID: " + file.id());

// Mount in a session
var session = client.beta().sessions().create(SessionCreateParams.builder()
    .agent(agent.id())
    .environmentId(environment.id())
    .addResource(BetaManagedAgentsFileResourceParams.builder()
        .type(BetaManagedAgentsFileResourceParams.Type.FILE)
        .fileId(file.id())
        .mountPath("/data.csv")
        .build())
    .build());
```

### Add and Manage Resources on an Existing Session

```java
import com.anthropic.models.beta.sessions.resources.ResourceAddParams;
import com.anthropic.models.beta.sessions.resources.ResourceDeleteParams;

// Attach an additional file to an open session
var resource = client.beta().sessions().resources().add(session.id(), ResourceAddParams.builder()
    .betaManagedAgentsFileResourceParams(BetaManagedAgentsFileResourceParams.builder()
        .type(BetaManagedAgentsFileResourceParams.Type.FILE)
        .fileId(file.id())
        .build())
    .build());
System.out.println(resource.id()); // "sesrsc_01ABC..."

// List resources on the session - entries are a discriminated union
var listed = client.beta().sessions().resources().list(session.id());
for (var entry : listed.data()) {
    if (entry.isFile()) {
        var fileResource = entry.asFile();
        System.out.println(fileResource.id() + " " + fileResource.type());
    } else if (entry.isGitHubRepository()) {
        var repoResource = entry.asGitHubRepository();
        System.out.println(repoResource.id() + " " + repoResource.type());
    }
}

// Detach a resource
client.beta().sessions().resources().delete(resource.id(), ResourceDeleteParams.builder()
    .sessionId(session.id())
    .build());
```

---

## List and Download Session Files

> Note: Listing and downloading files an agent wrote during a session is not yet documented for Java in this skill or in the apps source examples. See `shared/managed-agents-events.md` and the `anthropic-java` repository for the file list/download bindings.

---

## Session Management

```java
// List environments
var environments = client.beta().environments().list();

// Retrieve a specific environment
var env = client.beta().environments().retrieve(environment.id());

// Archive an environment (read-only, existing sessions continue)
client.beta().environments().archive(environment.id());

// Delete an environment (only if no sessions reference it)
client.beta().environments().delete(environment.id());

// Delete a session
client.beta().sessions().delete(session.id());
```

---

## MCP Server Integration

```
/* Detailed source-code truncated for AI context efficiency. */
```

See `shared/managed-agents-tools.md` §Vaults for creating vaults and adding credentials.

---

## Vaults

```
/* Detailed source-code truncated for AI context efficiency. */
```

---

## GitHub Repository Integration

Mount a GitHub repository as a session resource (a vault holds the GitHub MCP credential):

```java
import com.anthropic.models.beta.sessions.BetaManagedAgentsGitHubRepositoryResourceParams;

var session = client.beta().sessions().create(SessionCreateParams.builder()
    .agent(agent.id())
    .environmentId(environment.id())
    .addVaultId(vault.id())
    .addResource(BetaManagedAgentsGitHubRepositoryResourceParams.builder()
        .type(BetaManagedAgentsGitHubRepositoryResourceParams.Type.GITHUB_REPOSITORY)
        .url("https://github.com/org/repo")
        .mountPath("/workspace/repo")
        .authorizationToken("ghp_your_github_token")
        .build())
    .build());
```

Multiple repositories on the same session:

```java
import java.util.List;

var resources = List.of(
    BetaManagedAgentsGitHubRepositoryResourceParams.builder()
        .type(BetaManagedAgentsGitHubRepositoryResourceParams.Type.GITHUB_REPOSITORY)
        .url("https://github.com/org/frontend")
        .mountPath("/workspace/frontend")
        .authorizationToken("ghp_your_github_token")
        .build(),
    BetaManagedAgentsGitHubRepositoryResourceParams.builder()
        .type(BetaManagedAgentsGitHubRepositoryResourceParams.Type.GITHUB_REPOSITORY)
        .url("https://github.com/org/backend")
        .mountPath("/workspace/backend")
        .authorizationToken("ghp_your_github_token")
        .build());
```

Rotating a repository's authorization token:

```java
import com.anthropic.models.beta.sessions.resources.ResourceUpdateParams;

var listed = client.beta().sessions().resources().list(session.id());
var repoResourceId = listed.data().get(0).asGitHubRepository().id();

client.beta().sessions().resources().update(repoResourceId, ResourceUpdateParams.builder()
    .sessionId(session.id())
    .authorizationToken("ghp_your_new_github_token")
    .build());
```

---

### Overview (`skills/claude-api/php/claude-api/README.md`)

# Claude API - PHP

> **Note:** The PHP SDK is the official Anthropic SDK for PHP. A beta tool runner is available via `$client->beta->messages->toolRunner()`. Structured output helpers are supported via `StructuredOutputModel` classes. Agent SDK is not available. Bedrock, Vertex AI, and Foundry clients are supported.

## Installation

```bash
composer require "anthropic-ai/sdk"
```

## Client Initialization

```php
use Anthropic\Client;

// Using API key from environment variable
$client = new Client(apiKey: getenv("ANTHROPIC_API_KEY"));
```

### Amazon Bedrock

```php
use Anthropic\Bedrock\MantleClient;

// Messages-API Bedrock endpoint. Reads AWS credentials from env.
$client = new MantleClient(awsRegion: 'us-east-1');
```

Model IDs on Bedrock take an `anthropic.` prefix - e.g. `model: 'anthropic.claude-opus-5-5'`.

### Google Vertex AI

```php
use Anthropic\Vertex;

// Constructor is private. Parameter is `location`, not `region`.
$client = Vertex\Client::fromEnvironment(
    location: 'us-east5',
    projectId: 'my-project-id',
);
```

### Anthropic Foundry

```php
use Anthropic\Foundry;

// Constructor is private. baseUrl or resource is required.
$client = Foundry\Client::withCredentials(
    apiKey: getenv('ANTHROPIC_FOUNDRY_API_KEY'),
    baseUrl: 'https://<resource>.services.ai.azure.com/anthropic/v1',
);
```

---

## Basic Message Request

```php
$message = $client->messages->create(
    model: 'claude-opus-5-5',
    maxTokens: 16000,
    messages: [
        ['role' => 'user', 'content' => 'What is the capital of France?'],
    ],
);

// content is an array of polymorphic blocks (TextBlock, ToolUseBlock,
// ThinkingBlock). Accessing ->text on content[0] without checking the block
// type will throw if the first block is not a TextBlock (e.g., when extended
// thinking is enabled and a ThinkingBlock comes first). Always guard:
foreach ($message->content as $block) {
    if ($block->type === 'text') {
        echo $block->text;
    }
}
```

If you only want the first text block:

```php
foreach ($message->content as $block) {
    if ($block->type === 'text') {
        echo $block->text;
        break;
    }
}
```

---

## Extended Thinking

**Adaptive thinking is the recommended mode for Claude 4.6+ models.** Claude decides dynamically when and how much to think.

```php
use Anthropic\Messages\ThinkingBlock;

$message = $client->messages->create(
    model: 'claude-opus-5-5',
    maxTokens: 16000,
    thinking: ['type' => 'adaptive', 'display' => 'summarized'], // display opt-in: default is omitted (empty thinking text) on Fable 5/5.1, Mythos 5/5.1, Claude Opus 5.5, Claude Opus 5, Opus 4.8/4.7, Claude Sonnet 5.5, and Claude Sonnet 5
    messages: [
        ['role' => 'user', 'content' => 'Solve: 27 * 453'],
    ],
);

// ThinkingBlock(s) precede TextBlock in content
foreach ($message->content as $block) {
    if ($block instanceof ThinkingBlock) {
        echo "Thinking:\n{$block->thinking}\n\n";
        // $block->signature is an opaque string - preserve verbatim if
        // passing thinking blocks back in multi-turn conversations
    } elseif ($block->type === 'text') {
        echo "Answer: {$block->text}\n";
    }
}
```

> **Fable 5, Claude Opus 5.5, Claude Opus 5, Opus 4.8, Opus 4.7, Opus 4.6, and Sonnet 4.6:** Use adaptive thinking (above). `['type' => 'enabled', 'budgetTokens' => N]` is removed on Fable 5, Claude Opus 5.5, Claude Opus 5, Opus 4.8, and 4.7 (400 if sent); deprecated on Opus 4.6 and Sonnet 4.6.
> **Claude Opus 5.5:** thinking is always on - omit `thinking:` (or send `['type' => 'adaptive']`, which is equivalent); `['type' => 'disabled']` returns a 400 at every effort, as does a thinking budget. Control depth with the `outputConfig:` effort instead - the default is `medium` on this model, where Claude Opus 5 defaults to `high`.
> **Claude Opus 5:** thinking is on by default - omitting `thinking:` runs adaptive (`['type' => 'adaptive']` is equivalent), unlike Opus 4.8/4.7 where omitting it meant no thinking. `['type' => 'disabled']` is accepted only at effort `high` or lower; pairing it with `xhigh`/`max` returns a 400.
> **Older models:** Use `thinking: ['type' => 'enabled', 'budgetTokens' => N]` (budget must be < `maxTokens`, min 1024).

`$block->type === 'thinking'` also works for the check; `instanceof` narrows for PHPStan.

---

## Prompt Caching

`system:` takes an array of text blocks; set `cacheControl` on the last block. Array-shape syntax (camelCase keys) is idiomatic. For placement patterns and the silent-invalidator audit checklist, see `shared/prompt-caching.md`.

```php
$message = $client->messages->create(
    model: 'claude-opus-5-5',
    maxTokens: 16000,
    system: [
        ['type' => 'text', 'text' => $longSystemPrompt, 'cacheControl' => ['type' => 'ephemeral']],
    ],
    messages: [['role' => 'user', 'content' => 'Summarize the key points']],
);
```

For 1-hour TTL: `'cacheControl' => ['type' => 'ephemeral', 'ttl' => '1h']`. There's also a top-level `cacheControl:` on `messages->create(...)` that auto-places on the last cacheable block.

Verify hits via `$message->usage->cacheCreationInputTokens` / `$message->usage->cacheReadInputTokens`.

---

## Stop Details

When `stopReason` is `'refusal'`, the response includes structured `stopDetails`:

```php
if ($message->stopReason === 'refusal' && $message->stopDetails !== null) {
    echo "Category: " . $message->stopDetails->category . "\n";     // e.g. "cyber", "bio", "reasoning_extraction", "frontier_llm", or null - see docs for the full set
    echo "Explanation: " . $message->stopDetails->explanation . "\n";
}
```

**Refusal fallbacks (Claude Fable 5.1) - opt in by default.** Fallbacks are opt-in: without them a refused request simply stops. New `claude-fable-5-1` code should include the server-side `fallbacks` parameter (beta header `server-side-fallback-2026-06-01`, fallback model `claude-opus-4-8`, on the beta messages call) by default. The exact PHP binding (and the client-side middleware for providers without server-side support) is not documented here - WebFetch the PHP SDK repo's `examples/` from `shared/live-sources.md`; full semantics in `shared/model-migration.md` -> Migrating to Claude Fable 5.1 -> `refusal` stop reason.

---

## Error Type

`APIStatusException` exposes a `->type` property for programmatic error classification:

```php
try {
    $client->messages->create(...);
} catch (\Anthropic\Core\Exceptions\APIStatusException $e) {
    echo $e->type?->value;  // "rate_limit_error", "overloaded_error", etc.
}
```

---

### Overview (`skills/claude-api/php/managed-agents/README.md`)

# Managed Agents - PHP

> **Bindings not shown here:** This README covers the most common managed-agents flows for PHP. If you need a class, method, namespace, field, or behavior that isn't shown, WebFetch the PHP SDK repo **or the relevant docs page** from `shared/live-sources.md` rather than guess. Do not extrapolate from cURL shapes or another language's SDK.

> **Agents are persistent - create once, reference by ID.** Store the agent ID returned by `$client->beta->agents->create` and pass it to every subsequent `->sessions->create`; do not call `agents->create` in the request path. **Recommended:** define agents and environments as version-controlled files synced with `ant apply` - see `shared/anthropic-cli.md` (its live-docs URL is in `shared/live-sources.md`). The CLI owns the control plane (create/update); your code owns the data plane (sessions with the stored ID). The examples below show in-code creation for when you must provision programmatically; in production the create call belongs in setup, not in the request path.

## Installation

```bash
composer require "anthropic-ai/sdk" "guzzlehttp/guzzle:^7"
```

## Client Initialization

```php
use Anthropic\Client;

// Default (uses ANTHROPIC_API_KEY env var)
$client = new Client();

// Explicit API key
$client = new Client(apiKey: 'your-api-key');
```

---

## Create an Environment

```php
$environment = $client->beta->environments->create(
    name: 'my-dev-env',
    config: ['type' => 'cloud', 'networking' => ['type' => 'limited', 'allow_package_managers' => true, 'allow_mcp_servers' => true]],
);
echo "Environment ID: {$environment->id}\n"; // env_...
```

---

## Create an Agent (required first step)

> Warning: **There is no inline agent config.** `model`/`system`/`tools` live on the agent object, not the session. Always start with `$client->beta->agents->create()` - the session takes either `agent: $agent->id` or the typed `BetaManagedAgentsAgentParams::with(type: 'agent', id: $agent->id, version: $agent->version)`.

### Minimal

The examples on this page turn both web tools off. Set `enabled` to true on `web_fetch` / `web_search` only when the job as described needs the web (a general-purpose or open-ended job stays off; tell the user how to switch it on) - see `shared/managed-agents-tools.md` § Agent Toolset. They also set the `auto` permission policy (PHP SDK 0.48.0+), under which a call can pause for your approval - the event loop under Stream Events answers it. When nobody is watching the run, answer `deny`; never answer `allow` to every paused call.

```php
use Anthropic\Beta\Agents\BetaManagedAgentsAgentToolset20260401Params;
use Anthropic\Beta\Agents\BetaManagedAgentsAgentToolsetDefaultConfigParams;
use Anthropic\Beta\Agents\BetaManagedAgentsAutoPolicy;
use Anthropic\Beta\Agents\BetaManagedAgentsWebFetchToolConfigParams;
use Anthropic\Beta\Agents\BetaManagedAgentsWebSearchToolConfigParams;

// 1. Create the agent (reusable, versioned)
$agent = $client->beta->agents->create(
    name: 'Coding Assistant',
    model: 'claude-opus-5-5',
    system: 'You are a helpful coding assistant.',
    tools: [
        BetaManagedAgentsAgentToolset20260401Params::with(
            type: 'agent_toolset_20260401',
            defaultConfig: BetaManagedAgentsAgentToolsetDefaultConfigParams::with(
                permissionPolicy: BetaManagedAgentsAutoPolicy::with(),
            ),
            configs: [
                BetaManagedAgentsWebFetchToolConfigParams::with(enabled: false),
                BetaManagedAgentsWebSearchToolConfigParams::with(enabled: false),
            ],
        ),
    ],
);

// 2. Start a session
$session = $client->beta->sessions->create(
    agent: ['type' => 'agent', 'id' => $agent->id, 'version' => $agent->version],
    environmentID: $environment->id,
    title: 'Quickstart session',
);
echo "Session ID: {$session->id}\n";
echo "Trace: https://platform.claude.com/workspaces/default/sessions/{$session->id}\n"; // swap 'default' for your workspace ID if the API key is not in the Default workspace
```

### Updating an Agent

Updates create new versions; the agent object is immutable per version.

```php
$updatedAgent = $client->beta->agents->update(
    $agent->id,
    version: $agent->version,
    system: 'You are a helpful coding agent. Always write tests.',
);
echo "New version: {$updatedAgent->version}\n";

// List all versions
foreach ($client->beta->agents->versions->list($agent->id)->pagingEachItem() as $version) {
    echo "Version {$version->version}: {$version->updatedAt->format(DateTimeInterface::ATOM)}\n";
}

// Archive the agent
$archived = $client->beta->agents->archive($agent->id);
echo "Archived at: {$archived->archivedAt->format(DateTimeInterface::ATOM)}\n";
```

---

## Send a User Message

```php
$client->beta->sessions->events->send(
    $session->id,
    events: [
        [
            'type' => 'user.message',
            'content' => [['type' => 'text', 'text' => 'Review the auth module']],
        ],
    ],
);
```

> Tip: **Stream-first:** Open the stream *before* (or concurrently with) sending the message. The stream only delivers events that occur after it opens - stream-after-send means early events arrive buffered in one batch. See [Steering Patterns](../../shared/managed-agents-events.md#steering-patterns).

---

## Stream Events (SSE)

> Note: **Streaming transporter:** PHP's default buffered PSR-18 client never returns for the open-ended session event stream. Use a streaming Guzzle transporter for `streamStream()` calls - other calls keep the default client.

```
/* Detailed source-code truncated for AI context efficiency. */
```

### Reconnecting and Tailing

When reconnecting mid-session, list past events first to dedupe, then tail live events. Answer paused calls as in the loop above, including an `ask` in the history that no `user.tool_confirmation` follows.

```php
$stream = $client->beta->sessions->events->streamStream(
    $session->id,
    requestOptions: ['transporter' => $streamingClient],
);

// Stream is open and buffering. List history before tailing live.
$seenEventIds = [];
foreach ($client->beta->sessions->events->list($session->id)->pagingEachItem() as $event) {
    $seenEventIds[$event->id] = true;
}

// Tail live events, skipping anything already seen
foreach ($stream as $event) {
    if (isset($seenEventIds[$event->id])) {
        continue;
    }
    $seenEventIds[$event->id] = true;
    match ($event->type) {
        'agent.message' => array_walk(
            $event->content,
            static fn($block) => $block->type === 'text' ? print($block->text) : null,
        ),
        default => null,
    };
    // requires_action: answer the paused call as under Stream Events, then keep streaming
    if ($event->type === 'session.status_idle' && $event->stopReason->type !== 'requires_action') {
        break;
    }
}
$stream->close();
```

---

## Provide Custom Tool Result

> Note: The PHP managed-agents bindings for `user.custom_tool_result` are not yet documented in this skill or in the apps source examples. Refer to `shared/managed-agents-events.md` for the wire format and the `anthropic-ai/sdk` PHP repository for the corresponding params.

---

## Poll Events

```php
foreach ($client->beta->sessions->events->list($session->id)->pagingEachItem() as $event) {
    echo "{$event->type}: {$event->id}\n";
}
```

---

## Upload a File

> Note: **PHP file upload:** The PHP SDK's beta managed-agents file upload binding is not shown in the apps source examples; the canonical PHP example uses raw cURL against `POST /v1/files`. If your codebase prefers the SDK, WebFetch the `anthropic-ai/sdk` PHP repository for the latest binding before writing code.

```php
use Anthropic\Beta\Sessions\BetaManagedAgentsFileResourceParams;

// Raw cURL upload (canonical example from the apps source)
$csvPath = 'data.csv';
$ch = curl_init('https://api.anthropic.com/v1/files');
curl_setopt_array($ch, [
    CURLOPT_RETURNTRANSFER => true,
    CURLOPT_POST => true,
    CURLOPT_HTTPHEADER => [
        'x-api-key: ' . getenv('ANTHROPIC_API_KEY'),
        'anthropic-version: 2023-06-01',
        'anthropic-beta: files-api-2025-04-14',
    ],
    CURLOPT_POSTFIELDS => ['file' => new CURLFile($csvPath, 'text/csv', 'data.csv')],
]);
$file = json_decode(curl_exec($ch));
echo "File ID: {$file->id}\n";

// Mount in a session
$session = $client->beta->sessions->create(
    agent: $agent->id,
    environmentID: $environment->id,
    resources: [
        BetaManagedAgentsFileResourceParams::with(
            type: 'file',
            fileID: $file->id,
            mountPath: '/data.csv',
        ),
    ],
);
```

### Add and Manage Resources on an Existing Session

```php
// Attach an additional file to an open session
$resource = $client->beta->sessions->resources->add(
    $session->id,
    type: 'file',
    fileID: $file->id,
);
echo "{$resource->id}\n"; // "sesrsc_01ABC..."

// List resources on the session
$listed = $client->beta->sessions->resources->list($session->id);
foreach ($listed->data as $entry) {
    echo "{$entry->id} {$entry->type}\n";
}

// Detach a resource
$client->beta->sessions->resources->delete($resource->id, sessionID: $session->id);
```

---

## List and Download Session Files

```php
$files = $client->beta->files->list(
    scopeID: 'sesn_abc123',
    betas: ['managed-agents-2026-04-01'],
);
$content = $client->beta->files->download($files->data[0]->id);
file_put_contents('output.txt', $content);
```

---

## Session Management

```php
// List environments
$environments = $client->beta->environments->list();

// Retrieve a specific environment
$env = $client->beta->environments->retrieve($environment->id);

// Archive an environment (read-only, existing sessions continue)
$client->beta->environments->archive($environment->id);

// Delete an environment (only if no sessions reference it)
$client->beta->environments->delete($environment->id);

// Delete a session
$client->beta->sessions->delete($session->id);
```

---

## MCP Server Integration

```
/* Detailed source-code truncated for AI context efficiency. */
```

See `shared/managed-agents-tools.md` §Vaults for creating vaults and adding credentials.

---

## Vaults

```php
// Create a vault
$vault = $client->beta->vaults->create(
    displayName: 'Alice',
    metadata: ['external_user_id' => 'usr_abc123'],
);
echo $vault->id . "\n"; // "vlt_01ABC..."

// Add an OAuth credential
$credential = $client->beta->vaults->credentials->create(
    vaultID: $vault->id,
    displayName: "Alice's Slack",
    auth: [
        'type' => 'mcp_oauth',
        'mcp_server_url' => 'https://mcp.slack.com/mcp',
        'access_token' => 'xoxp-...',
        'expires_at' => '2026-04-15T00:00:00Z',
        'refresh' => [
            'token_endpoint' => 'https://slack.com/api/oauth.v2.access',
            'client_id' => '1234567890.0987654321',
            'scope' => 'channels:read chat:write',
            'refresh_token' => 'xoxe-1-...',
            'token_endpoint_auth' => [
                'type' => 'client_secret_post',
                'client_secret' => 'abc123...',
            ],
        ],
    ],
);

// Rotate the credential (e.g., after a token refresh)
$client->beta->vaults->credentials->update(
    $credential->id,
    vaultID: $vault->id,
    auth: [
        'type' => 'mcp_oauth',
        'access_token' => 'xoxp-new-...',
        'expires_at' => '2026-05-15T00:00:00Z',
        'refresh' => ['refresh_token' => 'xoxe-1-new-...'],
    ],
);

// Archive a vault
$client->beta->vaults->archive($vault->id);
```

---

## GitHub Repository Integration

Mount a GitHub repository as a session resource (a vault holds the GitHub MCP credential):

```php
$session = $client->beta->sessions->create(
    agent: $agent->id,
    environmentID: $environment->id,
    vaultIDs: [$vault->id],
    resources: [
        [
            'type' => 'github_repository',
            'url' => 'https://github.com/org/repo',
            'mount_path' => '/workspace/repo',
            'authorization_token' => 'ghp_your_github_token',
        ],
    ],
);
```

Multiple repositories on the same session:

```php
$resources = [
    [
        'type' => 'github_repository',
        'url' => 'https://github.com/org/frontend',
        'mount_path' => '/workspace/frontend',
        'authorization_token' => 'ghp_your_github_token',
    ],
    [
        'type' => 'github_repository',
        'url' => 'https://github.com/org/backend',
        'mount_path' => '/workspace/backend',
        'authorization_token' => 'ghp_your_github_token',
    ],
];
```

Rotating a repository's authorization token:

```php
$listed = $client->beta->sessions->resources->list($session->id);
$repoResourceId = $listed->data[0]->id;

$client->beta->sessions->resources->update(
    $repoResourceId,
    sessionID: $session->id,
    authorizationToken: 'ghp_your_new_github_token',
);
```

---

### Overview (`skills/claude-api/python/claude-api/README.md`)

# Claude API - Python

## Installation

```bash
pip install anthropic
```

## Client Initialization

```python
import anthropic

# Default - resolves credentials from the environment:
# ANTHROPIC_API_KEY, or ANTHROPIC_AUTH_TOKEN, or an `ant auth login` profile.
# Prefer this for local dev; don't hardcode a key.
client = anthropic.Anthropic()

# Explicit API key (only when you must inject a specific key)
client = anthropic.Anthropic(api_key="your-api-key")

# Async client
async_client = anthropic.AsyncAnthropic()
```

---

## Client Configuration

### Per-request overrides

Use `with_options()` to override client settings for a single call without mutating the client:

```python
client.with_options(timeout=5.0, max_retries=5).messages.create(
    model="claude-opus-5-5",
    max_tokens=1024,
    messages=[{"role": "user", "content": "Hello"}],
)
```

### Timeouts

Default request timeout is 10 minutes. Pass a float (seconds) or an `anthropic.Timeout` for granular control. On timeout the SDK raises `anthropic.APITimeoutError` (and retries per `max_retries`).

```python
client = anthropic.Anthropic(timeout=20.0)
client = anthropic.Anthropic(
    timeout=anthropic.Timeout(60.0, read=5.0, write=10.0, connect=2.0),
)
```

`anthropic` 1.x is built on [`httpx2`](https://pypi.org/project/httpx2/), not `httpx`. `anthropic.Timeout` is `httpx2.Timeout`; if you import the HTTP library yourself, write `import httpx2 as httpx` - an object from the `httpx` package (`httpx.Timeout`, `httpx.Client`, transports, limits) is rejected or fails at request time. Existing `httpx`-era code is covered by the [v1 migration guide](https://github.com/anthropics/anthropic-sdk-python/blob/main/MIGRATION.md) and `/claude-api upgrade python`.

### Retries

The SDK auto-retries connection errors, 408, 409, 429, and >=500 with exponential backoff (default 2 retries). Set `max_retries` on the client or via `with_options()`; `max_retries=0` disables.

### Async performance (aiohttp backend)

For high-concurrency async workloads, install `anthropic[aiohttp]` and pass `DefaultAioHttpClient` instead of the default httpx2 backend:

```python
from anthropic import AsyncAnthropic, DefaultAioHttpClient

async with AsyncAnthropic(http_client=DefaultAioHttpClient()) as client:
    ...
```

### Custom HTTP client (proxy, base URL)

Use `DefaultHttpxClient` / `DefaultAsyncHttpxClient` - not a raw `httpx2.Client` (and never a client from the `httpx` package) - so the SDK's default timeouts and connection limits are preserved:

```python
from anthropic import Anthropic, DefaultHttpxClient

client = Anthropic(
    base_url="http://my.test.server.example.com:8083",  # or ANTHROPIC_BASE_URL env var
    http_client=DefaultHttpxClient(proxy="http://my.test.proxy.example.com"),
)
```

### Logging

Set `ANTHROPIC_LOG=debug` (or `info`) to enable SDK logging via the standard `logging` module.

---

## Basic Message Request

```python
response = client.messages.create(
    model="claude-opus-5-5",
    max_tokens=16000,
    messages=[
        {"role": "user", "content": "What is the capital of France?"}
    ]
)
# response.content is a list of content block objects (TextBlock, ThinkingBlock,
# ToolUseBlock, ...). Check .type before accessing .text.
for block in response.content:
    if block.type == "text":
        print(block.text)
```

---

## System Prompts

```python
response = client.messages.create(
    model="claude-opus-5-5",
    max_tokens=16000,
    system="You are a helpful coding assistant. Always provide examples in Python.",
    messages=[{"role": "user", "content": "How do I read a JSON file?"}]
)
```

### Mid-conversation system messages (model-gated)

For operator instructions that arrive mid-conversation (mode switches, injected state), append `{"role": "system", ...}` to `messages` instead of editing top-level `system` - this preserves the cached prefix and carries operator authority. Must follow a user message (or an `assistant` message ending in server-tool use), and must be either the last entry in `messages` or be followed by an `assistant` turn; cannot be `messages[0]`. Unsupported models return a 400 (`role 'system' is not supported on this model`). See `shared/prompt-caching.md` for when to use this vs. top-level `system`.

```python
response = client.messages.create(
    model=MODEL_ID,  # must support mid-conversation system messages
    max_tokens=16000,
    system=[{"type": "text", "text": STABLE_SYSTEM, "cache_control": {"type": "ephemeral"}}],
    messages=history + [
        {"role": "user", "content": user_message},
        {"role": "system", "content": "Terse mode enabled - keep responses under 40 words."},
    ],
)  # No beta header needed - use regular client.messages.create
```

---

## Vision (Images)

### Base64

```python
import base64

with open("image.png", "rb") as f:
    image_data = base64.standard_b64encode(f.read()).decode("utf-8")

response = client.messages.create(
    model="claude-opus-5-5",
    max_tokens=16000,
    messages=[{
        "role": "user",
        "content": [
            {
                "type": "image",
                "source": {
                    "type": "base64",
                    "media_type": "image/png",
                    "data": image_data
                }
            },
            {"type": "text", "text": "What's in this image?"}
        ]
    }]
)
```

### URL

```python
response = client.messages.create(
    model="claude-opus-5-5",
    max_tokens=16000,
    messages=[{
        "role": "user",
        "content": [
            {
                "type": "image",
                "source": {
                    "type": "url",
                    "url": "https://example.com/image.png"
                }
            },
            {"type": "text", "text": "Describe this image"}
        ]
    }]
)
```

---

## Prompt Caching

Cache large context to reduce costs (up to 90% savings). **Caching is a prefix match** - any byte change anywhere in the prefix invalidates everything after it. For placement patterns, architectural guidance (frozen system prompt, deterministic tool order, where to put volatile content), and the silent-invalidator audit checklist, read `shared/prompt-caching.md`.

### Automatic Caching (Recommended)

Use top-level `cache_control` to automatically cache the last cacheable block in the request - no need to annotate individual content blocks:

```python
response = client.messages.create(
    model="claude-opus-5-5",
    max_tokens=16000,
    cache_control={"type": "ephemeral"},  # auto-caches the last cacheable block
    system="You are an expert on this large document...",
    messages=[{"role": "user", "content": "Summarize the key points"}]
)
```

### Manual Cache Control

For fine-grained control, add `cache_control` to specific content blocks:

```python
response = client.messages.create(
    model="claude-opus-5-5",
    max_tokens=16000,
    system=[{
        "type": "text",
        "text": "You are an expert on this large document...",
        "cache_control": {"type": "ephemeral"}  # default TTL is 5 minutes
    }],
    messages=[{"role": "user", "content": "Summarize the key points"}]
)

# With explicit TTL (time-to-live)
response = client.messages.create(
    model="claude-opus-5-5",
    max_tokens=16000,
    system=[{
        "type": "text",
        "text": "You are an expert on this large document...",
        "cache_control": {"type": "ephemeral", "ttl": "1h"}  # 1 hour TTL
    }],
    messages=[{"role": "user", "content": "Summarize the key points"}]
)
```

### Verifying Cache Hits

```python
print(response.usage.cache_creation_input_tokens)  # tokens written to cache (~1.25x cost)
print(response.usage.cache_read_input_tokens)      # tokens served from cache (~0.1x cost)
print(response.usage.input_tokens)                 # uncached tokens (full cost)
```

If `cache_read_input_tokens` is zero across repeated identical-prefix requests, a silent invalidator is at work - `datetime.now()` or a UUID in the system prompt, unsorted `json.dumps()`, or a varying tool set. See `shared/prompt-caching.md` for the full audit table.

---

## Extended Thinking

> **Fable 5, Claude Opus 5.5, Claude Opus 5, Opus 4.8, Opus 4.7, Opus 4.6, and Sonnet 4.6:** Use adaptive thinking. `budget_tokens` is removed on Fable 5, Claude Opus 5.5, Claude Opus 5, Opus 4.8, and 4.7 (400 if sent); deprecated on Opus 4.6 and Sonnet 4.6.
> **Claude Opus 5.5:** thinking is always on - omit `thinking` (or send `{"type": "adaptive"}`, which is equivalent); `{"type": "disabled"}` returns a 400 at every effort, as does a thinking budget. Control depth with `output_config.effort` instead - the default is `medium` on this model, where Claude Opus 5 defaults to `high`.
> **Claude Opus 5:** thinking is on by default - omitting `thinking` runs adaptive (`{"type": "adaptive"}` is equivalent), unlike Opus 4.8/4.7 where omitting it meant no thinking. `{"type": "disabled"}` is accepted only at effort `high` or lower; pairing it with `xhigh`/`max` returns a 400.
> **Claude Haiku 5.5:** adaptive thinking only, on by default; `budget_tokens` returns a 400.
> **Older models:** Use `thinking: {type: "enabled", budget_tokens: N}` (must be < `max_tokens`, min 1024).

```python
# Fable 5 / Claude Opus 5.5 / Claude Opus 5 / Opus 4.8 / 4.7 / 4.6: adaptive thinking (recommended)
response = client.messages.create(
    model="claude-opus-5-5",
    max_tokens=16000,
    thinking={"type": "adaptive", "display": "summarized"},  # display opt-in: default is omitted (empty thinking text) on Fable 5/5.1, Mythos 5/5.1, Claude Opus 5.5, Claude Opus 5, Opus 4.8/4.7, Claude Sonnet 5.5, and Claude Sonnet 5
    output_config={"effort": "high"},  # low | medium | high | xhigh | max
    messages=[{"role": "user", "content": "Solve this step by step..."}]
)

# Access thinking and response
for block in response.content:
    if block.type == "thinking":
        print(f"Thinking: {block.thinking}")
    elif block.type == "text":
        print(f"Response: {block.text}")
```

---

## Error Handling

```python
import anthropic

try:
    response = client.messages.create(...)
except anthropic.BadRequestError as e:
    print(f"Bad request: {e.message}")
except anthropic.AuthenticationError:
    print("Invalid API key")
except anthropic.PermissionDeniedError:
    print("API key lacks required permissions")
except anthropic.NotFoundError:
    print("Invalid model or endpoint")
except anthropic.RateLimitError as e:
    retry_after = int(e.response.headers.get("retry-after", "60"))
    print(f"Rate limited. Retry after {retry_after}s.")
except anthropic.APIStatusError as e:
    if e.status_code >= 500:
        print(f"Server error ({e.status_code}). Retry later.")
    else:
        print(f"API error: {e.message}")
except anthropic.APIConnectionError:
    print("Network error. Check internet connection.")
```

---

## Response Helpers

Every response object exposes `_request_id` (populated from the `request-id` header) - log it when reporting failures to Anthropic. Despite the underscore prefix, this property is public.

```python
message = client.messages.create(...)
print(message._request_id)       # req_018EeWyXxfu5pfWkrYcMdjWG
print(message.to_json())          # serialize the Pydantic model
print(message.to_dict())          # plain dict
```

To access raw headers or other response metadata, use `.with_raw_response`:

```python
raw = client.messages.with_raw_response.create(
    model="claude-opus-5-5",
    max_tokens=1024,
    messages=[{"role": "user", "content": "Hello"}],
)
print(raw.headers.get("request-id"))
message = raw.parse()  # the Message object messages.create() would have returned
```

---

## Multi-Turn Conversations

The API is stateless - send the full conversation history each time.

```python
class ConversationManager:
    """Manage multi-turn conversations with the Claude API."""

    def __init__(self, client: anthropic.Anthropic, model: str, system: str = None):
        self.client = client
        self.model = model
        self.system = system
        self.messages = []

    def send(self, user_message: str, **kwargs) -> str:
        """Send a message and get a response."""
        self.messages.append({"role": "user", "content": user_message})

        response = self.client.messages.create(
            model=self.model,
            max_tokens=kwargs.get("max_tokens", 16000),
            system=self.system,
            messages=self.messages,
            **kwargs
        )

        assistant_message = next(
            (b.text for b in response.content if b.type == "text"), ""
        )
        self.messages.append({"role": "assistant", "content": assistant_message})

        return assistant_message

# Usage
conversation = ConversationManager(
    client=anthropic.Anthropic(),
    model="claude-opus-5-5",
    system="You are a helpful assistant."
)

response1 = conversation.send("My name is Alice.")
response2 = conversation.send("What's my name?")  # Claude remembers "Alice"
```

**Rules:**

- Consecutive same-role messages are allowed - the API combines them into a single turn
- First message must be `user`
- `role: "system"` messages are allowed mid-conversation on supporting models (no beta header needed) - see § Mid-conversation system messages above

---

### Compaction (long conversations)

> **Beta, Fable 5, Claude Opus 5.5, Claude Opus 5, Opus 4.8, Opus 4.7, Opus 4.6, and Sonnet 4.6.** When conversations approach the 200K context window, compaction automatically summarizes earlier context server-side. The API returns a `compaction` block; you must pass it back on subsequent requests - append `response.content`, not just the text.

```python
import anthropic

client = anthropic.Anthropic()
messages = []

def chat(user_message: str) -> str:
    messages.append({"role": "user", "content": user_message})

    response = client.beta.messages.create(
        betas=["compact-2026-01-12"],
        model="claude-opus-5-5",
        max_tokens=16000,
        messages=messages,
        context_management={
            "edits": [{"type": "compact_20260112"}]
        }
    )

    # Append full content - compaction blocks must be preserved
    messages.append({"role": "assistant", "content": response.content})

    return next(block.text for block in response.content if block.type == "text")

# Compaction triggers automatically when context grows large
print(chat("Help me build a Python web scraper"))
print(chat("Add support for JavaScript-rendered pages"))
print(chat("Now add rate limiting and error handling"))
```

---

## Stop Reasons

The `stop_reason` field in the response indicates why the model stopped generating:

| Value | Meaning |
|-------|---------|
| `end_turn` | Claude finished its response naturally |
| `max_tokens` | Hit the `max_tokens` limit - increase it or use streaming |
| `stop_sequence` | Hit a custom stop sequence |
| `tool_use` | Claude wants to call a tool - execute it and continue |
| `pause_turn` | Model paused and can be resumed (agentic flows) |
| `refusal` | Claude refused for safety reasons - check `stop_details` |

### Structured Stop Details

When `stop_reason` is `"refusal"`, the response includes a `stop_details` object with structured information about the refusal:

```python
if response.stop_reason == "refusal" and response.stop_details:
    print(f"Category: {response.stop_details.category}")   # e.g. "cyber", "bio", "reasoning_extraction", "frontier_llm", or None - see docs for the full set
    print(f"Explanation: {response.stop_details.explanation}")
```

### Refusal Fallbacks (Claude Fable 5.1) - opt in by default

Fallbacks are **opt-in**: without them a refused request simply stops. Include the server-side `fallbacks` parameter in `claude-fable-5-1` code by default - on a policy decline the API re-runs the same request on the fallback model inside the same call. A mid-stream decline is billed at normal rates, and the rescue bills at the fallback model's own rates, with cache repricing applied automatically; for a decline before any output, see [How refusals are billed](https://platform.claude.com/docs/en/build-with-claude/refusals-and-fallback#how-refusals-are-billed).

```python
response = client.beta.messages.create(
    model="claude-fable-5-1",
    max_tokens=16000,
    betas=["server-side-fallback-2026-06-01"],
    fallbacks=[{"model": "claude-opus-4-8"}],
    messages=[{"role": "user", "content": "..."}],
)

# Switch points: one fallback block per model that ran and declined this turn
for block in response.content:
    if block.type == "fallback":
        print(f"{block.from_.model} declined; {block.to.model} continued")

# Served-by signal - covers sticky turns, which carry no fallback block.
# Pair with stop_reason: the fallback model can itself refuse.
fallback_ran = any(
    entry.type == "fallback_message" for entry in response.usage.iterations or []
)
if fallback_ran and response.stop_reason != "refusal":
    print(f"Served by {response.model}")
```

A `stop_reason: "refusal"` on the final response means the whole chain refused. The header must be exactly `server-side-fallback-2026-06-01` **for this array form**; the newer `fallbacks: "default"` scalar form uses `server-side-fallback-2026-07-01` instead (see `shared/model-migration.md` -> Migrating to Claude Opus 5 -> New API features), and pairing either header with the other form returns a 400. The parameter is rejected on the Batches API and unavailable on Amazon Bedrock, Vertex AI, and Microsoft Foundry - register the client-side `BetaRefusalFallbackMiddleware` on the client there instead. Full semantics (sticky routing, billing, streaming, echoing fallback turns back): `shared/model-migration.md` -> Migrating to Claude Fable 5.1 -> `refusal` stop reason.

---

## Cost Optimization Strategies

### 1. Use Prompt Caching for Repeated Context

```python
# Automatic caching (simplest - caches the last cacheable block)
response = client.messages.create(
    model="claude-opus-5-5",
    max_tokens=16000,
    cache_control={"type": "ephemeral"},
    system=large_document_text,  # e.g., 50KB of context
    messages=[{"role": "user", "content": "Summarize the key points"}]
)

# First request: full cost
# Subsequent requests: ~90% cheaper for cached portion
```

### 2. Choose the Right Model

```python
# Default to Opus for most tasks
response = client.messages.create(
    model="claude-opus-5-5",  # $4.00/$20.00 per 1M tokens
    max_tokens=16000,
    messages=[{"role": "user", "content": "Explain quantum computing"}]
)

# Use Sonnet for high-volume production workloads
standard_response = client.messages.create(
    model="claude-sonnet-5-5",  # $2.00/$10.00 per 1M tokens
    max_tokens=16000,
    messages=[{"role": "user", "content": "Summarize this document"}]
)

# Use Haiku only for simple, speed-critical tasks
simple_response = client.messages.create(
    model="claude-haiku-5-5",  # $0.10/$0.50 per 1M tokens
    max_tokens=1024,
    output_config={"effort": "low"},
    messages=[{"role": "user", "content": "Classify this as positive or negative"}]
)
```

### 3. Use Token Counting Before Requests

```python
count_response = client.messages.count_tokens(
    model="claude-opus-5-5",
    messages=messages,
    system=system
)

estimated_input_cost = count_response.input_tokens * 0.000004  # $4/1M tokens
print(f"Estimated input cost: ${estimated_input_cost:.4f}")
```

---

## Retry with Exponential Backoff

> **Note:** The Anthropic SDK automatically retries rate limit (429) and server errors (5xx) with exponential backoff. You can configure this with `max_retries` (default: 2). Only implement custom retry logic if you need behavior beyond what the SDK provides.

```python
import time
import random
import anthropic

def call_with_retry(
    client: anthropic.Anthropic,
    max_retries: int = 5,
    base_delay: float = 1.0,
    max_delay: float = 60.0,
    **kwargs
):
    """Call the API with exponential backoff retry."""
    last_exception = None

    for attempt in range(max_retries):
        try:
            return client.messages.create(**kwargs)
        except anthropic.RateLimitError as e:
            last_exception = e
        except anthropic.APIStatusError as e:
            if e.status_code >= 500:
                last_exception = e
            else:
                raise  # Client errors (4xx except 429) should not be retried

        delay = min(base_delay * (2 ** attempt) + random.uniform(0, 1), max_delay)
        print(f"Retry {attempt + 1}/{max_retries} after {delay:.1f}s")
        time.sleep(delay)

    raise last_exception
```

---

### Overview (`skills/claude-api/python/managed-agents/README.md`)

# Managed Agents - Python

> **Bindings not shown here:** This README covers the most common managed-agents flows for Python. If you need a class, method, namespace, field, or behavior that isn't shown, WebFetch the Python SDK repo **or the relevant docs page** from `shared/live-sources.md` rather than guess. Do not extrapolate from cURL shapes or another language's SDK.

> **Agents are persistent - create once, reference by ID.** Store the agent ID returned by `agents.create` and pass it to every subsequent `sessions.create`; do not call `agents.create` in the request path. **Recommended:** define agents and environments as version-controlled files synced with `ant apply` - see `shared/anthropic-cli.md` (its live-docs URL is in `shared/live-sources.md`). The CLI owns the control plane (create/update); your code owns the data plane (sessions with the stored ID). The examples below show in-code creation for when you must provision programmatically; in production the create call belongs in setup, not in the request path.

## Installation

```bash
pip install anthropic
```

## Client Initialization

```python
import anthropic

# Default - resolves credentials from the environment:
# ANTHROPIC_API_KEY, or ANTHROPIC_AUTH_TOKEN, or an `ant auth login` profile.
# Prefer this for local dev; don't hardcode a key.
client = anthropic.Anthropic()

# Explicit API key (only when you must inject a specific key)
client = anthropic.Anthropic(api_key="your-api-key")
```

---

## Create an Environment

```python
environment = client.beta.environments.create(
    name="my-dev-env",
    config={
        "type": "cloud",
        "networking": {"type": "limited", "allow_package_managers": True, "allow_mcp_servers": True},
    },
)
print(environment.id)  # env_...
```

---

## Create an Agent (required first step)

> Warning: **There is no inline agent config.** `model`/`system`/`tools` live on the agent object, not the session. Always start with `agents.create()` - the session only takes `agent={"type": "agent", "id": agent.id}`.

### Minimal

The examples on this page turn both web tools off. Set `enabled` to true on `web_fetch` / `web_search` only when the job as described needs the web (a general-purpose or open-ended job stays off; tell the user how to switch it on) - see `shared/managed-agents-tools.md` § Agent Toolset. They also set the `auto` permission policy, under which a call can pause for your approval - the event loop under Stream Events answers it. When nobody is watching the run, answer `deny`; never answer `allow` to every paused call.

```python
# 1. Create the agent (reusable, versioned)
agent = client.beta.agents.create(
    name="Coding Assistant",
    model="claude-opus-5-5",
    tools=[
        {
            "type": "agent_toolset_20260401",
            "default_config": {"enabled": True, "permission_policy": {"type": "auto"}},
            "configs": [
                {"name": "web_fetch", "enabled": False},
                {"name": "web_search", "enabled": False},
            ],
        },
    ],
)

# 2. Start a session
session = client.beta.sessions.create(
    agent={"type": "agent", "id": agent.id, "version": agent.version},
    environment_id=environment.id,
)
print(session.id, session.status)
print(f"Trace: https://platform.claude.com/workspaces/default/sessions/{session.id}")  # swap 'default' for your workspace ID if the API key is not in the Default workspace
```

### With system prompt and custom tools

```python
import os

agent = client.beta.agents.create(
    name="Code Reviewer",
    model="claude-opus-5-5",
    system="You are a senior code reviewer.",
    tools=[
        {
            "type": "agent_toolset_20260401",
            "default_config": {"permission_policy": {"type": "auto"}},
            "configs": [
                {"name": "web_fetch", "enabled": False},
                {"name": "web_search", "enabled": False},
            ],
        },
        {
            "type": "custom",
            "name": "run_tests",
            "description": "Run the test suite",
            "input_schema": {
                "type": "object",
                "properties": {
                    "test_path": {"type": "string", "description": "Path to test file"}
                },
                "required": ["test_path"],
            },
        },
    ],
)

session = client.beta.sessions.create(
    agent={"type": "agent", "id": agent.id, "version": agent.version},
    environment_id=environment.id,
    title="Code review session",
    resources=[
        {
            "type": "github_repository",
            "url": "https://github.com/owner/repo",
            "mount_path": "/workspace/repo",
            "authorization_token": os.environ["GITHUB_TOKEN"],
            "checkout": {"type": "branch", "name": "main"},
        }
    ],
)
```

---

## Send a User Message

```python
client.beta.sessions.events.send(
    session_id=session.id,
    events=[
        {
            "type": "user.message",
            "content": [{"type": "text", "text": "Review the auth module"}],
        }
    ],
)
```

> Tip: **Stream-first:** Open the stream *before* (or concurrently with) sending the message. The stream only delivers events that occur after it opens - stream-after-send means early events arrive buffered in one batch. See [Steering Patterns](../../shared/managed-agents-events.md#steering-patterns).

---

## Define an Outcome (default kickoff for one deliverable)

When the session's job is one checkable deliverable - an artifact, a report, a PR - kick off with `user.define_outcome` instead of `user.message`: the harness grades each iteration against your rubric and the agent revises until it passes. Send one or the other, never both. See [Outcomes](../../shared/managed-agents-outcomes.md) for the event reference and rubric-writing guidance. The job below reads live prices, so its agent needs `web_search` and `web_fetch` set to `enabled: true`. Under the `limited` networking created above, those tools reach only the hosts in `allowed_hosts`, and with none listed they return no page or search result: list the sites the job reads there (a listed host is also open to the sandbox). When the sites can't be listed in advance, the other mode is `unrestricted`, which gives the whole sandbox full egress, not only the web tools: offer it to the user with that warning, do not choose it for them, and if they take it, keep secrets and sensitive files out of the sandbox (see [Environments](../../shared/managed-agents-environments.md)).

```python
STARTER_RUBRIC = """# Report rubric - starter, tune the criteria
- Output is a single `report.md` in /mnt/session/outputs/
- Every claim cites a source URL
- Includes a summary table with one row per competitor
- Prices are current as of the run date and each row says where it was read from
- No placeholder text, TODOs, or empty sections remain
"""

client.beta.sessions.events.send(
    session_id=session.id,
    events=[
        {
            "type": "user.define_outcome",
            "description": "Write a competitor-pricing report as report.md",
            "rubric": {"type": "text", "content": STARTER_RUBRIC},
            "max_iterations": 5,  # optional; default 3, max 20
        }
    ],
)
```

---

## Stream Events (SSE)

```
/* Detailed source-code truncated for AI context efficiency. */
```

---

## Provide Custom Tool Result

```python
client.beta.sessions.events.send(
    session_id=session.id,
    events=[
        {
            "type": "user.custom_tool_result",
            "custom_tool_use_id": "sevt_abc123",
            "content": [{"type": "text", "text": "All 42 tests passed."}],
        }
    ],
)
```

---

## Poll Events

```python
events = client.beta.sessions.events.list(
    session_id=session.id,
)
for event in events.data:
    print(f"{event.type}: {event.id}")
```

> Warning: **Prefer the SDK over raw `requests`/`httpx`.** If you hand-roll a poll loop, don't assume `timeout=(5, 60)` or `httpx.Timeout(120)` caps total call duration - both are **per-chunk** read timeouts (reset on every byte), so a trickling response can block forever. For a hard wall-clock deadline, track `time.monotonic()` at the loop level and bail explicitly, or wrap with `asyncio.wait_for()`. See [Receiving Events](../../shared/managed-agents-events.md#receiving-events).

---

## Full Streaming Loop with Custom Tools

```
/* Detailed source-code truncated for AI context efficiency. */
```

---

## Upload a File

```python
with open("data.csv", "rb") as f:
    file = client.beta.files.upload(
        file=f,
    )

# Use in a session
session = client.beta.sessions.create(
    agent={"type": "agent", "id": agent.id, "version": agent.version},
    environment_id=environment.id,
    resources=[{"type": "file", "file_id": file.id, "mount_path": "/data.csv"}],
)
```

---

## List and Download Session Files

List files the agent wrote to `/mnt/session/outputs/` during a session, then download them.

```python
# List files associated with a session
files = client.beta.files.list(
    scope_id=session.id,
    betas=["managed-agents-2026-04-01"],
)
for f in files.data:
    print(f.filename, f.size_bytes)
    # Download each file and save to disk
    file_content = client.beta.files.download(f.id)
    file_content.write_to_file(f.filename)
```

> Tip: There's a brief indexing lag (~1-3s) between `session.status_idle` and output files appearing in `files.list`. Retry once or twice if the list is empty.

---

## Session Management

```python
# Get session details
session = client.beta.sessions.retrieve(session_id="sesn_011CZxAbc123Def456")
print(session.status, session.usage)

# List sessions
sessions = client.beta.sessions.list()

# Delete a session
client.beta.sessions.delete(session_id="sesn_011CZxAbc123Def456")

# Archive a session
client.beta.sessions.archive(session_id="sesn_011CZxAbc123Def456")
```

---

## MCP Server Integration

```python
# Agent declares MCP server (no auth here - auth goes in a vault)
agent = client.beta.agents.create(
    name="MCP Agent",
    model="claude-opus-5-5",
    mcp_servers=[
        {"type": "url", "name": "my-tools", "url": "https://my-mcp-server.example.com/sse"},
    ],
    tools=[
        {
            "type": "agent_toolset_20260401",
            "default_config": {"enabled": True, "permission_policy": {"type": "auto"}},
            "configs": [
                {"name": "web_fetch", "enabled": False},
                {"name": "web_search", "enabled": False},
            ],
        },
        {"type": "mcp_toolset", "mcp_server_name": "my-tools"},
    ],
)

# Session attaches vault(s) containing credentials for those MCP server URLs
session = client.beta.sessions.create(
    agent=agent.id,
    environment_id=environment.id,
    vault_ids=[vault.id],
)
```

See `shared/managed-agents-tools.md` §Vaults for creating vaults and adding credentials.

---

### Overview (`skills/claude-api/ruby/claude-api/README.md`)

# Claude API - Ruby

> **Note:** The Ruby SDK supports the Claude API. A tool runner is available in beta via `client.beta.messages.tool_runner()`. Agent SDK is not yet available for Ruby.

## Installation

```bash
gem install anthropic
```

## Client Initialization

```ruby
require "anthropic"

# Default (uses ANTHROPIC_API_KEY env var)
client = Anthropic::Client.new

# Explicit API key
client = Anthropic::Client.new(api_key: "your-api-key")
```

---

## Basic Message Request

```ruby
message = client.messages.create(
  model: :"claude-opus-5-5",
  max_tokens: 16000,
  messages: [
    { role: "user", content: "What is the capital of France?" }
  ]
)
# content is an array of polymorphic block objects (TextBlock, ThinkingBlock,
# ToolUseBlock, ...). .type is a Symbol - compare with :text, not "text".
# .text raises NoMethodError on non-TextBlock entries.
message.content.each do |block|
  puts block.text if block.type == :text
end
```

---

## Extended Thinking

> **Fable 5, Claude Opus 5.5, Claude Opus 5, Opus 4.8, Opus 4.7, Opus 4.6, and Sonnet 4.6:** Use adaptive thinking. `budget_tokens` is removed on Fable 5, Claude Opus 5.5, Claude Opus 5, Opus 4.8, and 4.7 (400 if sent); deprecated on Opus 4.6 and Sonnet 4.6.
> **Claude Opus 5.5:** thinking is always on - omit `thinking` (or send `{ type: "adaptive" }`, which is equivalent); `{ type: "disabled" }` returns a 400 at every effort, as does a thinking budget. Control depth with `output_config.effort` instead - the default is `medium` on this model, where Claude Opus 5 defaults to `high`.
> **Claude Opus 5:** thinking is on by default - omitting `thinking:` runs adaptive (`{ type: "adaptive" }` is equivalent), unlike Opus 4.8/4.7 where omitting it meant no thinking. `{ type: "disabled" }` is accepted only at effort `high` or lower; pairing it with `xhigh`/`max` returns a 400.
> **Older models:** Use `thinking: { type: "enabled", budget_tokens: N }` (must be < `max_tokens`, min 1024).

```ruby
message = client.messages.create(
  model: :"claude-opus-5-5",
  max_tokens: 16000,
  thinking: { type: "adaptive" },
  messages: [{ role: "user", content: "Solve: 27 * 453" }]
)

message.content.each do |block|
  case block.type
  when :thinking then puts "Thinking: #{block.thinking}"
  when :text then puts "Response: #{block.text}"
  end
end
```

---

## Prompt Caching

`system_:` (trailing underscore - avoids shadowing `Kernel#system`) takes an array of text blocks; set `cache_control` on the last block. Plain hashes work via the `OrHash` type alias. For placement patterns and the silent-invalidator audit checklist, see `shared/prompt-caching.md`.

```ruby
message = client.messages.create(
  model: :"claude-opus-5-5",
  max_tokens: 16000,
  system_: [
    { type: "text", text: long_system_prompt, cache_control: { type: "ephemeral" } }
  ],
  messages: [{ role: "user", content: "Summarize the key points" }]
)
```

For 1-hour TTL: `cache_control: { type: "ephemeral", ttl: "1h" }`. There's also a top-level `cache_control:` on `messages.create` that auto-places on the last cacheable block.

Verify hits via `message.usage.cache_creation_input_tokens` / `message.usage.cache_read_input_tokens`.

---

## Stop Details

When `stop_reason` is `:refusal`, the response includes structured `stop_details`:

```ruby
if message.stop_reason == :refusal && message.stop_details
  puts "Category: #{message.stop_details.category}"     # e.g. :cyber, :bio, :reasoning_extraction, :frontier_llm, or nil - see docs for the full set
  puts "Explanation: #{message.stop_details.explanation}"
end
```

**Refusal fallbacks (Claude Fable 5.1) - opt in by default.** Fallbacks are opt-in: without them a refused request simply stops. New `claude-fable-5-1` code should include the server-side `fallbacks` parameter (beta header `server-side-fallback-2026-06-01`, `fallbacks: [{model: "claude-opus-4-8"}]` on the beta messages call) by default. The exact Ruby binding (and the client-side middleware for providers without server-side support) is not documented here - WebFetch the Ruby SDK repo's `examples/` from `shared/live-sources.md`; full semantics in `shared/model-migration.md` -> Migrating to Claude Fable 5.1 -> `refusal` stop reason.

---

## Beta Features

`betas:` is only valid on `client.beta.messages.create`, not the non-beta path.

### Task budgets

```ruby
response = client.beta.messages.create(
  model: :"claude-opus-5-5",
  max_tokens: 16000,
  output_config: { task_budget: { type: :tokens, total: 64_000 } },
  tools: [...],
  messages: [...],
  betas: ["task-budgets-2026-03-13"]
)
```

---

## Error Type

`APIStatusError` exposes a `.type` field for programmatic error classification:

```ruby
begin
  client.messages.create(...)
rescue Anthropic::Errors::APIStatusError => e
  puts e.type  # :rate_limit_error, :overloaded_error, etc.
end
```

---

### Overview (`skills/claude-api/ruby/managed-agents/README.md`)

# Managed Agents - Ruby

> **Bindings not shown here:** This README covers the most common managed-agents flows for Ruby. If you need a class, method, namespace, field, or behavior that isn't shown, WebFetch the Ruby SDK repo **or the relevant docs page** from `shared/live-sources.md` rather than guess. Do not extrapolate from cURL shapes or another language's SDK.

> **Agents are persistent - create once, reference by ID.** Store the agent ID returned by `client.beta.agents.create` and pass it to every subsequent `client.beta.sessions.create`; do not call `agents.create` in the request path. **Recommended:** define agents and environments as version-controlled files synced with `ant apply` - see `shared/anthropic-cli.md` (its live-docs URL is in `shared/live-sources.md`). The CLI owns the control plane (create/update); your code owns the data plane (sessions with the stored ID). The examples below show in-code creation for when you must provision programmatically; in production the create call belongs in setup, not in the request path.

## Installation

```bash
gem install anthropic
```

## Client Initialization

```ruby
require "anthropic"

# Default (uses ANTHROPIC_API_KEY env var)
client = Anthropic::Client.new

# Explicit API key
client = Anthropic::Client.new(api_key: "your-api-key")
```

> Warning: **Trailing underscores:** The Ruby SDK uses `system_:` and `send_(` (trailing underscore) to avoid shadowing `Kernel#system` and `Kernel#send`. Use these forms throughout managed-agents code.

---

## Create an Environment

```ruby
environment = client.beta.environments.create(
  name: "my-dev-env",
  config: {
    type: "cloud",
    networking: {type: "limited", allow_package_managers: true, allow_mcp_servers: true}
  }
)
puts "Environment ID: #{environment.id}" # env_...
```

---

## Create an Agent (required first step)

> Warning: **There is no inline agent config.** `model`/`system_`/`tools` live on the agent object, not the session. Always start with `client.beta.agents.create()` - the session takes either `agent: agent.id` or the typed hash form `agent: {type: "agent", id: agent.id, version: agent.version}`.

### Minimal

The examples on this page turn both web tools off. Set `enabled` to true on `web_fetch` / `web_search` only when the job as described needs the web (a general-purpose or open-ended job stays off; tell the user how to switch it on) - see `shared/managed-agents-tools.md` § Agent Toolset. They also set the `auto` permission policy, under which a call can pause for your approval - the event loop under Stream Events answers it. When nobody is watching the run, answer `deny`; never answer `allow` to every paused call.

```ruby
# 1. Create the agent (reusable, versioned)
agent = client.beta.agents.create(
  name: "Coding Assistant",
  model: :"claude-opus-5-5",
  system_: "You are a helpful coding assistant.",
  tools: [
    {
      type: "agent_toolset_20260401",
      default_config: {permission_policy: {type: "auto"}},
      configs: [
        {name: "web_fetch", enabled: false},
        {name: "web_search", enabled: false}
      ]
    }
  ]
)

# 2. Start a session
session = client.beta.sessions.create(
  agent: {type: "agent", id: agent.id, version: agent.version},
  environment_id: environment.id,
  title: "Quickstart session"
)
puts "Session ID: #{session.id}"
puts "Trace: https://platform.claude.com/workspaces/default/sessions/#{session.id}"  # swap 'default' for your workspace ID if the API key is not in the Default workspace
```

### Updating an Agent

Updates create new versions; the agent object is immutable per version.

```ruby
updated_agent = client.beta.agents.update(
  agent.id,
  version: agent.version,
  system_: "You are a helpful coding agent. Always write tests."
)
puts "New version: #{updated_agent.version}"

# List all versions
client.beta.agents.versions.list(agent.id).auto_paging_each do |version|
  puts "Version #{version.version}: #{version.updated_at.iso8601}"
end

# Archive the agent
archived = client.beta.agents.archive(agent.id)
puts "Archived at: #{archived.archived_at.iso8601}"
```

---

## Send a User Message

```ruby
client.beta.sessions.events.send_(
  session.id,
  events: [{
    type: "user.message",
    content: [{type: "text", text: "Review the auth module"}]
  }]
)
```

> Tip: **Stream-first:** Open the stream *before* (or concurrently with) sending the message. The stream only delivers events that occur after it opens - stream-after-send means early events arrive buffered in one batch. See [Steering Patterns](../../shared/managed-agents-events.md#steering-patterns).

---

## Stream Events (SSE)

```ruby
# Open the stream first, then send the user message
stream = client.beta.sessions.events.stream_events(session.id)

client.beta.sessions.events.send_(
  session.id,
  events: [{
    type: "user.message",
    content: [{type: "text", text: "Summarize the repo README"}]
  }]
)

stream.each do |event|
  case event.type
  in :"agent.message"
    event.content.each { |block| print block.text }
  in :"agent.tool_use" | :"agent.mcp_tool_use"
    puts "\n[Using tool: #{event.name}]"
    if event.evaluated_permission == :ask
      # Paused for your decision (always_ask, or auto with no determination)
      client.beta.sessions.events.send_(
        session.id,
        events: [{
          type: "user.tool_confirmation",
          tool_use_id: event.id,
          # you write approve: ask a person or apply your own rule; deny when unattended
          result: approve(event) ? "allow" : "deny"
        }]
      )
    end
  in :"session.status_idle"
    break unless event.stop_reason.type == :requires_action # waiting on you, keep streaming
  in :"session.error"
    puts "\n[Error: #{event.error&.message || "unknown"}]"
    break
  else
    # ignore other event types
  end
end
```

> Note: Event `.type` is a Symbol (compare with `:"agent.message"`, not `"agent.message"`).

### Reconnecting and Tailing

When reconnecting mid-session, list past events first to dedupe, then tail live events. Answer paused calls as in the loop above, including an `ask` in the history that no `user.tool_confirmation` follows.

```ruby
require "set"

stream = client.beta.sessions.events.stream_events(session.id)

# Stream is open and buffering. List history before tailing live.
seen_event_ids = Set.new
client.beta.sessions.events.list(session.id).auto_paging_each { |past| seen_event_ids << past.id }

# Tail live events, skipping anything already seen
stream.each do |event|
  next if seen_event_ids.include?(event.id)
  seen_event_ids << event.id
  case event.type
  in :"agent.message"
    event.content.each { |block| print block.text }
  in :"session.status_idle"
    # requires_action: answer the paused call as under Stream Events, then keep streaming
    break unless event.stop_reason.type == :requires_action
  else
    # ignore other event types
  end
end
```

---

## Provide Custom Tool Result

> Note: The Ruby managed-agents bindings for `user.custom_tool_result` are not yet documented in this skill or in the apps source examples. Refer to `shared/managed-agents-events.md` for the wire format and the `anthropic` Ruby gem repository for the corresponding params.

---

## Poll Events

```ruby
client.beta.sessions.events.list(session.id).auto_paging_each do |event|
  puts "#{event.type}: #{event.id}"
end
```

---

## Upload a File

```ruby
require "pathname"

file = client.beta.files.upload(file: Pathname("data.csv"))
puts "File ID: #{file.id}"

# Mount in a session
session = client.beta.sessions.create(
  agent: agent.id,
  environment_id: environment.id,
  resources: [
    {
      type: "file",
      file_id: file.id,
      mount_path: "/data.csv"
    }
  ]
)
```

### Add and Manage Resources on an Existing Session

```ruby
# Attach an additional file to an open session
resource = client.beta.sessions.resources.add(
  session.id,
  type: "file",
  file_id: file.id
)
puts resource.id # "sesrsc_01ABC..."

# List resources on the session
listed = client.beta.sessions.resources.list(session.id)
listed.data.each { |entry| puts "#{entry.id} #{entry.type}" }

# Detach a resource
client.beta.sessions.resources.delete(resource.id, session_id: session.id)
```

---

## List and Download Session Files

```ruby
files = client.beta.files.list(scope_id: "sesn_abc123", betas: ["managed-agents-2026-04-01"])
content = client.beta.files.download(files.data[0].id)
File.binwrite("output.txt", content.read)
```

---

## Session Management

```ruby
# List environments
environments = client.beta.environments.list

# Retrieve a specific environment
env = client.beta.environments.retrieve(environment.id)

# Archive an environment (read-only, existing sessions continue)
client.beta.environments.archive(environment.id)

# Delete an environment (only if no sessions reference it)
client.beta.environments.delete(environment.id)

# Delete a session
client.beta.sessions.delete(session.id)
```

---

## MCP Server Integration

```ruby
# Agent declares MCP server (no auth here - auth goes in a vault)
agent = client.beta.agents.create(
  name: "GitHub Assistant",
  model: :"claude-opus-5-5",
  mcp_servers: [
    {
      type: "url",
      name: "github",
      url: "https://api.githubcopilot.com/mcp/"
    }
  ],
  tools: [
    {
      type: "agent_toolset_20260401",
      default_config: {permission_policy: {type: "auto"}},
      configs: [
        {name: "web_fetch", enabled: false},
        {name: "web_search", enabled: false}
      ]
    },
    {type: "mcp_toolset", mcp_server_name: "github"}
  ]
)

# Session attaches vault(s) containing credentials for those MCP server URLs
session = client.beta.sessions.create(
  agent: {type: "agent", id: agent.id, version: agent.version},
  environment_id: environment.id,
  vault_ids: [vault.id]
)
```

See `shared/managed-agents-tools.md` §Vaults for creating vaults and adding credentials.

---

## Vaults

```ruby
# Create a vault
vault = client.beta.vaults.create(
  display_name: "Alice",
  metadata: {external_user_id: "usr_abc123"}
)
puts vault.id # "vlt_01ABC..."

# Add an OAuth credential
credential = client.beta.vaults.credentials.create(
  vault.id,
  display_name: "Alice's Slack",
  auth: {
    type: "mcp_oauth",
    mcp_server_url: "https://mcp.slack.com/mcp",
    access_token: "xoxp-...",
    expires_at: "2026-04-15T00:00:00Z",
    refresh: {
      token_endpoint: "https://slack.com/api/oauth.v2.access",
      client_id: "1234567890.0987654321",
      scope: "channels:read chat:write",
      refresh_token: "xoxe-1-...",
      token_endpoint_auth: {
        type: "client_secret_post",
        client_secret: "abc123..."
      }
    }
  }
)

# Rotate the credential (e.g., after a token refresh)
client.beta.vaults.credentials.update(
  credential.id,
  vault_id: vault.id,
  auth: {
    type: "mcp_oauth",
    access_token: "xoxp-new-...",
    expires_at: "2026-05-15T00:00:00Z",
    refresh: {refresh_token: "xoxe-1-new-..."}
  }
)

# Archive a vault
client.beta.vaults.archive(vault.id)
```

---

## GitHub Repository Integration

Mount a GitHub repository as a session resource (a vault holds the GitHub MCP credential):

```ruby
session = client.beta.sessions.create(
  agent: agent.id,
  environment_id: environment.id,
  vault_ids: [vault.id],
  resources: [
    {
      type: "github_repository",
      url: "https://github.com/org/repo",
      mount_path: "/workspace/repo",
      authorization_token: "ghp_your_github_token"
    }
  ]
)
```

Multiple repositories on the same session:

```ruby
resources = [
  {
    type: "github_repository",
    url: "https://github.com/org/frontend",
    mount_path: "/workspace/frontend",
    authorization_token: "ghp_your_github_token"
  },
  {
    type: "github_repository",
    url: "https://github.com/org/backend",
    mount_path: "/workspace/backend",
    authorization_token: "ghp_your_github_token"
  }
]
```

Rotating a repository's authorization token:

```ruby
listed = client.beta.sessions.resources.list(session.id)
repo_resource_id = listed.data.first.id

client.beta.sessions.resources.update(
  repo_resource_id,
  session_id: session.id,
  authorization_token: "ghp_your_new_github_token"
)
```

---

### Overview (`skills/claude-api/typescript/claude-api/README.md`)

# Claude API - TypeScript

| Feature | Namespace | Key types / call |
|---|---|---|
| User profiles | beta | `client.beta.userProfiles.create(...)` / `.retrieve(id)` / `.list()`. Pass the returned profile id on `client.beta.messages.create`. Requires a beta header - check the SDK's beta-headers reference for the current flag. |

## Installation

```bash
npm install @anthropic-ai/sdk
```

> **Reading local files (ESM):** `__dirname` and `__filename` are **undefined** in ES modules - using either throws `ReferenceError: __dirname is not defined` at runtime. For cwd-relative reads, pass the bare relative path (`fs.readFileSync("./sample.png")`). For script-relative paths, derive the directory from `import.meta.url`: `const here = path.dirname(fileURLToPath(import.meta.url))`. Never write `path.join(__dirname, ...)` in an ESM `.ts` file.

## Client Initialization

```typescript
import Anthropic from "@anthropic-ai/sdk";

// Default - resolves credentials from the environment:
// ANTHROPIC_API_KEY, or ANTHROPIC_AUTH_TOKEN, or an `ant auth login` profile.
// Prefer this for local dev; don't hardcode a key.
const client = new Anthropic();

// Explicit API key (only when you must inject a specific key)
const client = new Anthropic({ apiKey: "your-api-key" });
```

---

## Basic Message Request

```typescript
const response = await client.messages.create({
  model: "claude-opus-5-5",
  max_tokens: 16000,
  messages: [{ role: "user", content: "What is the capital of France?" }],
});
// response.content is ContentBlock[] - a discriminated union. Narrow by .type
// before accessing .text (TypeScript will error on content[0].text without this).
for (const block of response.content) {
  if (block.type === "text") {
    console.log(block.text);
  }
}
```

---

## System Prompts

```typescript
const response = await client.messages.create({
  model: "claude-opus-5-5",
  max_tokens: 16000,
  system:
    "You are a helpful coding assistant. Always provide examples in Python.",
  messages: [{ role: "user", content: "How do I read a JSON file?" }],
});
```

### Mid-conversation system messages (model-gated)

For operator instructions that arrive mid-conversation (mode switches, injected state), append `{role: "system", ...}` to `messages` instead of editing top-level `system` - this preserves the cached prefix and carries operator authority. Must follow a user message (or an `assistant` message ending in server-tool use), and must be either the last entry in `messages` or be followed by an `assistant` turn; cannot be `messages[0]`. Unsupported models return a 400 (`role 'system' is not supported on this model`). See `shared/prompt-caching.md` for when to use this vs. top-level `system`.

```typescript
// No beta header needed - use regular client.messages.create.
const response = await client.messages.create({
  model: MODEL_ID, // must support mid-conversation system messages
  max_tokens: 16000,
  system: [
    { type: "text", text: STABLE_SYSTEM, cache_control: { type: "ephemeral" } },
  ],
  messages: [
    ...history,
    { role: "user", content: userMessage },
    { role: "system", content: "Terse mode enabled - keep responses under 40 words." },
  ],
});
```

---

## Vision (Images)

### URL

```typescript
const response = await client.messages.create({
  model: "claude-opus-5-5",
  max_tokens: 16000,
  messages: [
    {
      role: "user",
      content: [
        {
          type: "image",
          source: { type: "url", url: "https://example.com/image.png" },
        },
        { type: "text", text: "Describe this image" },
      ],
    },
  ],
});
```

### Base64

```typescript
import fs from "fs";

const imageData = fs.readFileSync("image.png").toString("base64");

const response = await client.messages.create({
  model: "claude-opus-5-5",
  max_tokens: 16000,
  messages: [
    {
      role: "user",
      content: [
        {
          type: "image",
          source: { type: "base64", media_type: "image/png", data: imageData },
        },
        { type: "text", text: "What's in this image?" },
      ],
    },
  ],
});
```

---

## Prompt Caching

**Caching is a prefix match** - any byte change anywhere in the prefix invalidates everything after it. For placement patterns, architectural guidance (frozen system prompt, deterministic tool order, where to put volatile content), and the silent-invalidator audit checklist, read `shared/prompt-caching.md`.

### Automatic Caching (Recommended)

Use top-level `cache_control` to automatically cache the last cacheable block in the request:

```typescript
const response = await client.messages.create({
  model: "claude-opus-5-5",
  max_tokens: 16000,
  cache_control: { type: "ephemeral" }, // auto-caches the last cacheable block
  system: "You are an expert on this large document...",
  messages: [{ role: "user", content: "Summarize the key points" }],
});
```

### Manual Cache Control

For fine-grained control, add `cache_control` to specific content blocks:

```typescript
const response = await client.messages.create({
  model: "claude-opus-5-5",
  max_tokens: 16000,
  system: [
    {
      type: "text",
      text: "You are an expert on this large document...",
      cache_control: { type: "ephemeral" }, // default TTL is 5 minutes
    },
  ],
  messages: [{ role: "user", content: "Summarize the key points" }],
});

// With explicit TTL (time-to-live)
const response2 = await client.messages.create({
  model: "claude-opus-5-5",
  max_tokens: 16000,
  system: [
    {
      type: "text",
      text: "You are an expert on this large document...",
      cache_control: { type: "ephemeral", ttl: "1h" }, // 1 hour TTL
    },
  ],
  messages: [{ role: "user", content: "Summarize the key points" }],
});
```

### Verifying Cache Hits

```typescript
console.log(response.usage.cache_creation_input_tokens); // tokens written to cache (~1.25x cost)
console.log(response.usage.cache_read_input_tokens);     // tokens served from cache (~0.1x cost)
console.log(response.usage.input_tokens);                // uncached tokens (full cost)
```

If `cache_read_input_tokens` is zero across repeated identical-prefix requests, a silent invalidator is at work - `Date.now()` or a UUID in the system prompt, non-deterministic key ordering, or a varying tool set. See `shared/prompt-caching.md` for the full audit table.

---

## Extended Thinking

> **Fable 5, Claude Opus 5.5, Claude Opus 5, Opus 4.8, Opus 4.7, Opus 4.6, and Sonnet 4.6:** Use adaptive thinking. `budget_tokens` is removed on Fable 5, Claude Opus 5.5, Claude Opus 5, Opus 4.8, and 4.7 (400 if sent); deprecated on Opus 4.6 and Sonnet 4.6.
> **Claude Opus 5.5:** thinking is always on - omit `thinking` (or send `{ type: "adaptive" }`, which is equivalent); `{ type: "disabled" }` returns a 400 at every effort, as does a thinking budget. Control depth with `output_config.effort` instead - the default is `medium` on this model, where Claude Opus 5 defaults to `high`.
> **Claude Opus 5:** thinking is on by default - omitting `thinking` runs adaptive (`{ type: "adaptive" }` is equivalent), unlike Opus 4.8/4.7 where omitting it meant no thinking. `{ type: "disabled" }` is accepted only at effort `high` or lower; pairing it with `xhigh`/`max` returns a 400.
> **Claude Haiku 5.5:** adaptive thinking only, on by default; `budget_tokens` returns a 400.
> **Older models:** Use `thinking: {type: "enabled", budget_tokens: N}` (must be < `max_tokens`, min 1024).

```typescript
// Fable 5 / Claude Opus 5.5 / Claude Opus 5 / Opus 4.8 / 4.7 / 4.6: adaptive thinking (recommended)
const response = await client.messages.create({
  model: "claude-opus-5-5",
  max_tokens: 16000,
  thinking: { type: "adaptive", display: "summarized" }, // display opt-in: default is omitted (empty thinking text) on Fable 5/5.1, Mythos 5/5.1, Claude Opus 5.5, Claude Opus 5, Opus 4.8/4.7, Claude Sonnet 5.5, and Claude Sonnet 5
  output_config: { effort: "high" }, // low | medium | high | xhigh | max
  messages: [
    { role: "user", content: "Solve this math problem step by step..." },
  ],
});

for (const block of response.content) {
  if (block.type === "thinking") {
    console.log("Thinking:", block.thinking);
  } else if (block.type === "text") {
    console.log("Response:", block.text);
  }
}
```

---

## Error Handling

Use the SDK's typed exception classes - never check error messages with string matching:

```typescript
import Anthropic from "@anthropic-ai/sdk";

try {
  const response = await client.messages.create({...});
} catch (error) {
  if (error instanceof Anthropic.BadRequestError) {
    console.error("Bad request:", error.message);
  } else if (error instanceof Anthropic.AuthenticationError) {
    console.error("Invalid API key");
  } else if (error instanceof Anthropic.RateLimitError) {
    console.error("Rate limited - retry later");
  } else if (error instanceof Anthropic.APIError) {
    console.error(`API error ${error.status}:`, error.message);
  }
}
```

All classes extend `Anthropic.APIError` with a typed `status` field. Check from most specific to least specific. See [shared/error-codes.md](../../shared/error-codes.md) for the full error code reference.

---

## Multi-Turn Conversations

The API is stateless - send the full conversation history each time. Use `Anthropic.MessageParam[]` to type the messages array:

```typescript
const messages: Anthropic.MessageParam[] = [
  { role: "user", content: "My name is Alice." },
  { role: "assistant", content: "Hello Alice! Nice to meet you." },
  { role: "user", content: "What's my name?" },
];

const response = await client.messages.create({
  model: "claude-opus-5-5",
  max_tokens: 16000,
  messages: messages,
});
```

**Rules:**

- Consecutive same-role messages are allowed - the API combines them into a single turn
- First message must be `user`
- Use SDK types (`Anthropic.MessageParam`, `Anthropic.Message`, `Anthropic.Tool`, etc.) for all API data structures - don't redefine equivalent interfaces

---

### Compaction (long conversations)

> **Beta, Fable 5, Claude Opus 5.5, Claude Opus 5, Opus 4.8, Opus 4.7, Opus 4.6, and Sonnet 4.6.** When conversations approach the 200K context window, compaction automatically summarizes earlier context server-side. The API returns a `compaction` block; you must pass it back on subsequent requests - append `response.content`, not just the text.

```typescript
import Anthropic from "@anthropic-ai/sdk";

const client = new Anthropic();
const messages: Anthropic.Beta.BetaMessageParam[] = [];

async function chat(userMessage: string): Promise<string> {
  messages.push({ role: "user", content: userMessage });

  const response = await client.beta.messages.create({
    betas: ["compact-2026-01-12"],
    model: "claude-opus-5-5",
    max_tokens: 16000,
    messages,
    context_management: {
      edits: [{ type: "compact_20260112" }],
    },
  });

  // Append full content - compaction blocks must be preserved
  messages.push({ role: "assistant", content: response.content });

  const textBlock = response.content.find(
    (b): b is Anthropic.Beta.BetaTextBlock => b.type === "text",
  );
  return textBlock?.text ?? "";
}

// Compaction triggers automatically when context grows large
console.log(await chat("Help me build a Python web scraper"));
console.log(await chat("Add support for JavaScript-rendered pages"));
console.log(await chat("Now add rate limiting and error handling"));
```

---

## Stop Reasons

The `stop_reason` field in the response indicates why the model stopped generating:

| Value           | Meaning                                                         |
| --------------- | --------------------------------------------------------------- |
| `end_turn`      | Claude finished its response naturally                          |
| `max_tokens`    | Hit the `max_tokens` limit - increase it or use streaming       |
| `stop_sequence` | Hit a custom stop sequence                                      |
| `tool_use`      | Claude wants to call a tool - execute it and continue           |
| `pause_turn`    | Model paused and can be resumed (agentic flows)                 |
| `refusal`       | Claude refused for safety reasons - check `stop_details`        |

### Structured Stop Details

When `stop_reason` is `"refusal"`, the response includes a `stop_details` object with structured information about the refusal:

```typescript
if (response.stop_reason === "refusal" && response.stop_details) {
  console.log(`Category: ${response.stop_details.category}`); // e.g. "cyber", "bio", "reasoning_extraction", "frontier_llm", or null - see docs for the full set
  console.log(`Explanation: ${response.stop_details.explanation}`);
}
```

### Refusal Fallbacks (Claude Fable 5.1) - opt in by default

Fallbacks are **opt-in**: without them a refused request simply stops. Include the server-side `fallbacks` parameter in `claude-fable-5-1` code by default - on a policy decline the API re-runs the same request on the fallback model inside the same call. A mid-stream decline is billed at normal rates, and the rescue bills at the fallback model's own rates, with cache repricing applied automatically; for a decline before any output, see [How refusals are billed](https://platform.claude.com/docs/en/build-with-claude/refusals-and-fallback#how-refusals-are-billed).

```typescript
const response = await client.beta.messages.create({
  model: "claude-fable-5-1",
  max_tokens: 16000,
  betas: ["server-side-fallback-2026-06-01"],
  fallbacks: [{ model: "claude-opus-4-8" }],
  messages: [{ role: "user", content: "..." }],
});

// Switch points: one fallback block per model that ran and declined this turn
for (const block of response.content) {
  if (block.type === "fallback") {
    console.log(`${block.from.model} declined; ${block.to.model} continued`);
  }
}

// Served-by signal - covers sticky turns, which carry no fallback block.
// Pair with stop_reason: the fallback model can itself refuse.
const fallbackRan = (response.usage.iterations ?? []).some(
  (entry) => entry.type === "fallback_message",
);
if (fallbackRan && response.stop_reason !== "refusal") {
  console.log(`Served by ${response.model}`);
}
```

A `stop_reason: "refusal"` on the final response means the whole chain refused. The header must be exactly `server-side-fallback-2026-06-01` **for this array form**; the newer `fallbacks: "default"` scalar form uses `server-side-fallback-2026-07-01` instead (see `shared/model-migration.md` -> Migrating to Claude Opus 5 -> New API features), and pairing either header with the other form returns a 400. The parameter is rejected on the Batches API and unavailable on Amazon Bedrock, Vertex AI, and Microsoft Foundry - register the client-side `betaRefusalFallbackMiddleware` on the client there instead. Full semantics (sticky routing, billing, streaming, echoing fallback turns back): `shared/model-migration.md` -> Migrating to Claude Fable 5.1 -> `refusal` stop reason.

---

## Cost Optimization Strategies

### 1. Use Prompt Caching for Repeated Context

```typescript
// Automatic caching (simplest - caches the last cacheable block)
const response = await client.messages.create({
  model: "claude-opus-5-5",
  max_tokens: 16000,
  cache_control: { type: "ephemeral" },
  system: largeDocumentText, // e.g., 50KB of context
  messages: [{ role: "user", content: "Summarize the key points" }],
});

// First request: full cost
// Subsequent requests: ~90% cheaper for cached portion
```

### 2. Use Token Counting Before Requests

```typescript
const countResponse = await client.messages.countTokens({
  model: "claude-opus-5-5",
  messages: messages,
  system: system,
});

const estimatedInputCost = countResponse.input_tokens * 0.000004; // $4/1M tokens
console.log(`Estimated input cost: $${estimatedInputCost.toFixed(4)}`);
```

---

### Overview (`skills/claude-api/typescript/managed-agents/README.md`)

# Managed Agents - TypeScript

> **Bindings not shown here:** This README covers the most common managed-agents flows for TypeScript. If you need a class, method, namespace, field, or behavior that isn't shown, WebFetch the TypeScript SDK repo **or the relevant docs page** from `shared/live-sources.md` rather than guess. Do not extrapolate from cURL shapes or another language's SDK.

> **Agents are persistent - create once, reference by ID.** Store the agent ID returned by `agents.create` and pass it to every subsequent `sessions.create`; do not call `agents.create` in the request path. **Recommended:** define agents and environments as version-controlled files synced with `ant apply` - see `shared/anthropic-cli.md` (its live-docs URL is in `shared/live-sources.md`). The CLI owns the control plane (create/update); your code owns the data plane (sessions with the stored ID). The examples below show in-code creation for when you must provision programmatically; in production the create call belongs in setup, not in the request path.

## Installation

```bash
npm install @anthropic-ai/sdk
```

## Client Initialization

```typescript
import Anthropic from "@anthropic-ai/sdk";

// Default - resolves credentials from the environment:
// ANTHROPIC_API_KEY, or ANTHROPIC_AUTH_TOKEN, or an `ant auth login` profile.
// Prefer this for local dev; don't hardcode a key.
const client = new Anthropic();

// Explicit API key (only when you must inject a specific key)
const client = new Anthropic({ apiKey: "your-api-key" });
```

---

## Create an Environment

```typescript
const environment = await client.beta.environments.create(
  {
    name: "my-dev-env",
    config: {
      type: "cloud",
      networking: { type: "limited", allow_package_managers: true, allow_mcp_servers: true },
    },
  },
);
console.log(environment.id); // env_...
```

---

## Create an Agent (required first step)

> Warning: **There is no inline agent config.** `model`/`system`/`tools` live on the agent object, not the session. Always start with `agents.create()` - the session only takes `agent: { type: "agent", id: agent.id }`.

### Minimal

The examples on this page turn both web tools off. Set `enabled` to true on `web_fetch` / `web_search` only when the job as described needs the web (a general-purpose or open-ended job stays off; tell the user how to switch it on) - see `shared/managed-agents-tools.md` § Agent Toolset. They also set the `auto` permission policy, under which a call can pause for your approval - the event loop under Stream Events answers it. When nobody is watching the run, answer `deny`; never answer `allow` to every paused call.

```typescript
// 1. Create the agent (reusable, versioned)
const agent = await client.beta.agents.create(
  {
    name: "Coding Assistant",
    model: "claude-opus-5-5",
    tools: [
      {
        type: "agent_toolset_20260401",
        default_config: { enabled: true, permission_policy: { type: "auto" } },
        configs: [
          { name: "web_fetch", enabled: false },
          { name: "web_search", enabled: false },
        ],
      },
    ],
  },
);

// 2. Start a session
const session = await client.beta.sessions.create(
  {
    agent: { type: "agent", id: agent.id, version: agent.version },
    environment_id: environment.id,
  },
);
console.log(session.id, session.status);
console.log(`Trace: https://platform.claude.com/workspaces/default/sessions/${session.id}`); // swap 'default' for your workspace ID if the API key is not in the Default workspace
```

### With system prompt and custom tools

```typescript
const agent = await client.beta.agents.create(
  {
    name: "Code Reviewer",
    model: "claude-opus-5-5",
    system: "You are a senior code reviewer.",
    tools: [
      {
        type: "agent_toolset_20260401",
        default_config: { enabled: true, permission_policy: { type: "auto" } },
        configs: [
          { name: "web_fetch", enabled: false },
          { name: "web_search", enabled: false },
        ],
      },
      {
        type: "custom",
        name: "run_tests",
        description: "Run the test suite",
        input_schema: {
          type: "object",
          properties: {
            test_path: { type: "string", description: "Path to test file" },
          },
          required: ["test_path"],
        },
      },
    ],
  },
);

const session = await client.beta.sessions.create(
  {
    agent: { type: "agent", id: agent.id, version: agent.version },
    environment_id: environment.id,
    title: "Code review session",
    resources: [
      {
        type: "github_repository",
        url: "https://github.com/owner/repo",
        mount_path: "/workspace/repo",
        authorization_token: process.env.GITHUB_TOKEN,
        checkout: { type: "branch", name: "main" },
      },
    ],
  },
);
```

---

## Send a User Message

```typescript
await client.beta.sessions.events.send(
  session.id,
  {
    events: [
      {
        type: "user.message",
        content: [{ type: "text", text: "Review the auth module" }],
      },
    ],
  },
);
```

> Tip: **Stream-first:** Open the stream *before* (or concurrently with) sending the message. The stream only delivers events that occur after it opens - stream-after-send means early events arrive buffered in one batch. See [Steering Patterns](../../shared/managed-agents-events.md#steering-patterns).

---

## Define an Outcome (default kickoff for one deliverable)

When the session's job is one checkable deliverable - an artifact, a report, a PR - kick off with `user.define_outcome` instead of `user.message`: the harness grades each iteration against your rubric and the agent revises until it passes. Send one or the other, never both. See [Outcomes](../../shared/managed-agents-outcomes.md) for the event reference and rubric-writing guidance. The job below reads live prices, so its agent needs `web_search` and `web_fetch` set to `enabled: true`. Under the `limited` networking created above, those tools reach only the hosts in `allowed_hosts`, and with none listed they return no page or search result: list the sites the job reads there (a listed host is also open to the sandbox). When the sites can't be listed in advance, the other mode is `unrestricted`, which gives the whole sandbox full egress, not only the web tools: offer it to the user with that warning, do not choose it for them, and if they take it, keep secrets and sensitive files out of the sandbox (see [Environments](../../shared/managed-agents-environments.md)).

```typescript
const STARTER_RUBRIC = `# Report rubric - starter, tune the criteria
- Output is a single \`report.md\` in /mnt/session/outputs/
- Every claim cites a source URL
- Includes a summary table with one row per competitor
- Prices are current as of the run date and each row says where it was read from
- No placeholder text, TODOs, or empty sections remain
`;

await client.beta.sessions.events.send(
  session.id,
  {
    events: [
      {
        type: "user.define_outcome",
        description: "Write a competitor-pricing report as report.md",
        rubric: { type: "text", content: STARTER_RUBRIC },
        max_iterations: 5, // optional; default 3, max 20
      },
    ],
  },
);
```

---

## Stream Events (SSE)

```
/* Detailed source-code truncated for AI context efficiency. */
```

---

## Provide Custom Tool Result

```typescript
await client.beta.sessions.events.send(
  session.id,
  {
    events: [
      {
        type: "user.custom_tool_result",
        custom_tool_use_id: "sevt_abc123",
        content: [{ type: "text", text: "All 42 tests passed." }],
      },
    ],
  },
);
```

---

## Poll Events

```typescript
const events = await client.beta.sessions.events.list(
  session.id,
);
for (const event of events.data) {
  console.log(`${event.type}: ${event.id}`);
}
```

---

## Full Streaming Loop with Custom Tools

```
/* Detailed source-code truncated for AI context efficiency. */
```

---

## Upload a File

```typescript
import fs from "fs";

const file = await client.beta.files.upload({
  file: fs.createReadStream("data.csv"),
});

// Use in a session
const session = await client.beta.sessions.create(
  {
    agent: { type: "agent", id: agent.id, version: agent.version },
    environment_id: environment.id,
    resources: [{ type: "file", file_id: file.id, mount_path: "/data.csv" }],
  },
);
```

---

## List and Download Session Files

List files the agent wrote to `/mnt/session/outputs/` during a session, then download them.

```typescript
import fs from "fs";

// List files associated with a session
const files = await client.beta.files.list({
  scope_id: session.id,
  betas: ["managed-agents-2026-04-01"],
});
for (const f of files.data) {
  console.log(f.filename, f.size_bytes);

  // Download and save to disk
  const resp = await client.beta.files.download(f.id);
  const buffer = Buffer.from(await resp.arrayBuffer());
  fs.writeFileSync(f.filename, buffer);
}
```

> Tip: There's a brief indexing lag (~1-3s) between `session.status_idle` and output files appearing in `files.list`. Retry once or twice if the list is empty.

---

## Session Management

```typescript
// Get session details
const session = await client.beta.sessions.retrieve("sesn_011CZxAbc123Def456");
console.log(session.status, session.usage);

// List sessions
const sessions = await client.beta.sessions.list();

// Delete a session
await client.beta.sessions.delete("sesn_011CZxAbc123Def456");

// Archive a session
await client.beta.sessions.archive("sesn_011CZxAbc123Def456");
```

---

## MCP Server Integration

```typescript
// Agent declares MCP server (no auth here - auth goes in a vault)
const agent = await client.beta.agents.create({
  name: "MCP Agent",
  model: "claude-opus-5-5",
  mcp_servers: [
    { type: "url", name: "my-tools", url: "https://my-mcp-server.example.com/sse" },
  ],
  tools: [
    {
      type: "agent_toolset_20260401",
      default_config: { enabled: true, permission_policy: { type: "auto" } },
      configs: [
        { name: "web_fetch", enabled: false },
        { name: "web_search", enabled: false },
      ],
    },
    { type: "mcp_toolset", mcp_server_name: "my-tools" },
  ],
});

// Session attaches vault(s) containing credentials for those MCP server URLs
const session = await client.beta.sessions.create({
  agent: agent.id,
  environment_id: environment.id,
  vault_ids: [vault.id],
});
```

See `shared/managed-agents-tools.md` §Vaults for creating vaults and adding credentials.

---

--- METRICS ---
- Files Extracted: 14
- Estimated Token Budget: ~40864 tokens
- Canonical Reference: https://codewiki.google/github.com/anthropics/skills