The vendor’s own documentation for each objective of the exam, free to use.
Understand the message lifecycle, tool execution, context window, and architecture that power your SDK agents.
A guided walkthrough from a single tool call to a production-ready agentic loop.
Understand the agentic loop, built-in tools, and how Claude Code interacts with your project.
Compares agent architecture patterns and frameworks, contrasting workflows with autonomous agents and describing common composable building blocks for LLM systems.
Explains programmatic tool calling, where Claude invokes tools from code execution, and how to enable it on tool definitions.
Work with Anthropic-executed tools: server_tool_use blocks, pause_turn continuation, mixed server and client tool turns, and domain filtering.
Give Claude an advisory token budget for the full agentic loop to help the model self-regulate on long agentic tasks.
Use the SDK's tool runner to handle the agentic loop, error wrapping, and type safety automatically.
Scale Claude applications to thousands of tools using semantic embeddings for dynamic tool discovery.
Understand the tool use loop, where tools execute, and when to use tools instead of prose.
Shows how to verify managed agent outputs with an outcome grader, covering outcome definitions, rubrics, citation checks, and the iteration loop.
Two-model team economics via the multiagent coordinator config — a frontier coordinator plans and synthesizes while cheap search workers do the token-heavy web reading in parallel threads, compared against a rigor-matched solo-frontier control with real bills. Covers per-thread usage.list_cost metering, cost attribution, and a session budget as the fan-out guardrail.
Dynamic workflows orchestrate many subagents from a script Claude writes and you can rerun. Use them for codebase audits, large migrations, and cross-checked research.
Compares agent architecture patterns and frameworks, contrasting workflows with autonomous agents and describing common composable building blocks for LLM systems.
Coordinate multiple agents within a single session.
Understand the message lifecycle, tool execution, context window, and architecture that power your SDK agents.
Coordinate multiple Claude Code instances working together as a team, with shared tasks, inter-agent messaging, and centralized management.
Choose between tool search, programmatic tool calling, prompt caching, and context editing to manage context bloat.
Explains cross-session messaging in Claude Code, covering version requirements, how sessions find and message each other, and delivery behavior.
Dynamic workflows orchestrate many subagents from a script Claude writes and you can rerun. Use them for codebase audits, large migrations, and cross-checked research.
Explains using git worktrees to isolate Claude Code sessions, covering creation, trust requirements, switching, and permission prompts for paths.
Create and use specialized AI subagents in Claude Code for task-specific workflows and improved context management.
Define and invoke subagents to isolate context, run tasks in parallel, and apply specialized instructions in your Claude Agent SDK applications.
Introduces advanced tool use features for the Claude API, such as deferred tool loading, programmatic tool calling, and input examples, and their effect on efficiency.
Coordinate multiple Claude Code instances working together as a team, with shared tasks, inter-agent messaging, and centralized management.
Step-by-step guides for exploring codebases, fixing bugs, refactoring, testing, and other everyday tasks with Claude Code.
Follow an agent's workflow runs: their states and events, when the work is done, what a run blocks, budgets, and limits.
Dynamic workflows orchestrate many subagents from a script Claude writes and you can rerun. Use them for codebase audits, large migrations, and cross-checked research.
Trigger a dynamic workflow from the Agent SDK to fact-check a report against source documents with parallel verifier and skeptic subagents.
Let Claude call your tools from code in the code execution container, cutting model round trips and token use in multi-tool workflows.
Walk through a complete two-turn tool-use round trip that preserves thinking blocks correctly, and see how interleaved thinking changes the flow.
Intercept and customize agent behavior at key execution points with hooks
Load custom plugins to extend Claude Code with skills, agents, hooks, and MCP servers through the Agent SDK
Dynamic workflows orchestrate many subagents from a script Claude writes and you can rerun. Use them for codebase audits, large migrations, and cross-checked research.
Attach the ant CLI to a Claude Managed Agents session to follow its transcript live, send messages, allow or deny tool calls, or open the session viewer in your browser.
Track, rewind, and summarize Claude's edits and conversation to manage session state.
List the sessions your users run in Claude apps and agents, such as Claude Cowork and Claude Code, and retrieve their transcripts through the Compliance API.
List, read, rename, tag, and fork Agent SDK sessions on disk to build a conversation history sidebar without writing a transcript parser.
Reference for the removed V2 TypeScript Agent SDK session API, with session-based send/stream patterns for multi-turn conversations.
A guided walkthrough from a single tool call to a production-ready agentic loop.
Run Python and bash code in a sandboxed container to analyze data, generate files, and iterate on solutions.
Give Claude screenshot, mouse, and keyboard control of a desktop environment with the computer use tool, the computer_toolset_20260801 client toolset.
Scale your agent to thousands of tools by discovering and loading only what's needed, on demand.
Learn how to connect Claude Code to your tools with the Model Context Protocol.
Configure MCP servers to extend your agent with external tools. Covers transport types, tool search for large tool sets, authentication, and error handling.
Pair a faster executor model with a higher-intelligence advisor model that provides strategic guidance mid-generation.
A guided walkthrough from a single tool call to a production-ready agentic loop.
Run Python and bash code in a sandboxed container to analyze data, generate files, and iterate on solutions.
Define and invoke subagents to isolate context, run tasks in parallel, and apply specialized instructions in your Claude Agent SDK applications.
Specify tool schemas, write effective descriptions, and control when Claude calls your tools.
Load custom plugins to extend Claude Code with skills, agents, hooks, and MCP servers through the Agent SDK
Serve custom tools from a self-hosted sandbox worker, and wrap an MCP server inside your network as custom tools without running a tunnel.
Configure MCP servers to extend your agent with external tools. Covers transport types, tool search for large tool sets, authentication, and error handling.
Add an MCP server to Claude Code, verify the connection, and find the configuration on disk.
Connect Claude to third-party remote MCP servers through the MCP connector API. Browse example servers and review the steps to connect.
Build production AI agents with Claude Code as a library
Let Claude request shell commands that your application runs in a persistent bash session and returns as tool results.
Understand the agentic loop, built-in tools, and how Claude Code interacts with your project.
Complete reference for the tools Claude Code can use, including permission requirements and per-tool behavior.
Configure tools available to your agent.
A decision map for administrators deploying Claude Code, covering API providers, managed settings, policy enforcement, usage monitoring, and data handling.
Give Claude persistent instructions with CLAUDE.md or AGENTS.md files, and let Claude accumulate learnings automatically with auto memory.
Configure Claude Code for monorepos and large single-tree codebases with nested CLAUDE.md files, sparse worktrees, code intelligence, and per-package skills so Claude stays focused on the code you're working in.
Realistic settings.json files for a developer, a team, and an organization: copy one, keep the keys you want, and change the values.
Define custom tools with the Claude Agent SDK's in-process MCP server so Claude can call your functions, hit your APIs, and perform domain-specific operations.
Control which skills Claude can invoke in Claude Agent SDK sessions, dispatch commands by name, and author skills your sessions discover
Control which skills Claude can invoke in Claude Agent SDK sessions, dispatch commands by name, and author skills your sessions discover
Track token usage, set team spend limits, and reduce Claude Code costs with context management, model selection, extended thinking settings, and preprocessing hooks.
Change Claude Code settings, pick the scope a key belongs in, verify the change, and learn which value Claude Code uses when a key is set in several places.
Configure cloud environments for Claude Code cloud sessions: network access levels, environment variables, setup scripts, and environment caching.
Explains how Claude managed agents discover skills stored in a repository, covering required file locations, frontmatter description matching, and discovery limits.
Explains how to share Claude Code skills through repositories, plugins, and enterprise settings, and how subagents load skills.
Create, manage, and share skills to extend Claude's capabilities in Claude Code. Includes custom commands and bundled skills.
An interactive simulation of how Claude Code's context window fills during a session. See what loads automatically, what each file read costs, and when rules and hooks fire.
Give Claude persistent instructions with CLAUDE.md or AGENTS.md files, and let Claude accumulate learnings automatically with auto memory.
Resolve the Claude Code notice saying this session isn't eligible for auto mode's no-charge classifier requests: what it means, why it appears, and what to do.
Shares best practices for working effectively with Claude Code, including context management, plan mode, scripted headless usage, and permission modes.
Control whether Claude asks before acting. Switch permission modes with Shift+Tab in the CLI, the mode indicator in VS Code, or the mode selector in Desktop.
Build a session-level mode that grants standing consent for multiagent fan-out, switched on and off with mid-conversation system messages.
Outlines Anthropic's policy view on trustworthy AI agents, covering agent components, user permission controls, and defenses against prompt injection.
Behavioral differences and prompting patterns for Claude Fable 5 and Claude Mythos 5, covering effort, instruction following, long runs, memory, and scaffolding changes.
Minimize hallucinations in Claude's outputs by allowing uncertainty, grounding responses in direct quotes, and verifying claims with citations.
Use the Agent SDK to run Claude Code programmatically from the CLI, Python, or TypeScript.
Use channels to push messages, alerts, and webhooks into your Claude Code session from an MCP server. Forward CI results, chat messages, and monitoring events so Claude can react while you're away.
Shares best practices for working effectively with Claude Code, including context management, plan mode, scripted headless usage, and permission modes.
Learn about integrating Claude Code into your development workflow with GitLab CI/CD
Make Claude's outputs more consistent by specifying exact formats, prefilling responses, constraining with examples, and grounding answers in retrieval.
Return validated JSON from agent workflows using JSON Schema, Zod, or Pydantic. Get type-safe, structured data after multi-turn tool use.
Behavioral differences and prompting patterns for Claude Fable 5.1 and Claude Mythos 5.1, covering effort, progress updates, tool-call batching, conversation history, writing style, formatting, task completion, compaction summaries, scope and test coverage, search triggering, safeguard false positives, file edits, long outputs, subagents, and vision.
Comprehensive guide to prompt engineering techniques for Claude's latest models, covering clarity, examples, XML structuring, thinking, and agentic systems.
Set up automated PR reviews that catch logic errors, security vulnerabilities, and regressions using multi-agent analysis of your full codebase
Minimize hallucinations in Claude's outputs by allowing uncertainty, grounding responses in direct quotes, and verifying claims with citations.
Build classification systems with Claude using RAG and chain-of-thought for insurance tickets.
Define measurable success criteria for your LLM application and build evaluations to test it, from exact match checks to LLM-based grading.
Behavioral differences and prompting patterns for Claude Fable 5 and Claude Mythos 5, covering effort, instruction following, long runs, memory, and scaffolding changes.
Comprehensive guide to prompt engineering techniques for Claude's latest models, covering clarity, examples, XML structuring, thinking, and agentic systems.
Behavioral differences and prompting patterns for Claude Sonnet 5, covering effort, adaptive thinking defaults, and tool use.
Prompting patterns specific to Claude Sonnet 5.5: effort, initiative and scope, running without up-front thinking, JSON output, progress updates, tool use, mid-turn messages, coding verification, tool calls, visual inputs, and refusals.
Make Claude's outputs more consistent by specifying exact formats, prefilling responses, constraining with examples, and grounding answers in retrieval.
Build classification systems with Claude using RAG and chain-of-thought for insurance tickets.
This guide is designed to give Claude the basics of using the Claude API. It gives explanation and examples of model IDs/the basic messages API, tool use, streaming, thinking, and nothing else.
Enforce JSON Schema compliance on Claude's tool inputs with grammar-constrained sampling.
Get validated JSON results from agent workflows
Return validated JSON from agent workflows using JSON Schema, Zod, or Pydantic. Get type-safe, structured data after multi-turn tool use.
Parse tool_use blocks, format tool_result responses, and handle errors with is_error.
Understand the HTTP status codes, error response shape, and request IDs the Claude API returns, and handle errors with the SDK's typed exceptions.
Create documents, analyze data, automate workflows with Claude's Excel, PowerPoint, PDF skills.
Explains common Compliance API error responses, how to identify them by status and error type, and how query filters and pagination behave.
Make Claude's outputs more consistent by specifying exact formats, prefilling responses, constraining with examples, and grounding answers in retrieval.
Dynamic workflows orchestrate many subagents from a script Claude writes and you can rerun. Use them for codebase audits, large migrations, and cross-checked research.
Walks through a generator-evaluator feedback loop that retries output with prior attempts, using XML-tagged prompts and Python helper functions.
Define measurable success criteria for your LLM application and build evaluations to test it, from exact match checks to LLM-based grading.
Stream Messages API responses incrementally with server-sent events, including text, tool use, and thinking deltas.
Lists best practices for authoring Agent Skills, covering naming, description writing, file organization, and progressive loading of SKILL.md content.
Get validated JSON results from agent workflows
Return validated JSON from agent workflows using JSON Schema, Zod, or Pydantic. Get type-safe, structured data after multi-turn tool use.
Surface Claude's approval requests and clarifying questions to users, then return their decisions to the SDK.
Process large volumes of Messages requests asynchronously with the Message Batches API, cutting costs by 50% and increasing throughput.
Coordinate multiple Claude Code instances working together as a team, with shared tasks, inter-agent messaging, and centralized management.
Shares best practices for working effectively with Claude Code, including context management, plan mode, scripted headless usage, and permission modes.
Behavioral differences and prompting patterns for Claude Fable 5 and Claude Mythos 5, covering effort, instruction following, long runs, memory, and scaffolding changes.
Set up automated PR reviews that catch logic errors, security vulnerabilities, and regressions using multi-agent analysis of your full codebase
Build a session-level mode that grants standing consent for multiagent fan-out, switched on and off with mid-conversation system messages.
An interactive simulation of how Claude Code's context window fills during a session. See what loads automatically, what each file read costs, and when rules and hooks fire.
Automatically manage conversation context as it grows with context editing.
Choose between tool search, programmatic tool calling, prompt caching, and context editing to manage context bloat.
Understand how the context window works, how extended thinking and tool use count toward it, and how to manage context as conversations grow.
Manage long-running Claude conversations with instant session memory compaction using background threading and prompt caching.
Intercept and customize agent behavior at key execution points with hooks
Understand the message lifecycle, tool execution, context window, and architecture that power your SDK agents.
Coordinate multiple Claude Code instances working together as a team, with shared tasks, inter-agent messaging, and centralized management.
Explains how to configure and use experimental agent teams in Claude Code, covering enabling settings, team roles, and coordination tradeoffs versus subagents.
Understand the HTTP status codes, error response shape, and request IDs the Claude API returns, and handle errors with the SDK's typed exceptions.
To mitigate misuse and manage capacity on the API, limits are in place on how much an organization can use the Claude API.
Explains how Claude API refusals are signaled and billed, and how server-side and client-side model fallback handle safety-declined requests.
Reference for Claude Code hook events, configuration schema, JSON input/output formats, exit codes, async hooks, HTTP hooks, prompt hooks, and MCP tool hooks.
Look up Claude Code runtime error messages with what each one means and how to fix it.
Two-model team economics via the multiagent coordinator config — a frontier coordinator plans and synthesizes while cheap search workers do the token-heavy web reading in parallel threads, compared against a rigor-matched solo-frontier control with real bills. Covers per-thread usage.list_cost metering, cost attribution, and a session budget as the fan-out guardrail.
Follow an agent's workflow runs: their states and events, when the work is done, what a run blocks, budgets, and limits.
Dynamic workflows orchestrate many subagents from a script Claude writes and you can rerun. Use them for codebase audits, large migrations, and cross-checked research.
Describes how inference hooks send prompts and tool calls to an organization's security server for allow or deny verdicts, and how failures are handled.
Coordinate multiple agents within a single session.
Implementing structured error responses for MCP tools with proper categorisation and recovery metadata
Get validated JSON results from agent workflows
Create and use specialized AI subagents in Claude Code for task-specific workflows and improved context management.
An interactive simulation of how Claude Code's context window fills during a session. See what loads automatically, what each file read costs, and when rules and hooks fire.
Automatically manage conversation context as it grows with context editing.
Choose between tool search, programmatic tool calling, prompt caching, and context editing to manage context bloat.
Configure Claude Code for monorepos and large single-tree codebases with nested CLAUDE.md files, sparse worktrees, code intelligence, and per-package skills so Claude stays focused on the code you're working in.
Practical guides and code examples for building with Claude. Learn prompting techniques, tool use, multimodal capabilities, and more.
Explains core evaluation terminology for AI agents, such as tasks, trials, graders, and transcripts, and how evaluation harnesses and suites are structured.
Demonstrates the Claude API citations feature across plain text, PDF, and custom content documents, explaining citation location formats and how chunking affects what gets cited.
Describes how to enable and read citations in Claude API document blocks, covering citation fields and PDF versus plain text source handling.
Describes how the Citations feature attaches source references to Claude's answers, covering document setup and citation location formats for each document type.
Describes how a multi-agent research system coordinates lead and subagents, with analysis of token usage, performance drivers, and scaling effort to query complexity.
Explains how Claude's text watermarking works, what it can and cannot detect, and where its limitations lie, for readers interested in provenance of AI-generated text.
Make Claude's outputs more consistent by specifying exact formats, prefilling responses, constraining with examples, and grounding answers in retrieval.
Minimize hallucinations in Claude's outputs by allowing uncertainty, grounding responses in direct quotes, and verifying claims with citations.