992 pages for start.
Titles and paths first, then pages whose text carries it. Each row opens that page as this site last read it.
sent_count = len(messages sent in the compaction request)
Claude Developer Platform · in the page
build-with-claude/compaction
…compaction is triggered, the API returns a `compaction` block at the start of the assistant response. A long-running conversation might result in multiple compactions. The last compaction block refl…
compaction-background
Claude Developer Platform · in the page
build-with-claude/compaction-background
…each new turn, don't edit anything already in the history, and don't start another compaction request until this one is swapped in or has failed. 3. When the response arrives with `stop_reason` `"com…
compaction-keep-recent-turns
Claude Developer Platform · in the page
build-with-claude/compaction-keep-recent-turns
…ata model." }, { "role": "assistant", "content": "Start with Recipe, Ingredient, and Step. Add a RecipeIngredient entry that holds the quantity and unit for each ingredient in a re…
compaction-on-demand
Claude Developer Platform · in the page
build-with-claude/compaction-on-demand
…ities in the data model."}, {"role": "assistant", "content": "Start with Recipe, Ingredient, and Step. Add a RecipeIngredient entry that holds the quantity and unit for each ingredient in a re…
compaction-thinking-blocks
Claude Developer Platform · in the page
build-with-claude/compaction-thinking-blocks
…tly the `messages` of a request you already sent. The kept turns then start with Claude's reply to it. * **`system` and the `tools` not marked `defer_loading: true` don't change.** They are the same o…
compaction-threshold
Claude Developer Platform · in the page
build-with-claude/compaction-threshold
…compaction is triggered, the API returns a `compaction` block at the start of the assistant response. A long-running conversation might result in multiple compactions. The last compaction block refl…
context-editing
Claude Developer Platform · in the page
build-with-claude/context-editing
…strategy activates. Once the prompt exceeds this threshold, clearing starts. You can specify this value in either `input_tokens` or `tool_uses`.…
context-windows
Claude Developer Platform · in the page
build-with-claude/context-windows
…r state artifacts so that context recovery is fast when a new session starts. The [memory tool's multisession pattern](https://platform.claude.com/docs/en/agents-and-tools/tool-use/memory-tool#multise…
effort
Claude Developer Platform · in the page
build-with-claude/effort
…Claude Fable 5.1 Claude Fable 5.1 supports all five effort levels. **Start with `high`, the default.** Step up to `xhigh` or `max` for the most capability-sensitive agentic and coding work, and step…
This will automatically use the environment variable VOYAGE_API_KEY.
Claude Developer Platform · in the page
build-with-claude/embeddings
….com/docs/voyageai/models/) in the MongoDB documentation. ## Getting started with Voyage AI To access Voyage AI models, create a model API key in MongoDB Atlas: 1. Sign up for a MongoDB Atlas accou…
extended-thinking
Claude Developer Platform · in the page
build-with-claude/extended-thinking
…/Warning> <Note> If a request fails with a 400 error whose message starts with `"thinking.type.enabled" is not supported`, your model uses adaptive thinking instead. See [Troubleshooting thinking](…
fallback-credit
Claude Developer Platform · in the page
build-with-claude/fallback-credit
…billed as though the conversation had been on the new model from the start. You need this page only when you build the retry yourself: over raw HTTP or with custom retry logic. [Server-side fallback…
fast-mode
Claude Developer Platform · in the page
build-with-claude/fast-mode
…available. Claude Opus 5.5 has a default fast mode rate limit in the Start, Build, and Scale [usage tiers](https://platform.claude.com/docs/en/api/rate-limits), so you don't need to request access to…
files
Claude Developer Platform · in the page
build-with-claude/files
…n/build-with-claude/files#file-expiration)). ### SDK beta namespace Starting with Python SDK 1.2.0, TypeScript SDK 0.122.0, Go SDK 1.68.0, Java SDK 2.59.0, PHP SDK 0.45.0, Ruby SDK 1.67.0, and C# SD…
handling-stop-reasons
Claude Developer Platform · in the page
build-with-claude/handling-stop-reasons
…ames the `code_execution` block that called it. That code has already started: it is paused waiting for your `tool_result` blocks, and sending them resumes the execution instead of starting a deferred…
mid-conversation-effort-example
Claude Developer Platform · in the page
build-with-claude/mid-conversation-effort-example
…ectory you launch the example from, so pointing it at a project means starting it there; when `DOC_TEST_MODE` is set, the harness instead gives bash a small throwaway fixture directory that is removed…
mid-conversation-system-messages
Claude Developer Platform · in the page
build-with-claude/mid-conversation-system-messages
…ort. Every tool declared in `tools` is offered to the model from the start of the conversation unless it is declared with `defer_loading: true`, which keeps it withheld until a `tool_addition` block…
overview
Claude Developer Platform · in the page
build-with-claude/overview
…Manage the documents and data you provide to Claude. If you're new, start with [model capabilities](https://platform.claude.com/docs/en/build-with-claude/overview#model-capabilities) and [tools](htt…
pdf-support
Claude Developer Platform · in the page
build-with-claude/pdf-support
…onverting document information into structured formats ## Before you start ### Check PDF requirements Claude works with any standard PDF. Ensure your request size meets these requirements: | Requi…
preserved-thinking
Claude Developer Platform · in the page
build-with-claude/preserved-thinking
…del can use a thinking block that you send back from an earlier turn. Starting with Claude Fable 5.1, when a `thinking` or `redacted_thinking` block comes back in a request, the API checks the block's…
Request 2 caches its request content (not the response)
Claude Developer Platform · in the page
build-with-claude/prompt-caching
…e-grained control over exactly what gets cached. The simplest way to start is with automatic caching: <CodeGroup> ```bash cURL curl https://api.anthropic.com/v1/messages \ -H "content-type:…
claude-prompting-best-practices
Claude Developer Platform · in the page
build-with-claude/prompt-engineering/claude-prompting-best-practices
…where appropriate. ``` ### Migrating away from prefilled responses Starting with Claude 4.6 models and [Claude Mythos Preview](https://anthropic.com/glasswing), prefilled responses (providing a par…
overview
Claude Developer Platform · in the page
build-with-claude/prompt-engineering/overview
…"> For model-specific tuning guidance for Claude's latest models, start here. </Card> </CardGroup> *** ## When to prompt engineer This guide focuses on success criteria that are controllable…
prompting-claude-fable-5
Claude Developer Platform · in the page
build-with-claude/prompt-engineering/prompting-claude-fable-5
…ey ask for one. Before running a command that changes system state (restarts, deletes, config edits), check that the evidence actually supports that specific action. A signal that pattern-matches to a…
Delivering work
Claude Developer Platform · in the page
build-with-claude/prompt-engineering/prompting-claude-fable-5-1
…ges, but a handful of behavioral differences are worth knowing about. Start with the section that matches what you observe: * Unsure which effort level to run, or latency and cost are higher than the…
prompting-claude-haiku-5-5
Claude Developer Platform · in the page
build-with-claude/prompt-engineering/prompting-claude-haiku-5-5
…xisting Claude Haiku 4.5 prompts should perform well without changes. Start with the section that matches what you observe: * Unsure which effort level to run, or your Claude Haiku 4.5 requests set a…
prompting-claude-opus-4-8
Claude Developer Platform · in the page
build-with-claude/prompt-engineering/prompting-claude-opus-4-8
…token spend, trading off capability for faster speed and lower costs. Start with the `xhigh` effort level for coding and agentic use cases, and use a minimum of `high` effort for most intelligence-sen…
prompting-claude-opus-5
Claude Developer Platform · in the page
build-with-claude/prompt-engineering/prompting-claude-opus-5
…g quality at a fraction of the tokens and latency of higher settings. Start with the default (`high`) and adjust based on your evals: use `low` and `medium` liberally as your primary control for token…
prompting-claude-opus-5-5
Claude Developer Platform · in the page
build-with-claude/prompt-engineering/prompting-claude-opus-5-5
…laude/prompt-engineering/prompting-claude-opus-5) remain a reasonable starting point. Start with the section that matches what you observe: * You're unsure which effort level to run, or turns run lon…
prompting-claude-sonnet-5
Claude Developer Platform · in the page
build-with-claude/prompt-engineering/prompting-claude-sonnet-5
…little texts on corners bottom centre like that. For the structure, start with a hero section containing a strong product statement, one short supporting paragraph, and a clean product placeholder o…
prompting-claude-sonnet-5-5
Claude Developer Platform · in the page
build-with-claude/prompt-engineering/prompting-claude-sonnet-5-5
…ude/prompt-engineering/prompting-claude-sonnet-5) remain a reasonable starting point. For the hardest long-horizon work, an Opus model is the better choice. Start with the section that matches what yo…
refusals-and-fallback
Claude Developer Platform · in the page
build-with-claude/refusals-and-fallback
…e occurs. **When the decline occurs before any output:** * `message_start` names the fallback model, and the `fallback` block is the first content block. * Because `message_start` waits for the fall…
search-results
Claude Developer Platform · in the page
build-with-claude/search-results
…rue}, ), ] # Build up the conversation in a list, starting with the user's question messages: list[MessageParam] = [ {"role": "user", "content": "How do I configure the tim…
skills-guide
Claude Developer Platform · in the page
build-with-claude/skills-guide
…on). </Note> ## Quick links <CardGroup cols={2}> <Card title="Get started with Agent Skills in the API" icon="rocket" href="https://platform.claude.com/docs/en/agents-and-tools/agent-skills/quicks…
streaming
Claude Developer Platform · in the page
build-with-claude/streaming
…in its data. Each stream uses the following event flow: 1. `message_start`: contains a `Message` object with empty `content`. Under the [`thinking-binding-controls-2026-08-01`](https://platform.clau…
task-budgets
Claude Developer Platform · in the page
build-with-claude/task-budgets
…n span several requests. A user message that carries no tool results starts a new turn with a fresh budget. Today, the countdown still counts earlier turns' history while it remains in the context. A…
thinking
Claude Developer Platform · in the page
build-with-claude/thinking
…inking": "Let me break this down. The question has two parts, so I'll start with the simpler one and use its result to constrain the second...", "signature": "WaUjzkypQ2mUEVM36O2Txu...." },…
thinking-steering-and-cost
Claude Developer Platform · in the page
build-with-claude/thinking-steering-and-cost
…lock. Don't build application logic that assumes every assistant turn starts with one. The primary control over this determination is the [effort](https://platform.claude.com/docs/en/build-with-claud…
thinking-tool-workflows
Claude Developer Platform · in the page
build-with-claude/thinking-tool-workflows
…r. Where forced tool use is accepted, it skips thinking: the response starts with the tool call. * [Keep one thinking configuration per assistant turn](https://platform.claude.com/docs/en/build-with-c…
thinking-troubleshooting
Claude Developer Platform · in the page
build-with-claude/thinking-troubleshooting
…ng configurations and the ones it rejects; the sections after it each start from a symptom you observe, so you can match an error message or unexpected response directly to its cause and fix. To learn…
token-counting
Claude Developer Platform · in the page
build-with-claude/token-counting
…uests per minute (RPM) | | ---------- | ------------------------- | | Start | 5,000 | | Build | 10,000 | | Scale | 20,000 | <N…
vision
Claude Developer Platform · in the page
build-with-claude/vision
…"image" href="https://platform.claude.com/cookbook/multimodal-getting-started-with-vision"> Get tips and best-practice techniques for tasks such as interpreting charts and extracting content from…
working-with-messages
Claude Developer Platform · in the page
build-with-claude/working-with-messages
…laude 4.7 and later models and Claude Mythos Preview, or on any model starting with [API version `2026-10-21`](https://platform.claude.com/docs/en/api/migration-2026-10-21#sampling-parameters). Settin…
API usage primer for Claude
Claude Developer Platform · in the page
claude_api_primer
…Where it's accepted, a forced tool call skips thinking: the response starts with the tool call and has no `thinking` block. 2. **Preserving thinking blocks:** During tool use, you must pass `thinking…
apply
Claude Developer Platform · in the page
cli-sdks-libraries/cli/apply
…ing new ones. To install and authenticate the CLI, see the [CLI quickstart](https://platform.claude.com/docs/en/cli-sdks-libraries/cli/quickstart). `ant apply` requires CLI version 1.30.0 or later.…
On a remote host without a browser:
Claude Developer Platform · in the page
cli-sdks-libraries/cli/authentication
…on. --- The `ant` CLI supports several credential sources. The [Quickstart](https://platform.claude.com/docs/en/cli-sdks-libraries/cli/quickstart#authentication) covers the one-command login (`ant au…
scripting
Claude Developer Platform · in the page
cli-sdks-libraries/cli/scripting
…xist, you can drive a session from the shell: <Steps> <Step title="Start a session"> Pass the agent and environment IDs to the session create command. After `ant apply`, read them from `claude-…
sessions-connect
Claude Developer Platform · in the page
cli-sdks-libraries/cli/sessions-connect
…CLI. To install or update the CLI and authenticate, see the [CLI quickstart](https://platform.claude.com/docs/en/cli-sdks-libraries/cli/quickstart). ## Connect to a session Pass the ID of a session…
using
Claude Developer Platform · in the page
cli-sdks-libraries/cli/using
…ply across every endpoint. To install and authenticate, see the [Quickstart](https://platform.claude.com/docs/en/cli-sdks-libraries/cli/quickstart). To chain commands and version-control resources, se…
apple-foundation-models
Claude Developer Platform · in the page
cli-sdks-libraries/libraries/apple-foundation-models
…mport FoundationModels import ClaudeForFoundationModels ``` ## Quick start `ClaudeLanguageModel` is the entry point. Pass it to `LanguageModelSession` and use the session exactly as you would with a…