OpenAI Codex CLI: Lightweight coding agent that runs in your terminal

danenania · 2025-04-16T19:24:42 1744831482

Hi folks! If you might be interested in an open source alternative to this/claude code that is actually more capable, efficient, and cost-effective than what can be achieved with a single provider’s models, maybe give Plandex v2[1] a try.

I launched the v2 a few weeks ago and it is now running well—very stable and reliable overall. Benchmarks are on my todo list but I honestly think it’s SOTA tier for agentic AI codegen for large projects and complex tasks. It’s more agentic than aider, more configurable and tightly controlled than Devin, and more provider-agnostic/multi-provider/open source than Claude Code or this new competitor from OpenAI.

I’m still working on getting the very latest models integrated. Gemini Pro 2.5 and these new OpenAI models will be introduced by the end of the week I hope. Currently by default, it supports 2M tokens of context directly and can index massive projects of 20N tokens and beyond.

Very interested to hear HN’s thoughts and feedback. I’m planning a Show HN within the next few days.

1 - https://github.com/plandex-ai/plandex

gizmodo59 · 2025-04-16T19:16:29 1744830989

This is pretty neat! I was able to use it for few use cases where it got it right the first time. The ability to use a screenshot to create an application is nice for rapid prototyping. And good to see them open sourcing it unlike claude.

udbhavs · 2025-04-16T19:25:47 1744831547

Next, set your OpenAI API key as an environment variable:

export OPENAI_API_KEY="your-api-key-here"

Note: This command sets the key only for your current terminal session. To make it permanent, add the export line to your shell's configuration file (e.g., ~/.zshrc).

Can't any 3rd party utility running in the same shell session phone home with the API key? I'd ideally want only codex to be able to access this var

jsheard · 2025-04-16T19:28:59 1744831739

If you run malicious 3rd party code as your main user then you have bigger problems than an OpenAI API key leaking.

mark_mcnally_je · 2025-04-16T18:17:41 1744827461

If one of these tools has broad model support (like aider) it would be a game changer.

lifty · 2025-04-16T19:28:13 1744831693

If aiders creator sees this, any plans on implementing agentic mode, or something more autonomous like claude cli works? Would love to have an independent tool doing that.

terminaltrove · 2025-04-16T18:07:42 1744826862

It's wild that both OpenAI and Anthropic are releasing tools that run in the terminal, especially with a TUI which is what we showcase.

aider was one of the first we listed as terminal tool of the week (0) last year (1)

and we recently listed parllama (3) if you like to run offline and online models in the terminal with a full TUI.

(0) https://terminaltrove.com/tool-of-the-week/

(1) https://terminaltrove.com/aider/

(2) https://terminaltrove.com/parllama/

swyx · 2025-04-16T17:37:22 1744825042

related demo/intro video: https://x.com/OpenAIDevs/status/1912556874211422572

this is a direct answer to claude code which has been shipping furiously: https://x.com/_catwu/status/1903130881205977320

and is not open source; there are unverified comments that they have DMCA'ed decompilations https://x.com/vikhyatk/status/1899997417736724858?s=46

by total coincidence we're releasing our claude code interview later this week that touches on a lot of these points + why code agent CLIs are an actually underrated point in the SWE design space

(TLDR you can use it like a linux utility - similar to @simonw's `llm` - to sprinkle intelligence in all sorts of things like CI/PR review without the overhead of buying a Devin or a Copilot SaaS)

if you are a Claude Code (and now OAI Codex) power user we want to hear use cases - CFP closing soon, apply here https://sessionize.com/ai-engineer-worlds-fair-2025

bigyabai · 2025-04-16T17:44:21 1744825461

  RAM  4‑GB minimum (8‑GB recommended)

It's a CLI...

cryptoz · 2025-04-16T19:08:04 1744830484

Possibly the heaviest "lightweight" CLI tool ever made haha.

brap · 2025-04-16T17:38:00 1744825080

What's the point of making the gif run so fast you can't even see shit

dheera · 2025-04-16T19:24:28 1744831468

People somehow seem to be adverse to making the shift from GIF to H.264

yablak · 2025-04-16T17:42:45 1744825365

That's the model speed :)

brap · 2025-04-16T18:04:51 1744826691

Not really, they don't even give you a second to read the output before it loops back again.

（评论） (comments)

（评论）
(comments)