Every guide, reference, and example you need to build with Codex. From first install to advanced deployment.
Ten minutes from account creation to your first AI-generated feature — the quickstart guide covers every step.
The getting started documentation walks you through creating a Codex account, choosing a plan, installing the CLI or IDE plugin, and running your first code generation. Every step includes screenshots of the actual interface and copy-pasteable terminal commands. The quickstart is structured so you can follow along in your own project — by the end, you will have generated a working function, reviewed it with the automated code review tool, and pushed it to a repository. The guide also covers authentication setup, API key management, and basic project configuration so your second session is even faster than the first.
For teams onboarding together, the getting started guide includes a team setup section. It covers inviting members, configuring shared project settings, and setting up the GitHub integration so automated code review begins on the first pull request after onboarding. Engineering managers will find the team analytics section useful — it explains how to interpret velocity metrics and adoption dashboards once the team is actively using Codex.
The Codex documentation is organized into six sections — find what you need in seconds.
| Section | Contents | Best For |
|---|---|---|
| Getting Started | Account setup, installation, first generation, team onboarding | New users and team leads |
| CLI Reference | Complete command reference, flags, shell completion, configuration | Terminal-first developers |
| API Reference | REST endpoints, authentication, rate limits, error codes, webhooks | Integration engineers |
| SDK Guides | Python, JavaScript/TypeScript, Go, Java — installation and usage | Application developers |
| Architecture | Platform components, data flow, deployment topologies, scaling | Platform engineers and architects |
| Changelog | Version history, feature additions, breaking changes, deprecations | All users tracking updates |
Every Codex CLI command documented with flags, environment variables, and annotated usage examples.
The CLI reference covers every command in the Codex command-line interface. Each command page includes the full syntax, a table of flags with default values, environment variable equivalents, and at least three annotated examples showing practical usage. The reference is generated from the CLI source code, so it always matches the installed version. Commands are grouped by function: project management, code generation, review, deployment, and configuration. A quick-reference card at the top of the section lists the twenty most-used commands with one-line descriptions — most developers bookmark this page.
Advanced CLI features like custom templates, hook scripts, and output formatting each get dedicated sub-pages. Shell completion setup is documented for Bash, Zsh, Fish, and PowerShell with copy-pasteable configuration snippets. The CLI configuration file reference documents every option with valid values and examples — useful for teams standardizing their Codex setup across repositories.
Complete REST API documentation with request and response schemas for every Codex endpoint.
The API reference documents every REST endpoint in the Codex platform. Each endpoint page includes the HTTP method, path, authentication requirements, request body schema, response body schema, rate limit tier, and a list of possible error codes. Code examples are provided in curl, Python, JavaScript, and Go — all functional and tested against the current API version. The reference also covers webhook event types, payload schemas, and signature verification for building real-time integrations.
Authentication documentation explains API key generation, scoping, rotation, and revocation. Rate limiting is documented with tier-specific limits and headers that help you build well-behaved integrations. The pagination section covers cursor-based navigation for list endpoints. For teams building custom tooling, the API reference includes a Postman collection and an OpenAPI 3.1 specification file that can be imported into your API client of choice.
Idiomatic SDK documentation for four languages — install, authenticate, and start calling the Codex API in minutes.
Codex provides officially supported SDKs for Python, JavaScript/TypeScript, Go, and Java. Each SDK guide covers installation from the language's package registry, authentication using API keys or OAuth2 tokens, and idiomatic usage patterns. The Python SDK guide includes examples for async usage with asyncio, context managers for session handling, and integration with popular frameworks like FastAPI and Django. The TypeScript guide covers both CommonJS and ESM import styles, type definitions, and usage with Next.js and Express.
Every SDK guide includes a "common recipes" section with copy-pasteable code for frequent tasks: generating code from a prompt, running a review on a diff, listing team members, and configuring a project. Error handling patterns are documented per language — try/catch in Python and JavaScript, error return values in Go, and exception hierarchies in Java. The SDK reference is versioned alongside the platform, and the guide includes migration notes when SDK APIs change between major versions.
Understand how Codex components fit together — from the CLI client to the cloud inference engine.
The architecture documentation explains the Codex platform at a systems level. It covers the client layer (CLI, IDE plugins, desktop application), the API gateway, the inference engine, the code analysis pipeline, and the data storage layer. Architecture diagrams show data flow for common operations: a code generation request traveling from the CLI through the gateway to the inference engine and back, a code review scanning a pull request diff and producing a structured report, and a team dashboard query aggregating metrics across repositories.
Deployment topology guides cover single-region cloud deployments, multi-region setups for global teams, and on-premise configurations for regulated industries. The security architecture section documents encryption at rest and in transit, the authentication chain, and the isolation model for multi-tenant cloud deployments. Platform engineers evaluating Codex for enterprise deployment will find the scaling section particularly useful — it covers auto-scaling configuration, database replication, and load balancing strategies.
Every Codex release documented with dates, feature descriptions, and migration notes.
The changelog records every Codex platform release in reverse chronological order. Each entry includes the version number, release date, a summary of new features, a list of bug fixes, and any breaking changes with migration instructions. The changelog follows a structured format — feature additions, improvements, deprecations, and removals each get their own section within a release entry. Links to relevant documentation pages and GitHub issues accompany significant changes.
Codex uses semantic versioning: major versions introduce breaking API or CLI changes, minor versions add backward-compatible features, and patch versions contain bug fixes. The documentation version selector in the site header lets you view docs for any past release. API versions are specified in request headers, and deprecated API versions continue to function for a minimum of six months after a new major version ships. The deprecation schedule is published in the changelog and the API reference.
Start with the Getting Started guide — it takes under ten minutes, produces working code, and establishes a mental model for everything else in the docs.
The Getting Started guide is designed to be completed in a single sitting. You will create an account, install the Codex CLI, authenticate, generate a working function, and review it — all in under ten minutes. This hands-on introduction builds the context you need to navigate the rest of the documentation effectively. After completing the quickstart, the CLI reference and API reference are the natural next stops depending on whether you work primarily in the terminal or build integrations. The documentation is structured so each section stands alone — you do not need to read sequentially to find what you need.
Every endpoint is documented with request schemas, response schemas, and runnable examples in curl, Python, JavaScript, and Go.
The API reference is generated from the OpenAPI specification and tested against the live API. Each endpoint page documents authentication, rate limits, error codes, and pagination behavior. Code examples are functional — you can copy them into your terminal or editor and run them after replacing the placeholder API key. The reference includes a Postman collection for interactive exploration and an OpenAPI 3.1 spec file for generating client libraries in additional languages. Webhook documentation covers event types, payload schemas, delivery headers, and signature verification with code examples in all four SDK languages.
Documentation releases ship in lockstep with platform updates — every new feature, flag, or endpoint is documented before it reaches stable.
The documentation pipeline is integrated into the Codex release process. When a new feature is merged, its documentation must be completed and reviewed before the feature ships to stable. The CLI reference is partially generated from the CLI source code, so flag tables and command descriptions never drift from the actual binary. The API reference is generated from the OpenAPI specification, which is the source of truth for the API surface. The changelog is compiled from merged pull requests with human-written summaries. A documentation version selector lets users view docs for any past release — useful when running an older version of the platform.
Official SDK guides with installation, authentication, and recipes for Python, TypeScript, Go, and Java — each tested against the current API.
Four SDKs are officially supported and documented. The Python guide covers sync and async usage, context managers, and FastAPI/Django integration. The TypeScript guide covers ESM and CommonJS, type definitions, and Next.js/Express patterns. The Go guide emphasizes idiomatic error handling and context usage. The Java guide documents the builder pattern for request construction and exception handling. Each guide includes a "common recipes" section with copy-pasteable code for the most frequent API operations: generating code, running a review, listing team members, and managing project configuration.
The architecture overview covers platform components, data flow, security model, and deployment topologies — essential reading for enterprise platform engineers.
The architecture section of the documentation targets platform engineers and solutions architects evaluating Codex for enterprise deployment. It explains the layered architecture — client, gateway, inference engine, analysis pipeline, and storage — with diagrams for each. The security architecture subsection covers encryption, authentication, and tenant isolation. Deployment topology guides walk through single-region, multi-region, and on-premise configurations with scaling recommendations. For teams planning a production deployment, the architecture docs provide the detail needed to complete a security review and infrastructure provisioning plan.
Whether you are looking to download Codex for the first time, explore the Codex CLI for terminal-native development, or understand how Codex AI transforms your engineering practice, the platform provides integrated tools for every stage of software delivery. The AI code generation engine produces idiomatic code across 40+ languages, while intelligent code review catches bugs before they reach production. Teams can automate testing with the integrated testing suite, debug efficiently with automated debugging, and enforce quality standards with deep code analysis.
Developers integrating Codex into their toolchain start with CLI installation and IDE plugin setup for their preferred editor. The comprehensive API enables custom automation, CI/CD pipeline integration connects Codex to your deployment workflow, and Docker containerization simplifies environment configuration. For deeper integration, see the full documentation covering every feature in detail.