@JacobColling Good question. Not sure if you can override the default plan mode approach, but I'll test that. For now I use a skill to get there.
English
Jarad Johnson
9.6K posts
Jarad Johnson
@jdjohnson
Trying to be better. Building @mostly_serious (https://t.co/Ikj59v8exR).
Springfield, Missouri Katılım Nisan 2007
654 Takip Edilen3.2K Takipçiler
I wish plan mode in Claude Code / Codex asked more open ended questions instead of multiple choice with recommendations.
I think it unintentionally damages critical thinking. I'd prefer more of an interview approach.
English
Can be a real challenge when sharing skills across Claude Code and Codex.
Ethan Mollick@emollick
Very different philosophies for skills in Codex versus Claude Code OpenAI seems to conceive of skills functionally, mostly matter-of-fact technical references for Codex. Claude skills are more about giving the AI approaches to problems See the difference in skill creator skills
English
Jarad Johnson retweetledi
The way I use AI today reminds me of how people used Sonnet 3.5 for coding in 2024. I prompt the LLM, it generates useful output, I incorporate that output into my work product.
The next step is "Claude Code for lawyers", which I assume we'll be seeing before too long.
For now, I am definitely more productive and definitely not (yet) unnerved. Very curious to see how long it will take AI to proliferate through the profession.
English
Once a week I spend time in the Codex app and Claude desktop app to see how they're coming along.
After about an hour in each, I go back to Ghostty.
English
Jarad Johnson retweetledi
Enjoy this brief period of time when agents are powerful but they are not fully workable via mobile yet. Soon there will be no escape and the psychosis that Karpathy talked about won't leave you. Right now you can go to the park, go for drinks, meet your family and you can be safe in the knowledge that there's not much you can do productively.
In a matter of weeks this will go away, Anthropic and OpenAI will build out their mobile offerings properly and you'll have no excuse not to check in on your agents and see what else they can do for you - whether you are in shopping, walking your dog or on the toilet.
English
After going through a discovery phase, I often want to reset context before having Codex build the full plan. This is a small thing that I run into many times a day.
Bonus points if starting on the plan also resets context like Claude Code does.
English
Two quality of life improvements I'd love to see in Codex CLI, @OpenAIDevs:
1. While compacting, if I queue a message, it will break instead of continuing after compacting is complete.
2. You can't switch to plan mode while compacting.
English
For non-coding work, Claude Code has been worse than Codex recently. It regularly doesn't follow instructions, makes incorrect assumptions, and doesn't fully complete tasks.
The voice still slightly edges Codex, but even that advantage is narrowing.
English
@deanwball I’ve been very surprised by the lack of progress with Pulse. A daily briefing is very powerful and could be extended through apps/connectors to do real work overnight. Also surprised Claude hasn’t entered the space.
English
I hope that in the “refocusing” OpenAI does not drop Pulse, which I find insanely useful for surfacing important but under-the-radar news items almost daily.
English
Interested if OpenAI sees Codex as its Claude Cowork response.
I only know one non-technical person using Codex. The rest are in love with either Cowork or Perplexity Computer.
English
@ClementDelangue I think this is one of the worst use cases for generative AI. Can be helpful for taking brain dumps and organizing them, but AI generated pre-written replies just remove humanity and cause confusion.
English
why don't all messaging apps (slack, whatsapp, gmail, linkedin,...) have great AI powered suggested replies and great autocompletes? Sounds kind of trivial given they have your voice and history and would save everyone tons of time.
Too costly maybe with big models (😉😉😉 smaller open-source on-device models)?
English
@JacobColling FWIW Granola's founder seems to get it (full thread): x.com/cjpedregal/sta…
I'm an advocate for Granola having used quite a few of these. It's the best I've used and the MCP does what I need to work with agents. API would be great for special use cases, though.
Chris Pedregal@cjpedregal
@appenz @meetgranola Hey Guido, we're not going closed. Our MCP lets you access everything. If you have feedback on why MCP isn't helpful, or what else you'd prefer, let me know plz
English
I was just going to re-sub to Granola but reconsidering after seeing this.
Best note taking app that plays well with agents?
Guido Appenzeller@appenz
Sorry to see Granola @meetgranola going closed. They encrypted their local db, no local and no cloud API. In a world where notes are managed by agents, the app now has zero value. Any recommendations for good alternatives? What are you switching to?
English
@_simonsmith Do you use GPT-5.4 frequently in Codex as well? I find it to be less verbose there and capable of matching Opus. But still prefer Opus in the main apps.
English
I go back and forth between GPT-5.4 and Claude Opus 4.6. The former feels a bit smarter and more accurate with search, but can be too verbose. The latter is way better at design and a bit more personable, but can be sycophantic ("great question") and naive with search (e.g. giving answers that are obviously outdated based on what it finds, without critically assessing the search results).
If the next GPT has massively improved design skills and tighter writing, and the next Opus dials back the sycophancy and elevates reasoning and critical thinking, the model families will converge and usage will likely come down to switching inertia and non-model preferences like harnesses, UIs, and personality.
English
We're road tripping for spring break and the random Texas town we stopped in last night had an annual event like this going on. This version of America is everywhere unless you actively avoid participating in it.
English
Every time I see these posts they are just from people who seemingly make no effort at all to experience community.
Sassafrass84@Sassafrass_84
Me too.
English
Jarad Johnson retweetledi
"The sooner David Ellison takes over that network, the better."
The Defense Secretary calling for a billionaire ally of the government to take over a media outlet & make it more pro-government. This is what authoritarianism looks like. But in other countries they try & hide it.
Acyn@Acyn
Hegseth: No quarter, no mercy for our enemies. Yet some in the press just can't stop. More fake news from CNN reports that the Trump administration underestimated the Iran war's impact on the strait of hormuz. The sooner David Ellison takes over that network, the better.
English
Both models do this. GPT-5.4 skipped a step in my day recap process yesterday. When I asked why, it told me I needed rest.
Jared Sleeper@JaredSleeper
One of Claude’s best features is that it will often tell you to “go to sleep.” Can’t imagine chatGPT doing that- it acts like it is desperate for every last second of engagement. It’s so clear which one is healthier/better.
English
@JaredSleeper Both models do this. GPT-5.4 skipped a step in my day recap process yesterday. When I asked why, it told me I needed rest.
English
One of Claude’s best features is that it will often tell you to “go to sleep.”
Can’t imagine chatGPT doing that- it acts like it is desperate for every last second of engagement.
It’s so clear which one is healthier/better.
English
