React
React Native
Node.js
Python
Rust
PostgreSQL
Solana
Claude

Claude SME Cookbook

Sections

Claude & LLM Fundamentals

How Large Language Models Actually Generate Text Claude & LLM Fundamentals Basics Inside the Transformer: How Neural Networks Power Claude Generative AI Terms Every Tech Team Should Know LLMs vs Traditional Rule-Based Software: Key Differences How Anthropic Trains and Ships New Claude Models Why Claude Sometimes Gets Things Wrong Claude Compared to Search Engines and Rule-Based Chatbots Claude & LLM Fundamentals Best Practices

How a Claude Conversation Actually Works Context, Tokens & Conversations Basics What Counts as a Token? Tokenization Explained Context Windows: Why Claude Has a Memory Limit System Prompts vs User and Assistant Turns How Claude Remembers (and Forgets) Within a Chat Session Temperature, Top-p, and Top-k: Controlling Claude's Randomness When to Rely on Claude's Extended Thinking Context, Tokens & Conversations Best Practices

Claude Model Family Basics Why Anthropic Offers Four Different Claude Models Claude Fable 5: Inside Anthropic's Top-Tier Mythos-Class Model Claude Opus 4.8: Anthropic's Flagship Reasoning Model Claude Haiku 4.5: Built for Speed and Scale Claude Model Pricing Compared: Fable, Opus, Sonnet, and Haiku Choosing the Right Claude Model for Your Team's Task How Claude Sonnet 5 Became the New Default Model Claude Model Family Best Practices

How Claude Interprets Your Instructions Prompting Claude Basics Why Specificity Beats Cleverness in Prompts Role and Context Framing: Telling Claude Who to Be Few-Shot Examples: Teaching Claude by Demonstration XML Tags for Structuring Complex Prompts: A Quick Reference Common Prompt Ambiguities and How to Avoid Them Iterative Refinement: Turning a Rough Prompt into a Great One Writing a Prompt for a Multi-Step Task in Claude.ai Prompting Claude Best Practices

How Anthropic Approaches AI Safety with Claude AI Safety & Responsible Use Basics Constitutional AI: How Anthropic Trains Claude's Values Claude's Knowledge Cutoff and Why It Matters Why Claude Can't Browse the Web by Default in Chat Anthropic's Usage Policies: What Claude Won't Do A Responsible-Use Checklist for Teams Adopting Claude Reporting and Handling a Claude Safety Concern at Work AI Safety & Responsible Use Best Practices

Everyday Claude & AI Fluency

The 4D AI Fluency Framework Explained AI Fluency Framework Basics Delegation: Deciding What to Hand Off to Claude Description: Crafting Clear Instructions for Claude Delegation Decision Checklist for Everyday Tasks Writing a Project Brief Claude Can Act On Context Windows and Why Description Quality Matters AI Fluency Variants: Students, Educators, Nonprofits, and Builders Delegation & Description Best Practices

Discernment and Diligence: Closing the AI Fluency Loop Discernment & Diligence Basics Why Claude Sometimes Hallucinates Bias in AI Outputs: What to Watch For A Fact-Checking Checklist for Claude's Answers Reviewing an Artifact Line by Line Before Shipping It Diligence as an Iterative Practice, Not a One-Time Check Signs an AI Output Needs a Second Look Discernment & Diligence Best Practices

Applying AI Fluency Basics AI Fluency Across Roles Best Practices How AI Fluency Looks Different Across Roles AI Fluency for Students: Learning With Claude, Not Around It AI Fluency for Educators: Rethinking Assignments and Assessment AI Fluency for Small Businesses and Nonprofits A Builder's Walkthrough: Applying Discernment to a Technical Task Choosing the Right AI Fluency Variant for Your Context Common AI Fluency Pitfalls by Role

Claude 101 Basics What Claude Is and How Conversations Work Claude's Interfaces: Web, Desktop, Mobile, and Slack Setting Up the Claude Desktop App Inside the Claude Desktop App: Chat, Cowork, and Code Getting Started with Claude on Mobile Using the Claude Chrome Extension for In-Browser Help Choosing the Right Claude Interface for the Task Claude in Slack: What It Can and Can't Do Claude 101 Best Practices

Projects and Artifacts: Two Ways to Organize Claude Work Projects & Artifacts Basics Setting Up a Project with Custom Instructions Uploading Knowledge Files to a Project What Makes a Good Project: Scope and Shared Context When to Use an Artifact Instead of a Chat Reply Editing and Iterating on an Artifact Project Setup Checklist: Instructions, Files, and Scope Artifact Types at a Glance: Documents, Code, Diagrams, Apps Projects & Artifacts Best Practices

How Research Mode Works Inside Claude Research Mode, Skills & Tools Basics Running a Research Mode Query Step by Step What Agent Skills Are and How Claude Uses Them Connectors and Enterprise Search: Bringing Outside Data into Claude Evaluating Research Mode Output: Sources and Confidence Connected Tools Checklist Before You Start a Task When to Use Research Mode vs a Regular Conversation Research Mode, Skills & Tools Best Practices

How Claude Adapts to Different Job Functions Role-Specific Use Cases Basics Claude for Writers: Drafting and Editing With Projects Claude for Business Analysts: Summarizing Data and Reports Claude for Marketers: Campaign Ideation and Content at Scale Claude for Customer Support: Drafting Responses and Summarizing Tickets Claude for Sales and Legal Teams: Research and Review Role-Based Prompt Checklist by Job Function Choosing Projects vs Artifacts by Role Role-Specific Use Cases Best Practices

Teams & Enterprise

How Claude Cowork's Local Folder Access Works Claude Cowork Basics Understanding Cowork Subagent Coordination for Parallel Tasks How Cowork Differs from Claude Projects and Plain Chat Types of Tasks Well-Suited to Claude Cowork Why Autonomous File Edits Require Careful Scoping Configuring Scheduled Tasks: Hourly, Daily, and Weekly Cadences Cowork Safety Checklist Before Granting Folder Access Claude Cowork Best Practices

Enterprise Search & Connectors Basics How Claude's Enterprise Search Connects to Your Company's Knowledge First-Party vs MCP-Based Connectors: What's the Difference How Claude Cites Internal Sources in Answers Connectors Available for Claude: Google Drive, Slack, Notion, and More Understanding Connector Permission Scopes and Data Access Connecting Slack and Notion to Claude Step by Step Connector Rollout Checklist for IT Admins Enterprise Search & Connectors Best Practices

How the Claude Console Organizes Workspaces, Seats, and Roles Team & Enterprise Admin Basics Understanding Claude's Plan Tiers: Free, Pro, Team, and Enterprise How Roles and Permissions Control Workspace Access Steps to Provision New Seats in the Claude Console How SSO and SCIM Provisioning Work Together Pulling Per-User Cost and Usage Data with the Enterprise Analytics API Workspace Setup Checklist for New Enterprise Admins Team & Enterprise Admin Best Practices

Governance & Compliance Best Practices Governance & Compliance Basics How Claude Handles Data Retention and Training Opt-Outs Understanding Claude's SOC2 and GDPR Compliance Posture What Audit Logs Capture in the Claude Console Data Handling Controls Every Admin Should Configure How DLP Considerations Apply to Claude Conversations Writing PII Handling Guidance for Teams Pasting Sensitive Data Compliance Checklist Before Rolling Out Claude Enterprise-Wide

Why Claude Answers Get Worse: A Mental Model Fixing Hit Context and Length Limits in Long Conversations What to Do When Claude Hallucinates Citations or Sources Walkthrough: Recovering a Conversation That Lost Context Fixing Inconsistent Formatting in Claude's Responses Projects vs Plain Chats: Choosing the Right One Walkthrough: Diagnosing Rate Limits on Lower Plan Tiers Troubleshooting Quick-Reference Checklist

How Team Rollouts Succeed: A Mental Model for Claude Adoption Onboarding a Team to Claude Basics Identifying Champions and Power Users for Your Claude Rollout How to Measure Adoption Across a Team Steps to Run a Successful Claude Pilot Group Integrating Claude into Slack, Email, and Existing Docs Workflows Usage Policy Checklist for Team-Wide Claude Rollouts Onboarding a Team to Claude Best Practices

Skills & Automation

What Are Agent Skills?Agent Skills Basics How Claude Discovers and Loads Skills Automatically Skills vs Projects vs Custom Instructions: What's the Difference Anatomy of a SKILL.md File Enabling and Managing Skills Across Claude Code, Cowork, and Claude.ai Common Types of Agent Skills Teams Build First Signs Your Skill Isn't Being Discovered by Claude Agent Skills Best Practices

How a Skill Comes Together: From Idea to Package Skill Design & Packaging Basics Writing an Effective Skill Description That Claude Will Trigger On Structuring Step-by-Step Instructions Claude Can Follow Reliably Scoping a Skill's Tool Access With allowed-tools Walkthrough: Bundling Reference Files and Scripts Into a Skill Package Naming Conventions and Metadata Patterns for Discoverable Skills Common Mistakes That Make a Skill Hard for Claude to Find or Use Skill Design & Packaging Best Practices

How Scheduled Tasks and Routines Work in Cowork Scheduled Tasks & Routines Basics Choosing the Right Cadence: Hourly, Daily, or Weekly Routines What Happens When a Scheduled Task Runs Without You Watching Connected Tools and Skills: What a Routine Can Access While Running Setting Up a Weekly Digest Routine Step by Step Common Triggers Worth Automating with Scheduled Claude Tasks Scheduled Tasks & Routines Best Practices

What Multi-Step Delegation Looks Like with Claude Multi-Step Delegation Basics Breaking a Large Task into a Plan Claude Can Execute Checkpoints and Human Review Gates in a Delegated Workflow When to Delegate Fully vs Keep Claude in a Tight Loop Walking Through an Iterative Refinement Cycle with Claude Common Orchestration Patterns for Chaining Skills and Routines Together Failure Modes When Delegation Goes Wrong Multi-Step Delegation & Orchestration Best Practices

Claude Code

Claude Code 101 Best Practices The Claude Code Mental Model: Agent, Not Autocomplete Claude Code 101 Basics Installing Claude Code and Connecting Your First Project Navigating the Claude Code REPL: Prompts and Slash Commands Permission Modes Explained: Auto-Accept vs Ask-Every-Time Switching Between Plan Mode and Execution Mode Terminal, Shell and Hotkey Setup for Claude Code Extended Thinking Mode: When Claude Code Reasons Longer

How Claude Code Executes the Read-Edit-Run-Test Loop Core Claude Code Workflows Basics Searching a Codebase with Claude Code's Built-In Grep Tool Making Multi-File Edits in a Single Claude Code Turn Why Claude Code Re-Reads Files Before Editing Them Running Tests and Iterating on Failures Automatically A Checklist for Reviewing Claude Code's Proposed Diffs How Claude Code Decides When to Run a Command vs Ask Permission Connecting an MCP Server to Extend Claude Code's Toolset Core Claude Code Workflows Best Practices

Context Management Best Practices Context Management Basics What CLAUDE.md Actually Does: Persistent Project Memory Explained Bootstrapping a New Project with /init Referencing Specific Files with @mentions in Your Prompts Why /compact and /clear Exist: Managing the Context Window Writing a CLAUDE.md File That Actually Gets Followed Auditing What's Eating Your Claude Code Context Window Nested CLAUDE.md Files and How Scope Inheritance Works

Custom Commands, Hooks, and Subagents: What Each One Is For Custom Commands & Subagents Basics Writing Your First Custom Slash Command with Markdown Frontmatter Firing a Shell Command on PostToolUse with Hooks How Subagents Get Their Own Context and Tools Blocking Risky File Edits with a PreToolUse Hook Common Hook Events and When to Use Each One Subagents vs the Agent SDK: How Claude Code Is Built Spawning Parallel Subagents for Isolated Research Tasks Custom Commands, Hooks & Subagents Best Practices

How Claude Code Fits Into Your GitHub Pull Request Workflow GitHub Integration Basics Creating a Pull Request from a Claude Code Session with gh Responding to PR Review Comments Automatically Headless Mode Explained: Running Claude Code Without a Terminal Running Claude Code as an Automated Code Review Bot in CI Setting Up a GitHub Actions Job That Runs Claude Code How Claude Code Decides What to Flag in an Automated Review GitHub Integration Best Practices

Claude API

Understanding the Claude API Request Lifecycle Claude API Fundamentals Basics Authenticating Claude API Calls with the x-api-key Header Installing and Configuring the Anthropic Python SDK Claude API Request Parameters Cheatsheet Selecting a Model: Claude Fable 5, Opus 4.8, Sonnet 5, and Haiku 4.5 Building a Reusable Claude API Client Wrapper Handling Rate Limits with Exponential Backoff Claude API Error Codes and Troubleshooting Reference Claude API Fundamentals Best Practices

Understanding Roles and Turns in the Messages API Messages API Basics Building Multi-Turn Conversations with the Messages Array Messages API Roles and Content Block Types Reference Formatting System Prompts for Consistent Claude Behavior Handling Mixed Content Blocks: Text, Image, and Tool Results in One Turn Managing Conversation History and Trimming for the Context Window Implementing Mid-Conversation System Messages on Claude Opus 4.8 Messages API Best Practices

How Server-Sent Events Power Claude's Streaming Responses Streaming Responses Basics Building a Chat UI with Incremental Token Deltas Handling Partial JSON During Streamed Tool Calls Streaming Event Types Reference Streaming Extended Thinking Blocks to the Frontend Combining Streaming with the Tool Use Loop Streaming Best Practices

How Claude Decides When to Call a Tool Tool Use Basics Defining Tool Schemas with name, description, and input_schema Handling tool_use Requests and Returning tool_result Blocks Tool Choice Options Reference Building a Multi-Turn Tool Use Loop Executing Parallel Tool Calls in a Single Turn Scaling to Hundreds of Tools with the Tool Search Tool Tool Search Tool defer_loading Parameters Cheatsheet Connecting to Remote Tools via the MCP Connector Tool Use Best Practices

Structured Outputs Basics Why Structured Outputs Beat Prompt-Based JSON Formatting Defining a JSON Schema for output_config.format Validating and Parsing Structured Responses in Python Structured Output Field Types and Constraints Reference Handling Malformed or Truncated JSON Output Combining Strict Tool Use with Schema-Constrained Outputs Extracting Structured Data from Unstructured Documents Structured Outputs Best Practices

How Prompt Caching's Prefix Match Actually Works Prompt Caching Basics Placing cache_control Breakpoints on System Prompts and Tools Cache TTL Options and Pricing Reference Verifying Cache Hits with cache_read_input_tokens Debugging Silent Cache Invalidators in Long Prompts Caching Strategy for Multi-Turn Agent Conversations Using the Cache Diagnostics Beta to Trace cache_miss_reason Prompt Caching Best Practices

Extended Thinking and the Effort Parameter Explained Extended Thinking, Effort & Multimodal Basics Enabling Adaptive Thinking with thinking: {type: 'adaptive'}Tuning the effort Parameter for Cost and Speed Effort Levels and Thinking Display Options Reference Sending Images for Vision Analysis with Base64 Encoding Uploading and Referencing PDFs with the Files API Analyzing Multi-Page Documents Across Repeated Requests Extended Thinking, Effort & Multimodal Best Practices

SDKs & MCP

The Anthropic Python SDK Mental Model Python SDK Basics Choosing Between Anthropic() and AsyncAnthropic() in Production Code Streaming Responses with the Python SDK Python SDK Retry and Timeout Configuration Reference Type-Hinted Response Models in the Python SDK Handling Streaming Tool Use in Async Python Workflows Python SDK Exception Types at a Glance Building a Custom HTTP Client for the Python SDK Python SDK Best Practices

The Anthropic TypeScript SDK Mental Model TypeScript/JavaScript SDK Basics Async Iterators for Streaming in the TypeScript SDK Running the Anthropic SDK on Edge Runtimes TypeScript SDK Type Reference for Content Blocks Strongly Typed Tool Definitions with Zod in TypeScript Retry, Timeout, and AbortController Patterns in the TypeScript SDK TypeScript SDK Error Class Reference Building a Streaming Chat Endpoint with Next.js and the TypeScript SDK TypeScript/JavaScript SDK Best Practices

Beyond Python and TypeScript: The Other Official Claude SDKs Other Official SDKs Basics Calling Claude from Go with the Official SDK Idiomatic Tool Use Across Go, Java, C#, PHP, and Ruby Migrating an OpenAI SDK Codebase to Claude OpenAI Compatibility Layer Parameter Mapping Reference Streaming Responses in Java and C# with the Official SDKs Other SDKs and OpenAI Compatibility Best Practices

Claude Agent SDK Basics The Claude Agent SDK Mental Model Enabling File Edit, Bash, and Web Tools in the Agent SDK Delegating Work to Subagents in the Claude Agent SDK Claude Agent SDK Configuration Options Reference Adding Human-in-the-Loop Checkpoints to an Agent SDK Workflow Persisting Sessions Across Runs with the Claude Agent SDK Connecting the Claude Agent SDK to MCP Servers Claude Agent SDK Deployment Patterns Reference Claude Agent SDK Best Practices

Understanding MCP: Tools, Resources, and Prompts MCP Core Concepts Basics Defining Callable Tools in an MCP Server Exposing Readable Data as MCP Resources MCP Tools vs Resources vs Prompts Comparison Building Reusable Prompt Templates with MCP Prompts Choosing Between stdio and HTTP/SSE Transports for MCP MCP Tunnels for Private Network Access Reference MCP Core Concepts Best Practices

How MCP Servers Handle Requests MCP Server Basics Building an MCP Server with the Python SDK Building an MCP Server with the TypeScript SDK Building a Custom MCP Client with the Agent SDK stdio vs HTTP/SSE MCP Deployment Comparison Authenticating Remote MCP Servers Testing MCP Servers Before Deployment Versioning and Rate Limiting MCP Tool Schemas Reference MCP Server and Client Best Practices

Production Engineering

How Claude Token Pricing Actually Works Token Economics Basics Sonnet 5 vs Opus 4.8 vs Fable 5: A Cost-Per-Task Comparison Building a Pre-Request Cost Calculator with the Token Counting API Model Tiering: Routing Simple Tasks to Haiku, Hard Tasks to Opus Batch API Economics: When 50 Percent Off Beats Real-Time Calls Prompt Caching ROI: When Cache Writes Actually Pay Off ADR Template: Choosing a Default Model Tier for Your Team Token Economics Best Practices

How Context Engineering Reduces Token Spend Prompt and Context Engineering Basics Trimming Context: Dependency Graphs Instead of Full Repos Summarization Before the Call: Condensing Long Documents for Claude Tuning the Effort Parameter for Cost vs Reasoning Depth Avoiding Redundant Tool Calls in Multi-Turn Agent Loops Context Rot: Why More Tokens Doesn't Mean Better Answers Model Tiering Checklist: Matching Task Complexity to Model Cost ADR Template: Setting a Context Window Budget for a Production Agent Prompt & Context Engineering Best Practices

How the Usage & Cost Admin API Models Your Spend Usage & Cost Tracking Basics Querying Usage by API Key and Workspace with the Admin API Building a Cost Dashboard from Uncached, Cached, and Output Token Breakdowns Filtering Usage by Model, Service Tier, and Data Residency Enterprise Analytics API: Per-User Cost Attribution Across Chat and Claude Code Comparison: Admin API vs Console Usage and Cost Pages ADR Template: A Chargeback Model for Multi-Team API Key Usage Usage & Cost Tracking Best Practices

Observability Basics How Observability Works for LLM Applications Instrumenting Agent Loops with OpenTelemetry Tracing Structured Logging of Prompts, Responses, and Token Counts Integrating Claude Usage and Cost Dashboards with Datadog Alerting on Spend Spikes and Rising Cache-Miss Rates Correlating Latency and Errors with Code Deploys ADR Template: Defining Your OTel Span Schema for Tool Calls Observability Best Practices

Single-Agent Loops vs Multi-Agent Systems: A Mental Model Agentic Orchestration Basics Orchestrator/Worker Patterns with the Claude Agent SDK Building Subagents for Parallel Research and Delegation Prompt Chaining vs Routing: Choosing Your Orchestration Pattern Guardrails for Multi-Agent Systems: Bounding Cost and Scope Error Recovery and Retry Strategies in Agent Loops Evaluating Agent Quality: A Checklist for Production Readiness ADR Template: Single Agent vs Orchestrator for a New Workflow Agentic Orchestration & Multi-Agent Best Practices

Tech-Lead Craft

Understanding Why Claude Agents Fail in Production Troubleshooting & Reliability Basics Diagnosing 429 Rate-Limit Errors and Quota Exhaustion Under Load Retry and Exponential Backoff Strategies for Transient Claude API Failures Handling Malformed Tool-Use JSON and Schema Validation Failures Context Overflow and Streaming Disconnects: A Prevention Checklist Prompt Cache Miss Storms: Diagnosing Sudden Cost and Latency Spikes Root-Cause-Analysis Template for Agent Failures Troubleshooting & Reliability Best Practices

Understanding the Claude Security and RAG Threat Model Security & RAG Pipelines Basics Defending Against Indirect Prompt Injection in RAG-Retrieved Tool Results Least-Privilege Tool-Scoping Checklist for Production Claude Agents Secrets Handling and Preventing Data Exfiltration Through Tool Use SOC2 and GDPR Considerations for PII in Prompts and Logs Chunking Strategies and Hybrid Search for Retrieval-Augmented Claude Pipelines Citation and Grounding Checklist to Reduce RAG Hallucinations Combining RAG Retrieval with Prompt Caching for Cost-Efficient Pipelines Security, Compliance & RAG Best Practices

Choosing Between Direct API, Bedrock, and Vertex AI for Claude Enterprise Deployment Basics Deploying Claude on Amazon Bedrock: IAM, Provisioning, and Cross-Region Inference Running Claude on Google Cloud Vertex AI: Quotas and Service Accounts Data Residency and Region-Pinning Checklist for Regulated Workloads Building a Self-Hosted Gateway for Routing, Caching, and Fallback Comparing Bedrock, Vertex AI, and Direct Anthropic API Access Multi-Provider Fallback Patterns for Gateway Outage Resilience Enterprise Deployment Best Practices

How Opinionated Rules Prevent Claude Production Incidents The Core Claude Rules: A Quick-Reference List Model Selection Rules: When to Use Opus, Sonnet, or Haiku Prompt and Context Hygiene: Rules for Keeping Agents Reliable Tool Design Rules for Safe, Composable Claude Agents Cost Control Rules: Caching, Batching, and Token Budget Discipline Security Guardrail Rules Every Tech Lead Should Enforce Claude Rules Best Practices

What the Claude Certified Architect Exam Actually Tests CCA Foundations Exam Domains at a Glance Studying Agentic Architecture and Orchestration: the CCA Exam's Largest Domain Tool Design, MCP Integration, and Context Management: CCA Domain Checklist Case Study: Building a Production Customer Support Agent End to End Case Study: An Internal Knowledge Assistant with RAG and Citations Case Study: Architecting a Code-Review Bot and Document-Processing Pipeline Before-and-After Case Study: Refactoring a Claude Agent to Cut Cost CCA Foundations Best Practices

/

/

© 2026 @csjcode|SystemsArchitect.io: Cloud Systems Tools|Terms

An independent, unofficial educational resource. Not affiliated with, endorsed, or sponsored by the brand/trademark maintainers/owners.

Not logged in

Search Documentation

Search across all documentation pages

Claude SME/Tags/Python

#Python127 pages

Pages tagged Python across 22 sections.

Agentic Orchestration

Agentic Orchestration & Multi-Agent Best Practices Agentic Orchestration Basics Building Subagents for Parallel Research and Delegation Error Recovery and Retry Strategies in Agent Loops Evaluating Agent Quality: A Checklist for Production Readiness Orchestrator/Worker Patterns with the Claude Agent SDK Single-Agent Loops vs Multi-Agent Systems: A Mental Model

Building MCP Servers

Building a Custom MCP Client with the Agent SDK Building an MCP Server with the Python SDK MCP Server Basics

Claude Agent SDK

Claude Agent SDK Basics Claude Agent SDK Best Practices Claude Agent SDK Configuration Options Reference Connecting the Claude Agent SDK to MCP Servers Delegating Work to Subagents in the Claude Agent SDK Enabling File Edit, Bash, and Web Tools in the Agent SDK The Claude Agent SDK Mental Model

API Fundamentals

Authenticating Claude API Calls with the x-api-key Header Claude API Fundamentals Basics Claude API Fundamentals Best Practices Claude API Request Parameters Cheatsheet Installing and Configuring the Anthropic Python SDK Selecting a Model: Claude Fable 5, Opus 4.8, Sonnet 5, and Haiku 4.5 Understanding the Claude API Request Lifecycle

Context & Tokens

Context, Tokens & Conversations Basics Context, Tokens & Conversations Best Practices How a Claude Conversation Actually Works How Claude Remembers (and Forgets) Within a Chat Session System Prompts vs User and Assistant Turns What Counts as a Token? Tokenization Explained

Enterprise Deployment

Choosing Between Direct API, Bedrock, and Vertex AI for Claude Comparing Bedrock, Vertex AI, and Direct Anthropic API Access Deploying Claude on Amazon Bedrock: IAM, Provisioning, and Cross-Region Inference Enterprise Deployment Basics Enterprise Deployment Best Practices Running Claude on Google Cloud Vertex AI: Quotas and Service Accounts

Extended Thinking & Multimodal

Effort Levels and Thinking Display Options Reference Enabling Adaptive Thinking with thinking: {type: 'adaptive'}Extended Thinking and the Effort Parameter Explained Extended Thinking, Effort & Multimodal Basics Extended Thinking, Effort & Multimodal Best Practices Tuning the effort Parameter for Cost and Speed

Defining Callable Tools in an MCP Server Exposing Readable Data as MCP Resources MCP Core Concepts Basics MCP Core Concepts Best Practices MCP Tools vs Resources vs Prompts Comparison Understanding MCP: Tools, Resources, and Prompts

Building Multi-Turn Conversations with the Messages Array Formatting System Prompts for Consistent Claude Behavior Messages API Basics Messages API Best Practices Messages API Roles and Content Block Types Reference

Instrumenting Agent Loops with OpenTelemetry Tracing Structured Logging of Prompts, Responses, and Token Counts

Calling Claude from Go with the Official SDK Idiomatic Tool Use Across Go, Java, C#, PHP, and Ruby Other Official SDKs Basics Other SDKs and OpenAI Compatibility Best Practices Streaming Responses in Java and C# with the Official SDKs

Cache TTL Options and Pricing Reference How Prompt Caching's Prefix Match Actually Works Placing cache_control Breakpoints on System Prompts and Tools Prompt Caching Basics Prompt Caching Best Practices Verifying Cache Hits with cache_read_input_tokens

Context Engineering

Context Rot: Why More Tokens Doesn't Mean Better Answers How Context Engineering Reduces Token Spend Model Tiering Checklist: Matching Task Complexity to Model Cost Prompt & Context Engineering Best Practices Prompt and Context Engineering Basics Summarization Before the Call: Condensing Long Documents for Claude Trimming Context: Dependency Graphs Instead of Full Repos

Prompting Claude

Common Prompt Ambiguities and How to Avoid Them How Claude Interprets Your Instructions Prompting Claude Basics Prompting Claude Best Practices Role and Context Framing: Telling Claude Who to Be Why Specificity Beats Cleverness in Prompts Writing a Prompt for a Multi-Step Task in Claude.ai

Choosing Between Anthropic() and AsyncAnthropic() in Production Code Python SDK Basics Python SDK Best Practices Python SDK Exception Types at a Glance Python SDK Retry and Timeout Configuration Reference Streaming Responses with the Python SDK The Anthropic Python SDK Mental Model

Citation and Grounding Checklist to Reduce RAG Hallucinations Defending Against Indirect Prompt Injection in RAG-Retrieved Tool Results Least-Privilege Tool-Scoping Checklist for Production Claude Agents Security & RAG Pipelines Basics Security, Compliance & RAG Best Practices SOC2 and GDPR Considerations for PII in Prompts and Logs Understanding the Claude Security and RAG Threat Model

Building a Chat UI with Incremental Token Deltas Handling Partial JSON During Streamed Tool Calls Streaming Best Practices Streaming Event Types Reference Streaming Responses Basics

Token Economics

Building a Pre-Request Cost Calculator with the Token Counting API How Claude Token Pricing Actually Works Model Tiering: Routing Simple Tasks to Haiku, Hard Tasks to Opus Sonnet 5 vs Opus 4.8 vs Fable 5: A Cost-Per-Task Comparison Token Economics Basics Token Economics Best Practices

Building a Multi-Turn Tool Use Loop Defining Tool Schemas with name, description, and input_schema Handling tool_use Requests and Returning tool_result Blocks How Claude Decides When to Call a Tool Tool Choice Options Reference Tool Search Tool defer_loading Parameters Cheatsheet Tool Use Basics Tool Use Best Practices

Context Overflow and Streaming Disconnects: A Prevention Checklist Diagnosing 429 Rate-Limit Errors and Quota Exhaustion Under Load Handling Malformed Tool-Use JSON and Schema Validation Failures Troubleshooting & Reliability Basics Troubleshooting & Reliability Best Practices

Async Iterators for Streaming in the TypeScript SDK TypeScript/JavaScript SDK Basics TypeScript/JavaScript SDK Best Practices

Usage & Cost Tracking

Comparison: Admin API vs Console Usage and Cost Pages Filtering Usage by Model, Service Tier, and Data Residency How the Usage & Cost Admin API Models Your Spend Querying Usage by API Key and Workspace with the Admin API Usage & Cost Tracking Basics Usage & Cost Tracking Best Practices

Related tags

List Basics Best Practices Claude Streaming API Checklist Context