ChatGPT Work Tool and Skill Reference(codex-tool-reference.simonw.chatgpt.site) |
ChatGPT Work Tool and Skill Reference(codex-tool-reference.simonw.chatgpt.site) |
It tells ChatGPT Work how to launch a Playwright instance via its Node.js REPL, and tells it to run:
nodeRepl.write(await browser.documentation());
That method returns this text with full details about how to use the browser: https://codex-tool-reference.simonw.chatgpt.site/skills/cont...
I'd love to know why it's defined like that, as opposed to including all of the instructions directly in the skill Markdown file.
Simply loading the browser Skill alone, doesn't always mean a browser is actually necessary (or supported). The user might've want them to control their main desktop browser; or the user might be pointing them toward a read-only HTML file that doesn't need rendering; or the sandbox might prevent network access; etc...
Codex is especially 'keen' to use Skills (especially by comparison to Claude), I often see it load Skills which it doesn't go on to use, because the main Skill content says "when not to use this". So I'm sure this is a product design decision to help with those kinds of circumstances.
Skill file could indicate general instructions for all browsers. This provides guidance for the browser that's about to actually be used.
Honestly, this is one of the things OpenAI does pretty well, I think, these tiny design decisions that you wouldn't notice until it becomes super annoying.
Are they following some sort of common guideline?
ChatGPT Work in the ChatGPT web and mobile app experience is very different.
If you like, you can think of it as a cloud-hosted version of Codex that's skinned to look more like consumer ChatGPT.
It even shares the same login state and conversation history with /resume, at least when running it in a "project" (which seems to largely map to a directory/repo, at least the one selected as "primary"?)
(1) https://www.newyorker.com/magazine/2026/04/13/sam-altman-may... (2) https://thenextweb.com/news/openai-preparedness-team-disband... (3) https://thezvi.wordpress.com/2026/08/29/metr-and-redwood-off...
That left-hand sidebar needs to be able to scroll on its own, unless you're on a 4k monitor. Otherwise, there's no way to view the bottom half of the list.
Scroll down to the "Establish the visual direction" section.
I have to say, having a background in doing detailed design work makes it much easier to write the right prompt. Also, having the ability to deliver a theme-switcher like this is very handy. There is a tradeoff. Adding this switching took about 1hr 10mins. Last 10% of the result always the most time-consuming.
Dark/light switching is also important.
If I had built this from the ground up manually, I could implement my own design and fiddle with things on a minute level but this type of granularity gets lost when building at speed.
Mind you I don’t spend a lot of time on ai generated websites or the kinds of places you might find those. Well except for HN but you get my point.
Once I can run the Chinese models locally at 60t/sec, they will move to my #1.
I am a bit confused on how Anthropics relationship with Palantir doesn't disqualify them for you.
Anthropic kept changing their terms of service (openclaw agent use) while only using X.com to make/counter their own announcements, doesn’t offer Fable with ZDR, use quota and refuse to give you the work due to guardrails triggering, had some crazy unreliable service periods, continually change what’s offered in their base paid plan, had worse quotas.
Anthropic models ALSO hacked other companies. Antrhopic model was used in the decision making that led to the killing of those iraqi school girls. Talk about blood on their hands.
Anthropic is anti open weights models. OpenAI has at least released several.
If i’m gonna make a devils bargain, I’m gonna pick the one that is more reliable and a better deal.
I argue that you see daylight between them where there is none.