OpenAI Codex CLI
https://github.com/openai/codex
OpenAI's open-source terminal coding agent — released April 2025 alongside o3/o4-mini. Counterpart to Claude Code in the OpenAI stack. Memory via ~/.codex/memories (file-backed); routes to OpenAI o-series and Codex (model) for execution. Distinct from the older OpenAI Codex API (deprecated 2023) and from the separate ChatGPT "Codex" cloud agent product.
At a glance
- Type
- Terminal-native agentic coding harness (OpenAI)
- Tier
- T1
- Section
- Agent IDEs & coding harnesses
- Created
- 2025-04
- Latest release
- codex-cli rolling releases (2025-2026)
- License
- Apache-2.0
- GitHub
- ~40k★, Rust + TypeScript
- Pricing
- Free (OSS); usage via OpenAI API
- Funding
- OpenAI (parent) — $40B Series I 2025-03 $300B val
Taxonomy
- storage
- file
- retrieval
- injection
- persistence
- cross-session
- update
- hybrid
- unit
- document
- governance
- inspectable
- conflict
- human-arbitrate
When to use
Optimised for: OpenAI o-series + Codex model as long-running coding agent
Anti-fit: OpenAI-model-only; CLI-only
Pros & cons
Pros
OSS Apache-2.0 means hackable by community; first-party OpenAI distribution + auto-routing to Codex / o-series.
Cons
OpenAI-only model lock; younger than Claude Code with less ecosystem (skills/hooks); naming collision with deprecated old Codex API and ChatGPT "Codex" cloud agent.
Claims & capabilities
Released 2025-04 as OSS Apache-2.0 CLI; positioned as OpenAI's answer to Claude Code; ~40k★ in first 6 months.
Technical surface
- API surface
- CLI (Rust); MCP support
- Backend storage
- local ~/.codex/ + git
- Deployment
- Self-host CLI; API calls to OpenAI cloud
- Embedding model
- not applicable — not a memory product
- Multi-tenancy
- per-user (terminal-local)
- MCP
- native — Codex CLI is an MCP client
- A2A
- searched not found
- OpenTelemetry
- searched not found
Similar systems
Other agent ides & coding harnesses in the catalog, ranked by inbound references.
- Bedrock AgentCore (AWS) T2
AWS Bedrock AgentCore — managed runtime to deploy + scale + observe agents. Multi-component: Runtime, Identity, Memory, Gateway, Browser, Code Interpreter, Observability. Distinct from Bedrock Agents (older fixed-config product); AgentCore is the 2025 framework-agnostic runtime for any agent (LangGraph, CrewAI, Strands, OpenAI Agents SDK).
- Claude Code (Anthropic) T1
Anthropic's official agentic coding CLI — runs Claude as a long-running coding agent in the developer's terminal. Memory layer = CLAUDE.md (project + user scoped), agent skills, MCP servers. Released GA October 2024; widely adopted as the reference harness for Claude-as-coder. Plan Mode and Compute Use surfaces extend it from coding to broader desktop automation.
- GitHub Copilot (Agent Mode) T1
The original AI coding assistant (Copilot in editor; 2021). 2025 evolved into a full agentic surface: chat, completions, Agent Mode (multi-step), and Copilot Workspace. Backed by Microsoft. Distinct row from Workspace — this row covers the everyday in-editor + agent mode product, Workspace covers the hosted spec-driven workflow.
- Aider (harness) T2
OSS terminal coding agent by Paul Gauthier — one of the original AI-pair-programmer CLIs (2023). Architect/Editor modes; repo-map for context; uses any LiteLLM-compatible model. Memory via CONVENTIONS.md and .aider.conf.yml — author-edited markdown. Distinct from the separate Aider memory-row in "Coding-agent memory"; this row characterises the harness itself.
- Amazon Q Developer T1
Amazon's developer-focused agent — formerly CodeWhisperer (renamed 2024-Q4). Available as IDE plugins (VS Code, JetBrains, Visual Studio, Eclipse), in AWS Console, and CLI. Strong AWS-tooling integration: Q can read CloudWatch, troubleshoot IAM, suggest AWS-SDK code. Companion to Kiro for spec-driven flows.
- Anthropic Computer Use T1
Anthropic's API-surface capability that lets Claude control a virtual desktop — see screenshots, type, click. Launched 2024-10 as a public beta. Not a finished IDE itself; it's the substrate that other harnesses (and Anthropic's own Claude apps) use to operate a computer.
Related systems
References (2)
- Model Context Protocol (MCP spec) depends on at runtime — adjacent-infrastructure cell: OpenAI API (Codex / o-series); MCP for tool extensibility; runs locally with TLS-only network egress
- OpenAI GPT family (GPT-5 / GPT-4o / o3 / o4) depends on at runtime — adjacent-infrastructure cell: OpenAI API (Codex / o-series); MCP for tool extensibility; runs locally with TLS-only network egress