Now that you’ve completed the Getting Started guide and read about Claude Code, this is the recommended reading order. Each phase builds on the last — beginners can follow it straight through, experienced engineers can skip to what they need.Documentation Index
Fetch the complete documentation index at: https://docs.output.ai/llms.txt
Use this file to discover all available pages before exploring further.
Phase 1: Core Concepts
Understand the building blocks:- Workflows - Where you orchestrate the work
- Steps - Where the actual work happens
- Prompts - How to manage your LLM interactions
- Evaluators - Quality control for your AI outputs
- API Clients - Call external APIs
Phase 2: Building Real Workflows
Best practices and patterns for building real workflows:- Step Best Practices - Patterns & organization
- Evaluator Best Practices - LLM-as-judge patterns
- Child Workflows - Compose workflows by calling other workflows
- Workflow Context - Understand the execution context
- Running Steps in Parallel - How to execute steps concurrently
- Credentials - Manage secrets and API keys across environments
Phase 3: Production Readiness
Ship with confidence:- Error Handling - Retries, failures, recovery
- Error Hooks - React to failures with custom logic
- Testing - Testing deterministic and non-deterministic code
- Tracing - Observability and debugging
- Cost Estimation - Track and control your AI spend
- Evaluation Workflow - Test workflow quality across datasets
Phase 4: Integrate with Your App
Connect Output to your product:- Integration Overview - How your app talks to Output over HTTP
- API Configuration - Base URL, ports, environment variables
- Authentication - Dev vs production auth
- Error Responses - Error codes and response handling
Reference (As Needed)
- Packages - Package documentation
- External Integration - Send data out, receive input while running
- Prompt Templating - Advanced prompt patterns