Dark | Light
# ![@andyrewlee Avatar](https://lunarcrush.com/gi/w:26/cr:twitter::31178584.png) @andyrewlee andyrewlee

andyrewlee posts on X about codex, open ai, claude code, code the most. They currently have [---] followers and [---] posts still getting attention that total [------] engagements in the last [--] hours.

### Engagements: [------] [#](/creator/twitter::31178584/interactions)
![Engagements Line Chart](https://lunarcrush.com/gi/w:600/cr:twitter::31178584/c:line/m:interactions.svg)

- [--] Week [---------] +49%
- [--] Month [---------] +20,387%
- [--] Months [---------] +272,842%

### Mentions: [--] [#](/creator/twitter::31178584/posts_active)
![Mentions Line Chart](https://lunarcrush.com/gi/w:600/cr:twitter::31178584/c:line/m:posts_active.svg)

- [--] Week [--] +112%
- [--] Month [--] +242%
- [--] Months [---] +8,150%

### Followers: [---] [#](/creator/twitter::31178584/followers)
![Followers Line Chart](https://lunarcrush.com/gi/w:600/cr:twitter::31178584/c:line/m:followers.svg)

- [--] Week [---] +7.70%
- [--] Month [---] +20%
- [--] Months [---] +199%

### CreatorRank: [---------] [#](/creator/twitter::31178584/influencer_rank)
![CreatorRank Line Chart](https://lunarcrush.com/gi/w:600/cr:twitter::31178584/c:line/m:influencer_rank.svg)

### Social Influence

**Social category influence**
[technology brands](/list/technology-brands)  [stocks](/list/stocks)  [nba](/list/nba)  [social networks](/list/social-networks)  [finance](/list/finance)  [cryptocurrencies](/list/cryptocurrencies)  [celebrities](/list/celebrities)  [countries](/list/countries)  [gaming](/list/gaming)  [vc firms](/list/vc-firms) 

**Social topic influence**
[codex](/topic/codex) #160, [open ai](/topic/open-ai), [claude code](/topic/claude-code), [code](/topic/code), [build](/topic/build), [this is](/topic/this-is), [openclaw](/topic/openclaw), [ai](/topic/ai), [agent](/topic/agent), [model](/topic/model)

**Top accounts mentioned or mentioned by**
[@steipete](/creator/undefined) [@convex](/creator/undefined) [@openclaw](/creator/undefined) [@openai](/creator/undefined) [@markk](/creator/undefined) [@jackfriks](/creator/undefined) [@conductorbuild](/creator/undefined) [@aisdk](/creator/undefined) [@steveyegge](/creator/undefined) [@factoryai](/creator/undefined) [@customclaw](/creator/undefined) [@0xlalice](/creator/undefined) [@winigoat7](/creator/undefined) [@joshbickett](/creator/undefined) [@llmwizard](/creator/undefined) [@catchphotons](/creator/undefined) [@ctodotnew](/creator/undefined) [@graphite](/creator/undefined) [@jshchnz](/creator/undefined) [@rayfernando1337](/creator/undefined)

**Top assets mentioned**
[Alphabet Inc Class A (GOOGL)](/topic/$googl) [Cogito Finance (CGV)](/topic/cogito) [DoorDash Inc. (DASH)](/topic/doordash)
### Top Social Posts
Top posts by engagements in the last [--] hours

"should consider changing its name. since the name is the url posts will always be nerfed here. that being said its vibe coding experience far exceeds devin v0 cursor and claude code. http://phoenix.new http://phoenix.new"  
[X Link](https://x.com/andyrewlee/status/1941597486604517563)  2025-07-05T20:38Z [--] followers, [----] engagements


"@nizzyabi @vercel vercel is eating javascript"  
[X Link](https://x.com/andyrewlee/status/1942591052831154325)  2025-07-08T14:26Z [--] followers, [---] engagements


"@theo still waiting for cogito"  
[X Link](https://x.com/andyrewlee/status/1960910459525324896)  2025-08-28T03:41Z [---] followers, [---] engagements


"4Cs of vibe coding - convex - codex - clerk - cursor"  
[X Link](https://x.com/andyrewlee/status/1961713557218992453)  2025-08-30T08:52Z [---] followers, [--] engagements


"@TheAhmadOsman next up: ollama gets faster ✅"  
[X Link](https://x.com/andyrewlee/status/1963788981860188314)  2025-09-05T02:19Z [---] followers, [---] engagements


"@HoopMixOnly lakers turning into weight loss camp"  
[X Link](https://x.com/andyrewlee/status/1965840638139900223)  2025-09-10T18:11Z [---] followers, [----] engagements


"@dyfrai @AlexReibman @talents_kids nano banana is pretty new will check them out"  
[X Link](https://x.com/andyrewlee/status/1966640047089999976)  2025-09-12T23:08Z [---] followers, [--] engagements


"@noticeUsual @winigoat7 luka getting better. nominated for best defensive player during eurobasket"  
[X Link](https://x.com/andyrewlee/status/1967485195906650193)  2025-09-15T07:06Z [---] followers, [---] engagements


"i miss game of zones @BleacherReport. nba offseason had so much material"  
[X Link](https://x.com/andyrewlee/status/1970242238270914690)  2025-09-22T21:42Z [---] followers, [--] engagements


"want to try alpha releases of codex on a mac [--]. [--]. download codex-aarch64-apple-darwin.tar.gz [--]. unpack for codex-aarch64-apple-darwin executable [--]. xattr -d ./codex-aarch64-apple-darwin [--]. ./codex-aarch64-apple-darwin http://com.apple https://github.com/openai/codex/releases http://com.apple https://github.com/openai/codex/releases"  
[X Link](https://x.com/andyrewlee/status/1970761244736131311)  2025-09-24T08:04Z [---] followers, [--] engagements


"@josh_bickett i don't see the appeal of CLI agents from providers without their own models. makes sense for codex claude code qwen and gemini. they optimize it for their models and dogfood internally. does it make sense for cursor and copilot"  
[X Link](https://x.com/andyrewlee/status/1971332802630324559)  2025-09-25T21:55Z [---] followers, [---] engagements


"context anxiety is real. a new term has entered the lexicon. i've experienced context anxiety with codex i've experienced context anxiety with codex"  
[X Link](https://x.com/andyrewlee/status/1973642979320725786)  2025-10-02T06:55Z [---] followers, [---] engagements


"who will capture the most attention this week OpenAI Anthropic Google OpenAI Anthropic Google"  
[X Link](https://x.com/andyrewlee/status/1975114900981903767)  2025-10-06T08:24Z [---] followers, [--] engagements


"openai has devday anthropic has hats and google hyping up the week https://x.com/osanseviero/status/1973437740210594119 Excited for next week :) https://x.com/osanseviero/status/1973437740210594119 Excited for next week :)"  
[X Link](https://x.com/andyrewlee/status/1975117116622934364)  2025-10-06T08:33Z [---] followers, [--] engagements


"openai really doesn't want you to leave chatgpt third time's the charm ❌ chatgpt plugins ❌ custom gpts ❔ chatgpt apps"  
[X Link](https://x.com/andyrewlee/status/1975250787128517068)  2025-10-06T17:24Z [---] followers, [--] engagements


"familiar faces as launch partners. devs will be able to publish their own end of year"  
[X Link](https://x.com/andyrewlee/status/1975251837940080722)  2025-10-06T17:28Z [---] followers, [--] engagements


"apps sdk examples now available on github https://github.com/openai/openai-apps-sdk-examples https://github.com/openai/openai-apps-sdk-examples"  
[X Link](https://x.com/andyrewlee/status/1975256731380031685)  2025-10-06T17:47Z [---] followers, [--] engagements


"oh snap all the openai language libs updated at the same time"  
[X Link](https://x.com/andyrewlee/status/1975257930107920606)  2025-10-06T17:52Z [---] followers, [--] engagements


"canva devs having to integrate with chatgpt a third time"  
[X Link](https://x.com/andyrewlee/status/1975263264838984073)  2025-10-06T18:13Z [---] followers, [--] engagements


"@boneGPT was hoping for more with codex. i could care less about slack sdk and enterprise controls"  
[X Link](https://x.com/andyrewlee/status/1975264957278396463)  2025-10-06T18:20Z [---] followers, [--] engagements


"@gregisenberg prompt engineering context engineering agent engineering"  
[X Link](https://x.com/andyrewlee/status/1975265569139270093)  2025-10-06T18:23Z [---] followers, [---] engagements


"@shauvikkumar isn't it better to launch first i think it's because meta vibes had the usual AI slop. sora [--] provided us with slop we haven't seen before with higher quality cameos and shocking disregard for copyright"  
[X Link](https://x.com/andyrewlee/status/1975443893618651167)  2025-10-07T06:11Z [---] followers, [--] engagements


"@llm_wizard days are limited. next generation of vibe coders are claiming the term "coding". https://x.com/a16z/status/1971259522670088514 The default way to code is vibecoding. OpenAI chief research officer Mark Chen (@markchen90) says high schoolers already see coding from scratch as weird. If coding is vibes what does great research look like https://t.co/xLSKCbtUxJ https://x.com/a16z/status/1971259522670088514 The default way to code is vibecoding. OpenAI chief research officer Mark Chen (@markchen90) says high schoolers already see coding from scratch as weird. If coding is vibes what"  
[X Link](https://x.com/andyrewlee/status/1975597712293929201)  2025-10-07T16:22Z [---] followers, [--] engagements


"@llm_wizard nothing wrong with that. can't blame a bad diff on vibes"  
[X Link](https://x.com/andyrewlee/status/1975598218261242081)  2025-10-07T16:24Z [---] followers, [--] engagements


"@tszzl openai is nothing without its people"  
[X Link](https://x.com/andyrewlee/status/1975993125601673614)  2025-10-08T18:34Z [---] followers, [---] engagements


"keeping up with codex has me learning rust through osmosis Codex is so good and is going to get so amazing. I am having a hard time imagining what creating software at the end of [----] is going to look like. Codex is so good and is going to get so amazing. I am having a hard time imagining what creating software at the end of [----] is going to look like"  
[X Link](https://x.com/andyrewlee/status/1977284512301359407)  2025-10-12T08:05Z [---] followers, [---] engagements


"@josh_bickett you can also select a previous session from a list with: codex resume"  
[X Link](https://x.com/andyrewlee/status/1977957092129640909)  2025-10-14T04:38Z [---] followers, [---] engagements


"plan mode coming soon to codex in openai's latest video on codex cli one of previous sessions via codex resume starts with: "Make a plan to implement a plan mode in plan-mode-planmd for a 'planning' mode that runs via codex plan similar to /review.""  
[X Link](https://x.com/andyrewlee/status/1978219941120692489)  2025-10-14T22:02Z [---] followers, [---] engagements


"@steipete are you maxing out [--] chatgpt pro accounts"  
[X Link](https://x.com/andyrewlee/status/1978342709787074939)  2025-10-15T06:10Z [---] followers, [---] engagements


"@steipete holy crap here i was thinking chatgpt pro limits were generous"  
[X Link](https://x.com/andyrewlee/status/1978355988513120534)  2025-10-15T07:03Z [---] followers, [---] engagements


"@AniC_dev looking forward to ariana"  
[X Link](https://x.com/andyrewlee/status/1979458097681961200)  2025-10-18T08:02Z [---] followers, [--] engagements


"@zeeg AGI delayed by [--] years"  
[X Link](https://x.com/andyrewlee/status/1979709701916115016)  2025-10-19T00:42Z [---] followers, [--] engagements


"@krishna_avi at least friend is explicit in nature. disguised alternatives are more sinister"  
[X Link](https://x.com/andyrewlee/status/1979844735088140799)  2025-10-19T09:39Z [---] followers, [---] engagements


"upgrading my AI workflow requires effort. in the short term i slow down but velocity compounds. here are my recent upgrades: - sectioned checklist to guide [--] hr+ tasks - git worktrees to parallelize ambitious features - delegating small tasks to codex cloud next up: tmux"  
[X Link](https://x.com/andyrewlee/status/1980355236293406961)  2025-10-20T19:27Z [---] followers, [--] engagements


"@fearedlaker @BallsackSports @hoodg3nius @Ar15Mvp @winigoat7 i'm sitting next to woj"  
[X Link](https://x.com/andyrewlee/status/1980478045967556943)  2025-10-21T03:35Z [---] followers, [--] engagements


"@steipete focus dogfooding and model expertise matters. hard to beat OpenAI at getting the most out of gpt-5-codex"  
[X Link](https://x.com/andyrewlee/status/1980495792462004402)  2025-10-21T04:46Z [---] followers, [---] engagements


"first class codex cli integration on chatgpt atlas will unlock vibe testing"  
[X Link](https://x.com/andyrewlee/status/1980702309152944561)  2025-10-21T18:26Z [---] followers, [--] engagements


"@AgentifySH i want codex atlas not chatgpt atlas"  
[X Link](https://x.com/andyrewlee/status/1980944248724980153)  2025-10-22T10:28Z [---] followers, [--] engagements


"codex CLI weekly limit with chatgpt pro sub: - cost: $1000 - total tokens: 5.2B hit the limit and verified with ccusage/codex"  
[X Link](https://x.com/andyrewlee/status/1981599674370023575)  2025-10-24T05:52Z [---] followers, [---] engagements


"@Thetruth8240 shaq needed someone to push him. kobe motivated him even when he was on the heat"  
[X Link](https://x.com/andyrewlee/status/1981914786041667895)  2025-10-25T02:44Z [---] followers, [---] engagements


"@catchphotons @mark_k @OpenAI good point. noticed droid auto compacts proactively"  
[X Link](https://x.com/andyrewlee/status/1981945302098579956)  2025-10-25T04:45Z [---] followers, [--] engagements


"@catchphotons @mark_k @OpenAI did codex auto compact before i ran into many situations where it asked me to /compact with 0% left. what's worse is it sometimes wouldn't even let me compact. will give that a try and see if quality improves"  
[X Link](https://x.com/andyrewlee/status/1981963447110943102)  2025-10-25T05:58Z [---] followers, [--] engagements


"@growing_daniel lets go san francisco"  
[X Link](https://x.com/andyrewlee/status/1982329141161783398)  2025-10-26T06:11Z [---] followers, [---] engagements


"@CompLikeDrew wemby has [--] blocks"  
[X Link](https://x.com/andyrewlee/status/1982353210217296154)  2025-10-26T07:46Z [---] followers, [--] engagements


"i'll take the other side to this argument with a twist. @claude_code and @OpenAI will always have the best coding agent for its own models. independent coding agents can get close but their main disadvantage is the lack of focus. jack of all trades master of none. however i am bullish on independent coding agent orchestrators. for example if @conductor_build supported codex and gemini CLI as well. trying to build a better coding agent than the creators of the models is a losing game. instead orchestrators can focus on the following: - allow engineers to choose the best coding agent(s) for the"  
[X Link](https://x.com/andyrewlee/status/1983121777762202010)  2025-10-28T10:40Z [---] followers, [---] engagements


"@TheDunkCentral @hoopinghollerin white chocolate might be better at a pickup game but john stockton is indubitably a better nba player"  
[X Link](https://x.com/andyrewlee/status/1983256123844198856)  2025-10-28T19:34Z [---] followers, 14.6K engagements


"list of free coding agents that doesn't require waitlist [--]. gemini CLI"  
[X Link](https://x.com/andyrewlee/status/1983794054115811615)  2025-10-30T07:12Z [---] followers, [--] engagements


"2. qwen CLI which is a fork of gemini CLI"  
[X Link](https://x.com/andyrewlee/status/1983794055491547566)  2025-10-30T07:12Z [---] followers, [--] engagements


"@Rome_Beast lakers ain't the knicks"  
[X Link](https://x.com/andyrewlee/status/1983945355537608807)  2025-10-30T17:13Z [---] followers, [---] engagements


"@conductor_build here's the script not sure if user should worry about script at all. i just want to specify the commands. this is cursor's. "setup-worktree": "pnpm i" "cp $ROOT_WORKTREE_PATH/.env .env" "cp -r $ROOT_WORKTREE_PATH/.clerk"  https://gist.github.com/andyrewlee/b5c9ad77e969ed801f2a7ab06f606238 https://gist.github.com/andyrewlee/b5c9ad77e969ed801f2a7ab06f606238"  
[X Link](https://x.com/andyrewlee/status/1984340236336578877)  2025-10-31T19:22Z [---] followers, [--] engagements


"been testing @ctodotnew and it works as advertised. cursor and windsurf are able to launch quality models due to their access to data. cto .new provides access for free and makes money by selling anonymized training data to others. here's my observations:"  
[X Link](https://x.com/andyrewlee/status/1984480013513867746)  2025-11-01T04:37Z [---] followers, [---] engagements


"@LakeShowYo put nico's jersey up in the rafters"  
[X Link](https://x.com/andyrewlee/status/1984481958542655937)  2025-11-01T04:45Z [---] followers, [----] engagements


"@ctodotnew oh wow i just gave it a try and it works thanks for the tip. it's not clear from the UI/UX. i think the chat and task interface can be better unified into a single agent experience"  
[X Link](https://x.com/andyrewlee/status/1984590395360379380)  2025-11-01T11:56Z [---] followers, [--] engagements


"@bwinterrose pomelli seems tasteful https://blog.google/technology/google-labs/pomelli/ https://blog.google/technology/google-labs/pomelli/"  
[X Link](https://x.com/andyrewlee/status/1985316241779933476)  2025-11-03T12:00Z [---] followers, [---] engagements


"nice i can stop wasting tokens testing my UI mock your @aisdk responses with loremllm. create ui demos write tests build your frontend without having to worry about the llm costs npm i @loremllm/transport https://t.co/9VTepX788E mock your @aisdk responses with loremllm. create ui demos write tests build your frontend without having to worry about the llm costs npm i @loremllm/transport https://t.co/9VTepX788E"  
[X Link](https://x.com/andyrewlee/status/1985417162555867262)  2025-11-03T18:41Z [---] followers, [--] engagements


"big PRs are common. @graphite solves this with stacked diffs but it's hard to stack retroactively. it's a mindset shift that takes time to build. a stack agent that breaks a large PR into stacked diffs would be a big unlock"  
[X Link](https://x.com/andyrewlee/status/1985735259015823462)  2025-11-04T15:45Z [---] followers, [---] engagements


"@Polymarket call it"  
[X Link](https://x.com/andyrewlee/status/1985933973110013952)  2025-11-05T04:55Z [---] followers, [----] engagements


"@DiscussingFilm gpt [--] before gta 6"  
[X Link](https://x.com/andyrewlee/status/1986547825250111882)  2025-11-06T21:34Z [---] followers, [----] engagements


"i'm so glad i built on top of nano banana. nano banana [--] outputs look amazing. it's peanut butter jelly time"  
[X Link](https://x.com/andyrewlee/status/1987574265491956122)  2025-11-09T17:33Z [---] followers, [---] engagements


"trying 4o to see what the fuss is about"  
[X Link](https://x.com/andyrewlee/status/1987649032500732399)  2025-11-09T22:30Z [---] followers, [--] engagements


"@a16z AI might write 90% of the code but we all know the last 10% takes 90% of the time"  
[X Link](https://x.com/andyrewlee/status/1988084182388724013)  2025-11-11T03:19Z [---] followers, [---] engagements


"@jshchnz @Porsche depends on the sound it makes"  
[X Link](https://x.com/andyrewlee/status/1988296455409500356)  2025-11-11T17:23Z [---] followers, [---] engagements


"@NBAPR @NBA @NBAonNBC @NBAAllStar @NBCSportsPR top [--] players are all in world"  
[X Link](https://x.com/andyrewlee/status/1988474421531840851)  2025-11-12T05:10Z [---] followers, [--] engagements


"@RayFernando1337 4o lives on"  
[X Link](https://x.com/andyrewlee/status/1988697303113507194)  2025-11-12T19:55Z [---] followers, [---] engagements


"@siddhantparadox i do all my coding on gpt-5-codex high through the codex CLI. for harder brainstorming/exploration i've been relying on gpt-5-pro. i'll have to compare gpt-5.1 thinking with gpt-5-pro to see which is better"  
[X Link](https://x.com/andyrewlee/status/1988698022067503397)  2025-11-12T19:58Z [---] followers, [--] engagements


"@chatgpt21 couldve been gpt 6-7"  
[X Link](https://x.com/andyrewlee/status/1989068083512893931)  2025-11-13T20:29Z [---] followers, [---] engagements


"@Kalshi is this what happens when you go all-in on AI"  
[X Link](https://x.com/andyrewlee/status/1989093655249092778)  2025-11-13T22:10Z [---] followers, [--] engagements


"if y'all are in sf check out doordash's going out tab. big discounts if you upload a receipt from the spot. that receipt is the holy grail of marketing: attribution. bullish on doordash as it expands beyond delivery"  
[X Link](https://x.com/andyrewlee/status/1989405836612178285)  2025-11-14T18:51Z [---] followers, [---] engagements


"@Angaisb_ logan taught googl3 how to build hyp3 on twitt3r 🍌🍌"  
[X Link](https://x.com/andyrewlee/status/1989751828180078952)  2025-11-15T17:46Z [---] followers, [---] engagements


"@EnoReyes @bentossell columns for ccusage/codex are: Input Output Reasoning Cache Read Total Tokens Cost (USD) i added up Total Tokens across the [--] days. ah i see will run some experiments on my end. @ryoppippi ccusage/droid soon"  
[X Link](https://x.com/andyrewlee/status/1989789735448117419)  2025-11-15T20:16Z [---] followers, [--] engagements


"@KobeissiLetter rumor has it buffett decided to invest after a private demo of gemini [--] and nano banana 2"  
[X Link](https://x.com/andyrewlee/status/1989797734250103023)  2025-11-15T20:48Z [---] followers, [---] engagements


"@vikhyatk nano banana is the other onewhats the third"  
[X Link](https://x.com/andyrewlee/status/1989851678317973938)  2025-11-16T00:23Z [---] followers, [----] engagements


"@vettmoney @SpursReporter for now yes but skys the limit for this duo"  
[X Link](https://x.com/andyrewlee/status/1989861435527368837)  2025-11-16T01:01Z [---] followers, [---] engagements


"@RobinhoodApp may i merch you"  
[X Link](https://x.com/andyrewlee/status/1990860281359253759)  2025-11-18T19:10Z [---] followers, [--] engagements


"of every frontier model open shot the fate of the universe on the line the martians have the death beam pointed at earth you better hit it I WANT GPT-5.1-CODEX"  
[X Link](https://x.com/andyrewlee/status/1990920265099059397)  2025-11-18T23:09Z [---] followers, [--] engagements


"codex is down time to touch grass https://github.com/openai/codex/issues/7156 https://github.com/openai/codex/issues/7156"  
[X Link](https://x.com/andyrewlee/status/1992307517201031443)  2025-11-22T19:01Z [---] followers, [---] engagements


"antigravity is anti serious without openai models"  
[X Link](https://x.com/andyrewlee/status/1992354051821052166)  2025-11-22T22:06Z [---] followers, [--] engagements


"3. handoff mode # Handoff guidelines You are producing a concise handoff for the next coding agent. Save it to handoffs/file_name.md. These are default rules for handoffs. They are not the final word: any higher-priority instruction in system/developer/user messages or task-specific guidance overrides them. What the handoff must cover 1) Current objective and status what we were doing and how far we got. 2) Decisions made key choices rationales and any reversals. 3) Code and assets files/paths touched branches feature flags and important diffs or todos. 4) Running context commands run (and"  
[X Link](https://x.com/andyrewlee/status/1992836397900656997)  2025-11-24T06:03Z [---] followers, [--] engagements


"based off of codex's review prompt https://github.com/openai/codex/blob/main/codex-rs/core/review_prompt.md https://github.com/openai/codex/blob/main/codex-rs/core/review_prompt.md"  
[X Link](https://x.com/andyrewlee/status/1992836399918383385)  2025-11-24T06:03Z [---] followers, [--] engagements


"@zephyr_z9 gpt-5.1-codex-max gemini [--] nano banana pro on the other hand"  
[X Link](https://x.com/andyrewlee/status/1992877390117388622)  2025-11-24T08:46Z [---] followers, [---] engagements


"week [--] of building $10k biz last week - integrate nano banana pro - pivot to all types of worksheets - iterate with SAT center on math + verbal - add free manipulatives and demo editor - improve landing this week - improve onboarding - post on etsy youtube instagram week [--] of building $10k biz last week - tailored solution for SAT tutoring center - started posting on X and Teachers Pay Teachers - free worksheets page this week - get feedback on SAT feature and iterate - upgrade to nano banana [--] - start posting on additional channels https://t.co/uZIzcKeT4k week [--] of building $10k biz last"  
[X Link](https://x.com/andyrewlee/status/1992963052761981169)  2025-11-24T14:26Z [---] followers, [---] engagements


"@RealSkipBayless skip to my lakers"  
[X Link](https://x.com/andyrewlee/status/1993623987600125967)  2025-11-26T10:12Z [---] followers, 12.2K engagements


"week [--] of building $10k biz last week - optimize onboarding - start posting on youtube instagram etsy - notes and answer key with nano banana pro this week - keep experimenting with social media - teacher interviews - deeper AI integration week [--] of building $10k biz last week - integrate nano banana pro - pivot to all types of worksheets - iterate with SAT center on math + verbal - add free manipulatives and demo editor - improve landing this week - improve onboarding - post on etsy youtube instagram https://t.co/RmVk45oTcu week [--] of building $10k biz last week - integrate nano banana pro -"  
[X Link](https://x.com/andyrewlee/status/1995558623817400812)  2025-12-01T18:20Z [---] followers, [---] engagements


"how to add frontend-design skill to claude code: /plugin marketplace add anthropics/claude-code /plugin install frontend-design@claude-code-plugins"  
[X Link](https://x.com/andyrewlee/status/1995850640732684660)  2025-12-02T13:40Z [---] followers, [--] engagements


"skill aka markdown can be found here: https://github.com/anthropics/claude-code/blob/main/plugins/frontend-design/skills/frontend-design/SKILL.md https://github.com/anthropics/claude-code/blob/main/plugins/frontend-design/skills/frontend-design/SKILL.md"  
[X Link](https://x.com/andyrewlee/status/1995852342898671912)  2025-12-02T13:47Z [---] followers, [--] engagements


"@RayFernando1337 cursor deserves its flowers but you're underestimating how on device inference will play out. there's a reason why openai considers apple as the biggest long term competitor and not google"  
[X Link](https://x.com/andyrewlee/status/1999389021186015658)  2025-12-12T08:01Z [---] followers, [--] engagements


"codex wrapped - joined: [---] days ago - streak: [---] days - tokens: 39.85B"  
[X Link](https://x.com/andyrewlee/status/2002860711714980086)  2025-12-21T21:56Z [---] followers, [--] engagements


"@TomasReimers @graphite @conductor_build you can create worktree off any branch. just missing some scripts/UI to make branching/syncing easy no"  
[X Link](https://x.com/andyrewlee/status/2004600367313723416)  2025-12-26T17:09Z [---] followers, [--] engagements


"hot takes from @Steve_Yegge - engineers with 12-15 years of exp hate AI the most - if you're still using an IDE you are a bad engineer - claude code ain't it agent orchestration is next - "never rewrite your code" is no longer true - models will be [--] times smarter in [--] years"  
[X Link](https://x.com/andyrewlee/status/2005453647459877343)  2025-12-29T01:39Z [---] followers, [--] engagements


"building perch: UI for your gas town Happy New Year I've just launched my coding agent orchestrator Gas Town for anyone crazy enough to try it. https://t.co/xWJLZzmpZH Happy New Year I've just launched my coding agent orchestrator Gas Town for anyone crazy enough to try it. https://t.co/xWJLZzmpZH"  
[X Link](https://x.com/andyrewlee/status/2007257561477853359)  2026-01-03T01:07Z [---] followers, [---] engagements


"review of gas town - first time having 15+ agents work on same project and merge without babysitting - issue tracker to sling code + merge queue to figure out the conflicts will go mainstream - try it out on a new project. high learning curve and requires shift in thinking"  
[X Link](https://x.com/andyrewlee/status/2008596766049136751)  2026-01-06T17:49Z [---] followers, [---] engagements


"@LyalinDotCom @geminicli 90k congrats. thankful it's open source"  
[X Link](https://x.com/andyrewlee/status/2008944894132539771)  2026-01-07T16:52Z [---] followers, [--] engagements


"tried GLM [---] with claude code opencode and droid. peak experience is with droid"  
[X Link](https://x.com/andyrewlee/status/2009091499305369951)  2026-01-08T02:35Z [---] followers, [---] engagements


"happy claude code and codex usage reset day for those who celebrate"  
[X Link](https://x.com/andyrewlee/status/2009280877704941777)  2026-01-08T15:07Z [---] followers, [--] engagements


"@benvargas @clawdbot what about running claude code in cloud sandboxes"  
[X Link](https://x.com/andyrewlee/status/2009468424158859468)  2026-01-09T03:33Z [---] followers, [---] engagements


"DMing @beyang for amp free is a no brainer. $300 worth of frontier models a month for free on a fundamentally different take on coding agents. i am forever indebted because @ampcode shipped an important feature for me that claude code and codex both failed at. For those running into stricter limits and enforcement on coding agent subscriptions check out Amp Free which offers the most powerful free model inference: https://t.co/qHUrbnDq3o And DM me if youd like to get access to our trial of free ad-supported Opus [---] For those running into stricter limits and enforcement on coding agent"  
[X Link](https://x.com/andyrewlee/status/2009488513587347756)  2026-01-09T04:52Z [---] followers, 35.9K engagements


"@0thernet 🤔 i don't find it simple or boring"  
[X Link](https://x.com/andyrewlee/status/2009501456081322429)  2026-01-09T05:44Z [---] followers, [---] engagements


"getting GLM [---] via has enabled me to experiment freely with gas town and other coding agents. i can save my expensive tokens for important features and let it rip with GLM [---]. http://z.ai Worried about token limits since you're Ralph'ing with Claude Code GLM [---] via https://t.co/BErUakJ9V2 coder is on sale - $288/year for a LOT of tokens You can plug this straight into Claude Code via a handy tool Turbo time Get 10% off with my link: https://t.co/wzCzm9Sm4v http://z.ai Worried about token limits since you're Ralph'ing with Claude Code GLM [---] via https://t.co/BErUakJ9V2 coder is on sale -"  
[X Link](https://x.com/andyrewlee/status/2009763262552826356)  2026-01-09T23:04Z [---] followers, [---] engagements


"@rayzhudev @jadenitripp @opencode @FactoryAI access to all models under one subscription. opencode is open source. droid beats claude code in benchmarks using opus [---] but do your own tests and judge for yourself. benchmarks are imperfect"  
[X Link](https://x.com/andyrewlee/status/2009985688859795487)  2026-01-10T13:48Z [---] followers, [---] engagements


"@Yuchenj_UW bit misleading since cowork is a wrapper on top of claude SDK and a tab inside claude desktop which have both existed for some time"  
[X Link](https://x.com/andyrewlee/status/2010951889307959436)  2026-01-13T05:47Z [---] followers, [---] engagements


"@kmdrfx @douglascamata i use all coding agent CLIs pretty frequently. claude code is the worst but not the only one (i.e. gemini CLI). never saw a flicker on codex CLI. maybe a combination of a simpler feature set + optimizations in rust"  
[X Link](https://x.com/andyrewlee/status/2011247254116651099)  2026-01-14T01:21Z [---] followers, [--] engagements


"@mikehostetler long way to go but has a generous free tier (gemini 3) that recharges daily. make sure you specify "Login with Google" as your authentication method using /auth. it's also open source https://github.com/google-gemini/gemini-cli https://github.com/google-gemini/gemini-cli"  
[X Link](https://x.com/andyrewlee/status/2011255285818749134)  2026-01-14T01:53Z [---] followers, [--] engagements


"@willccbb ralphheads will end up in gas town"  
[X Link](https://x.com/andyrewlee/status/2011529863820427368)  2026-01-14T20:04Z [---] followers, [---] engagements


"@thekitze is this the first killer apple vision pro app"  
[X Link](https://x.com/andyrewlee/status/2011530722667086025)  2026-01-14T20:07Z [---] followers, [--] engagements


"@steipete what do you mean by multiple checkouts don't you need either a worktree or a sandbox if you want to work on multiple checkouts at once"  
[X Link](https://x.com/andyrewlee/status/2011887546519969862)  2026-01-15T19:45Z [---] followers, [---] engagements


"@schrockn it's understandable but impact is undeniable - codex adds first class integration with opencode - opencode black takes off - stifled positive claude code momentum"  
[X Link](https://x.com/andyrewlee/status/2011918029211292158)  2026-01-15T21:46Z [---] followers, [----] engagements


"codex multi-agent (subagents) coming soon - collab tools: - multi-agent prompt: - orchestrator prompt: https://github.com/openai/codex/blob/main/codex-rs/core/templates/agents/orchestrator.md https://github.com/openai/codex/blob/main/codex-rs/core/templates/collab/experimental_prompt.md https://github.com/openai/codex/blob/main/codex-rs/core/src/tools/handlers/collab.rs https://github.com/openai/codex/blob/main/codex-rs/core/templates/agents/orchestrator.md https://github.com/openai/codex/blob/main/codex-rs/core/templates/collab/experimental_prompt.md"  
[X Link](https://x.com/andyrewlee/status/2012172550885999002)  2026-01-16T14:38Z [---] followers, [--] engagements


"@OpenAI chatgpt go to the polls"  
[X Link](https://x.com/andyrewlee/status/2012230659092205824)  2026-01-16T18:29Z [---] followers, [---] engagements


"been loving this new claude code feature. creates a new session with the plan as the initial prompt. context engineering made transparent and easy"  
[X Link](https://x.com/andyrewlee/status/2012319232420446441)  2026-01-17T00:21Z [---] followers, [---] engagements


"@jshchnz claude code is ralph pilled"  
[X Link](https://x.com/andyrewlee/status/2013352386073428208)  2026-01-19T20:46Z [---] followers, [--] engagements


"for about an hour i was disappointed that @GeoffreyHuntley and @Steve_Yegge both pumped meme coins. didn't we crucify hawk tuah girl over this since then i've come down from my high horse and realized they are all crazy.and craziness is a requirement for reimagining the future. squint your eyes and look past the $RALPH heads and $GAS bros and you might see a glimpse of what's next in software engineering. https://twitter.com/i/web/status/2013353657815781851 https://twitter.com/i/web/status/2013353657815781851"  
[X Link](https://x.com/andyrewlee/status/2013353657815781851)  2026-01-19T20:51Z [---] followers, [---] engagements


"sneak peak of codex subagents UI"  
[X Link](https://x.com/anyuser/status/2014163863407837242)  2026-01-22T02:31Z [---] followers, [---] engagements


"contrary to what others say this is a tough problem. seems like codex aborted their TUI2 experiment recently. 80k lines of code gone. "Why stop: making that experience feel fully native across the environment matrix (terminal emulator OS input modality multiplexer font/theme alt-screen behavior) creates a combinatorial explosion of edge cases." https://github.com/openai/codex/pull/9640 Our fix to remove flickering is rolled out again to everyone on Claude Code. Why did we roll it back If Claude is so good why were there bugs to begin with Why was it so complex 🧵"  
[X Link](https://x.com/anyuser/status/2014422174531256422)  2026-01-22T19:37Z [---] followers, [---] engagements


"@brexHQ @CapitalOne what's in your wallet"  
[X Link](https://x.com/andyrewlee/status/2014460699171553506)  2026-01-22T22:10Z [---] followers, [----] engagements


"claude code tastefully integrated two emerging orchestration tools: ralph and beads. easy to use and understand for everyone. [--]. ralph: clear context and send plan as initial prompt [--]. beads: todos tasks"  
[X Link](https://x.com/anyuser/status/2014649798901821578)  2026-01-23T10:41Z [---] followers, [----] engagements


"@jamwt @convex - skills - kebab-case - collaboration"  
[X Link](https://x.com/andyrewlee/status/2016346849805009042)  2026-01-28T03:05Z [---] followers, [---] engagements


"@jamwt @convex @waynesutton oo will try it out. i have a canvas app similar to figma. getting cursors and collaboration working on convex requires a lot of hacks + not great experience"  
[X Link](https://x.com/andyrewlee/status/2016348343988097525)  2026-01-28T03:11Z [---] followers, [--] engagements


"no components fit. i might be asking too much from convex. to get the figma experience i have to turn off convex realtime and bring my own websocket server. for example doing cursor this way feels hacky and the experience isn't great. bunch of db writes for a ephemeral cursor effect. https://www.youtube.com/watchv=Tx_YIN-87_M https://www.youtube.com/watchv=Tx_YIN-87_M https://www.youtube.com/watchv=Tx_YIN-87_M https://www.youtube.com/watchv=Tx_YIN-87_M"  
[X Link](https://x.com/andyrewlee/status/2016370076631629902)  2026-01-28T04:37Z [---] followers, [--] engagements


"@chris_mccord skills are worth exploring for all library maintainers. i use vercel's skills for react ai-sdk and react native every day. skills docs"  
[X Link](https://x.com/andyrewlee/status/2016561717191557535)  2026-01-28T17:19Z [---] followers, [----] engagements


"@windsurf i'm scared"  
[X Link](https://x.com/andyrewlee/status/2017370636491100454)  2026-01-30T22:53Z [---] followers, [--] engagements


"@Altimor no code was vibe code all along"  
[X Link](https://x.com/andyrewlee/status/2017767491171471590)  2026-02-01T01:10Z [---] followers, [---] engagements


"@GergelyOrosz @onchainbeliever @Steve_Yegge understandable if you dont want to cover it. maybe youre not fully aware of what happened. it is a topic many are curious about even for those who didnt buy the memecoin"  
[X Link](https://x.com/andyrewlee/status/2017908148615655769)  2026-02-01T10:29Z [---] followers, [--] engagements


"there's less value comparing two models on the same coding agent vs. comparing coding agents themselves (i.e. codex vs. claude code). models and coding agents are so intertwined that instead of choosing the best model choose the best coding agent for the job. Introducing Arena Mode in Windsurf: One prompt. Two models. Your vote. Benchmarks don't reflect real-world coding quality. The best model for you depends on your codebase and stack. So we made real-world coding the benchmark. Free for the next week. May the best model win. https://t.co/qXgd2K4Yf6 Introducing Arena Mode in Windsurf: One"  
[X Link](https://x.com/anyuser/status/2017360439223730286)  2026-01-30T22:13Z [---] followers, [----] engagements


"claude code should stick to being a coding agent instead of being an orchestrator. agent teams should be spun off into another project called claude town. Claude Code now supports agent teams (in research preview) Instead of a single agent working through a task sequentially a lead agent can delegate to multiple teammates that work in parallel to research debug and build while coordinating with each other. Try it out today by https://t.co/vi7lUJDOTi Claude Code now supports agent teams (in research preview) Instead of a single agent working through a task sequentially a lead agent can"  
[X Link](https://x.com/anyuser/status/2019624417417130203)  2026-02-06T04:09Z [---] followers, [---] engagements


"@bentossell i used 5B tokens in [--] week with $200 chatgpt pro subscription. am i comparing it correctly (Total Tokens in ccusage) they seem to be tightening up the usage lately so i'll have to measure again but seems like it's still way cheaper for GPT tokens. https://x.com/andyrewlee/status/1981599674370023575s=20 codex CLI weekly limit with chatgpt pro sub: - cost: $1000 - total tokens: 5.2B hit the limit and verified with ccusage/codex https://t.co/8cAWGsqfSY https://x.com/andyrewlee/status/1981599674370023575s=20 codex CLI weekly limit with chatgpt pro sub: - cost: $1000 - total tokens:"  
[X Link](https://x.com/andyrewlee/status/1989749844186206439)  2025-11-15T17:38Z [---] followers, [---] engagements


"@waynesutton @openclaw @convex @steipete convex is the best. just applied 🦞"  
[X Link](https://x.com/andyrewlee/status/2019603280817713458)  2026-02-06T02:45Z [---] followers, [---] engagements


"@ImSh4yy not sure why OP doesnt mention the main benefit which is your data is never sent to the cloud"  
[X Link](https://x.com/andyrewlee/status/2020595370397720758)  2026-02-08T20:27Z [---] followers, [----] engagements


"@alexpark @customclaw @convex @openclaw thanks. lobster movement is just getting started"  
[X Link](https://x.com/andyrewlee/status/2021714530074820670)  2026-02-11T22:34Z [---] followers, [--] engagements


"@Jason are your top team members using local models for privacy and security reasons local models are good but there's still a noticeable difference between the frontier models (for now). wondering how you decided on this tradeoff"  
[X Link](https://x.com/andyrewlee/status/2022737536796496035)  2026-02-14T18:19Z [---] followers, [----] engagements


"@jackfriks it has to be chatgpt models via subscription [--]. API prices are 10x more expensive than tokens subsidized via subscription [--]. openai is the only frontier lab that supports using its subscription on other harnesses [--]. frontier models are best prepared for prompt injection"  
[X Link](https://x.com/andyrewlee/status/2022755731846631703)  2026-02-14T19:32Z [---] followers, [----] engagements


"@0xLalice @jackfriks positive https://x.com/thsottiaux/status/2010064438033104966 Codex ❤ OSS. Over the coming days we are prioritizing working with open source coding agents and tools to support them in the same way as OpenCode so that codex users can benefit from their account and usage in those combined with using our models in codex directly. We are https://x.com/thsottiaux/status/2010064438033104966 Codex ❤ OSS. Over the coming days we are prioritizing working with open source coding agents and tools to support them in the same way as OpenCode so that codex users can benefit from their"  
[X Link](https://x.com/andyrewlee/status/2022761528420896895)  2026-02-14T19:55Z [---] followers, [---] engagements


"@0xLalice @jackfriks i'm using gpt-5.3-codex"  
[X Link](https://x.com/andyrewlee/status/2022775325529108874)  2026-02-14T20:49Z [---] followers, [--] engagements


"got accepted to Convex for 🦞Claw with @customclaw was going to use @convex anyway. for now i'm installing @openclaw for those in SF. i'll help you set up and customize your lobster (remote or in-person). lower fee for first handful of installs. Convex for 🦞Claw. If you're building with @openclaw & @convex this is for you We're backing developers and founders building with OpenClaw. Apply here https://t.co/JS4aLpsB1f If @steipete uses Convex you should too. https://t.co/3uM97p0e3e Convex for 🦞Claw. If you're building with @openclaw & @convex this is for you We're backing developers and"  
[X Link](https://x.com/anyuser/status/2021677256444457452)  2026-02-11T20:06Z [---] followers, [---] engagements


"access gpt-5.3-codex-spark on codex cli codex -m gpt-5.3-codex-spark if you want to set reasoning codex -m gpt-5.3-codex-spark --config model_reasoning_effort="high" The first time we demo'd GPT-5.3-Codex-Spark internally people thought it was fake or sped up. As someone commented on the team the infra for this is coming in screaming hot directly from the surface of the sun but it is a cool glimpse into what will become mainstream in the https://t.co/XOnbY8EQ9q The first time we demo'd GPT-5.3-Codex-Spark internally people thought it was fake or sped up. As someone commented on the team the"  
[X Link](https://x.com/anyuser/status/2022032776808280064)  2026-02-12T19:39Z [---] followers, [---] engagements


"separate usage limit for spark"  
[X Link](https://x.com/anyuser/status/2022033000473800957)  2026-02-12T19:40Z [---] followers, [--] engagements


"needed to run the CLI command for gpt-5.3-codex-spark to show up on the model selector in my CLI. just got the update on codex app and now i have access there too"  
[X Link](https://x.com/anyuser/status/2022038662901629259)  2026-02-12T20:02Z [---] followers, [--] engagements


"gpt-5.3-codex-spark is NOT gpt-5.3-codex but faster. difference in quality is obvious. use it for the right task. GPT-5.3-Codex-Spark is now in research preview. You can just build thingsfaster. https://t.co/85LzDOgcQj GPT-5.3-Codex-Spark is now in research preview. You can just build thingsfaster. https://t.co/85LzDOgcQj"  
[X Link](https://x.com/anyuser/status/2022053065248157971)  2026-02-12T20:59Z [---] followers, [----] engagements


"@benvargas i was confused by the output at first thinking it was going to be like opus [---] fast"  
[X Link](https://x.com/andyrewlee/status/2022106575281041495)  2026-02-13T00:32Z [---] followers, [--] engagements


"@juscallmevyom you think so wondering if inference from cerebras will always be degraded"  
[X Link](https://x.com/andyrewlee/status/2022139957679469034)  2026-02-13T02:45Z [---] followers, [--] engagements


"@leviath666 @0xSero bengalfox was a mystery model added to the codex repo in december. i just got this error when i hit the [--] hour limit for gpt-5.3-codex-spark. now we know what bengalfox was referring to"  
[X Link](https://x.com/andyrewlee/status/2022140397401878772)  2026-02-13T02:46Z [---] followers, [--] engagements


"@Hckytwn53 for some tasks yes. but for difficult tasks you still have to bring out the big guns with gpt-5.3-codex. i had initially thought this was like opus [---] fast which is just as good as opus [---] but faster. not the case here"  
[X Link](https://x.com/andyrewlee/status/2022167970961928654)  2026-02-13T04:36Z [---] followers, [--] engagements


"@Austen frontier and codex app are pretty close"  
[X Link](https://x.com/andyrewlee/status/2022205576739270961)  2026-02-13T07:05Z [---] followers, [---] engagements


"@zbruhnke @sarahfim @Scobleizer @openclaw only the composio api key is exposed to openclaw. now getting that leaked is still a concern but a bit better than your actual tokens/api keys"  
[X Link](https://x.com/andyrewlee/status/2022559527661682725)  2026-02-14T06:32Z [---] followers, [---] engagements


"@h_rrish @hermann_ai @JJismusic @jackfriks use anthropic subscription at your own risk. it is against their TOS and many people have been banned. https://x.com/trq212/status/2009689809875591565 Yesterday we tightened our safeguards against spoofing the Claude Code harness after accounts were banned for triggering abuse filters from third-party harnesses using Claude subscriptions. https://x.com/trq212/status/2009689809875591565 Yesterday we tightened our safeguards against spoofing the Claude Code harness after accounts were banned for triggering abuse filters from third-party harnesses using"  
[X Link](https://x.com/andyrewlee/status/2022781938587828570)  2026-02-14T21:16Z [---] followers, [---] engagements


"gpt-5.3-codex-spark is NOT gpt-5.3-codex but faster. difference in quality is obvious. use it for the right task. GPT-5.3-Codex-Spark is now in research preview. You can just build thingsfaster. https://t.co/85LzDOgcQj GPT-5.3-Codex-Spark is now in research preview. You can just build thingsfaster. https://t.co/85LzDOgcQj"  
[X Link](https://x.com/anyuser/status/2022053065248157971)  2026-02-12T20:59Z [---] followers, [----] engagements


"GPT-5.3-Codex-Spark is now in research preview. You can just build thingsfaster"  
[X Link](https://x.com/anyuser/status/2022009582210715925)  2026-02-12T18:07Z 4.6M followers, 1.4M engagements


"access gpt-5.3-codex-spark on codex cli codex -m gpt-5.3-codex-spark if you want to set reasoning codex -m gpt-5.3-codex-spark --config model_reasoning_effort="high" The first time we demo'd GPT-5.3-Codex-Spark internally people thought it was fake or sped up. As someone commented on the team the infra for this is coming in screaming hot directly from the surface of the sun but it is a cool glimpse into what will become mainstream in the https://t.co/XOnbY8EQ9q The first time we demo'd GPT-5.3-Codex-Spark internally people thought it was fake or sped up. As someone commented on the team the"  
[X Link](https://x.com/anyuser/status/2022032776808280064)  2026-02-12T19:39Z [---] followers, [---] engagements


"The first time we demo'd GPT-5.3-Codex-Spark internally people thought it was fake or sped up. As someone commented on the team the infra for this is coming in screaming hot directly from the surface of the sun but it is a cool glimpse into what will become mainstream in the next months. https://twitter.com/i/web/status/2022023219738554773 https://twitter.com/i/web/status/2022023219738554773"  
[X Link](https://x.com/anyuser/status/2022023219738554773)  2026-02-12T19:01Z 30.6K followers, 92.6K engagements


"separate usage limit for spark"  
[X Link](https://x.com/anyuser/status/2022033000473800957)  2026-02-12T19:40Z [---] followers, [--] engagements


"needed to run the CLI command for gpt-5.3-codex-spark to show up on the model selector in my CLI. just got the update on codex app and now i have access there too"  
[X Link](https://x.com/anyuser/status/2022038662901629259)  2026-02-12T20:02Z [---] followers, [--] engagements


"got accepted to Convex for 🦞Claw with @customclaw was going to use @convex anyway. for now i'm installing @openclaw for those in SF. i'll help you set up and customize your lobster (remote or in-person). lower fee for first handful of installs. Convex for 🦞Claw. If you're building with @openclaw & @convex this is for you We're backing developers and founders building with OpenClaw. Apply here https://t.co/JS4aLpsB1f If @steipete uses Convex you should too. https://t.co/3uM97p0e3e Convex for 🦞Claw. If you're building with @openclaw & @convex this is for you We're backing developers and"  
[X Link](https://x.com/anyuser/status/2021677256444457452)  2026-02-11T20:06Z [---] followers, [---] engagements


"Convex for 🦞Claw. If you're building with @openclaw & @convex this is for you We're backing developers and founders building with OpenClaw. Apply here If @steipete uses Convex you should too. https://www.convex.dev/claw https://www.convex.dev/claw"  
[X Link](https://x.com/anyuser/status/2019595574497144915)  2026-02-06T02:14Z 66K followers, 33.3K engagements


"open source AI spreadsheet finally got around to finishing this up free template to build your own ai spreadsheet with @shadcn ui @sadmann17 dice ui @tan_stack @aisdk. link in thread https://t.co/8rq937iBoK finally got around to finishing this up free template to build your own ai spreadsheet with @shadcn ui @sadmann17 dice ui @tan_stack @aisdk. link in thread https://t.co/8rq937iBoK"  
[X Link](https://x.com/anyuser/status/2021456629099790389)  2026-02-11T05:29Z [---] followers, [---] engagements


"finally got around to finishing this up free template to build your own ai spreadsheet with @shadcn ui @sadmann17 dice ui @tan_stack @aisdk. link in thread spreadsheet + data enrichment built with @shadcn @tan_stack table @aisdk. thinking about publishing it as a custom artifact. anyone interested https://t.co/e8iUG3EzdP spreadsheet + data enrichment built with @shadcn @tan_stack table @aisdk. thinking about publishing it as a custom artifact. anyone interested https://t.co/e8iUG3EzdP"  
[X Link](https://x.com/anyuser/status/2021284096970981708)  2026-02-10T18:04Z [----] followers, [---] engagements


"CLI (not MCP) is table stakes in the age of openclaw Anything you can do in Obsidian you can do from the command line. Obsidian CLI is now available in [----] (early access). https://t.co/B8ed2zrWHe Anything you can do in Obsidian you can do from the command line. Obsidian CLI is now available in [----] (early access). https://t.co/B8ed2zrWHe"  
[X Link](https://x.com/anyuser/status/2021314729134973092)  2026-02-10T20:06Z [---] followers, [---] engagements


"Anything you can do in Obsidian you can do from the command line. Obsidian CLI is now available in [----] (early access)"  
[X Link](https://x.com/anyuser/status/2021241384057930224)  2026-02-10T15:14Z 171.5K followers, 2.4M engagements


"glimpse of how @openclaw is about to shake up businesses"  
[X Link](https://x.com/anyuser/status/2020582912605995430)  2026-02-08T19:38Z [---] followers, [---] engagements


"setting this up is a no brainer. thanks for sharing @FactoryAI. one prompt and [--] hours later my repo is healthier and better set up for success. Our internal lint rules are now open source featuring 23+ custom rules we use to guide droids at @FactoryAI. These rules cover file organization React patterns testing error handling API conventions and more. Every rule is 100% droid-generated and includes detailed https://t.co/wEeoNXd666 Our internal lint rules are now open source featuring 23+ custom rules we use to guide droids at @FactoryAI. These rules cover file organization React patterns"  
[X Link](https://x.com/anyuser/status/2019868711398862929)  2026-02-06T20:20Z [---] followers, [---] engagements


"Our internal lint rules are now open source featuring 23+ custom rules we use to guide droids at @FactoryAI. These rules cover file organization React patterns testing error handling API conventions and more. Every rule is 100% droid-generated and includes detailed Markdown docs. While this isn't meant to be imported as-is we hope it inspires you to build custom linting rules tailored to your own codebase. Take what works ignore what doesnt and tweak as needed. You can leverage the docs as building blocks to suit your specific framework or language. Check it out:"  
[X Link](https://x.com/anyuser/status/2018754085948924136)  2026-02-03T18:30Z [----] followers, 58.7K engagements


"claude code should stick to being a coding agent instead of being an orchestrator. agent teams should be spun off into another project called claude town. Claude Code now supports agent teams (in research preview) Instead of a single agent working through a task sequentially a lead agent can delegate to multiple teammates that work in parallel to research debug and build while coordinating with each other. Try it out today by https://t.co/vi7lUJDOTi Claude Code now supports agent teams (in research preview) Instead of a single agent working through a task sequentially a lead agent can"  
[X Link](https://x.com/anyuser/status/2019624417417130203)  2026-02-06T04:09Z [---] followers, [---] engagements


"Claude Code now supports agent teams (in research preview) Instead of a single agent working through a task sequentially a lead agent can delegate to multiple teammates that work in parallel to research debug and build while coordinating with each other. Try it out today by enabling agent teams in your settings.json https://twitter.com/i/web/status/2019469032844587505 https://twitter.com/i/web/status/2019469032844587505"  
[X Link](https://x.com/anyuser/status/2019469032844587505)  2026-02-05T17:51Z 61.3K followers, 1.4M engagements


"wow so everyone who was using gpt-5.2-codex outside of codex had a much slower experience this whole time no wonder gpt-5.2-codex is underrated. 40% faster in cursor feels really nice 40% faster in cursor feels really nice"  
[X Link](https://x.com/anyuser/status/2019140431913841149)  2026-02-04T20:06Z [---] followers, [---] engagements


"40% faster in cursor feels really nice GPT-5.2 and GPT-5.2-Codex are now 40% faster. We have optimized our inference stack for all API customers. Same model. Same weights. Lower latency. GPT-5.2 and GPT-5.2-Codex are now 40% faster. We have optimized our inference stack for all API customers. Same model. Same weights. Lower latency"  
[X Link](https://x.com/anyuser/status/2019106751501725838)  2026-02-04T17:52Z 52.3K followers, 28.6K engagements


"i applaud codex app for imagining something new but workspaces is a better abstraction than threads for an agent orchestrator. parallelizing many agents gets way too chaotic with threads. i want to work on multiple workspaces with multiple agents in each one"  
[X Link](https://x.com/anyuser/status/2018572709886878119)  2026-02-03T06:30Z [---] followers, [---] engagements


"codex usage reset and vague posting. it can only mean one thing. @mark_k @OpenAI 👀 @mark_k @OpenAI 👀"  
[X Link](https://x.com/anyuser/status/2018387579926896859)  2026-02-02T18:14Z [---] followers, [---] engagements


"@mark_k @OpenAI 👀"  
[X Link](https://x.com/anyuser/status/2018369032576270657)  2026-02-02T17:00Z 37K followers, [----] engagements


"GUI lovers rejoice codex app is here https://x.com/thsottiaux/status/2018385183171592464s=20 I am Tibo and I have an incredible team. Codex would not exist without them and they cooked. Enjoy the new Codex app access through your free/go ChatGPT plan and 2X rate limits on other plans. Can't wait to hear what you do with it. https://t.co/Lwg13vEJDn https://t.co/c7AaRCenoQ https://x.com/thsottiaux/status/2018385183171592464s=20 I am Tibo and I have an incredible team. Codex would not exist without them and they cooked. Enjoy the new Codex app access through your free/go ChatGPT plan and 2X rate"  
[X Link](https://x.com/anyuser/status/2018388225728065954)  2026-02-02T18:17Z [---] followers, [---] engagements


"I am Tibo and I have an incredible team. Codex would not exist without them and they cooked. Enjoy the new Codex app access through your free/go ChatGPT plan and 2X rate limits on other plans. Can't wait to hear what you do with it. https://openai.com/codex/ https://openai.com/codex/"  
[X Link](https://x.com/anyuser/status/2018385183171592464)  2026-02-02T18:05Z 30.6K followers, 1.1M engagements


"there's less value comparing two models on the same coding agent vs. comparing coding agents themselves (i.e. codex vs. claude code). models and coding agents are so intertwined that instead of choosing the best model choose the best coding agent for the job. Introducing Arena Mode in Windsurf: One prompt. Two models. Your vote. Benchmarks don't reflect real-world coding quality. The best model for you depends on your codebase and stack. So we made real-world coding the benchmark. Free for the next week. May the best model win. https://t.co/qXgd2K4Yf6 Introducing Arena Mode in Windsurf: One"  
[X Link](https://x.com/anyuser/status/2017360439223730286)  2026-01-30T22:13Z [---] followers, [----] engagements


"Introducing Arena Mode in Windsurf: One prompt. Two models. Your vote. Benchmarks don't reflect real-world coding quality. The best model for you depends on your codebase and stack. So we made real-world coding the benchmark. Free for the next week. May the best model win"  
[X Link](https://x.com/anyuser/status/2017334552075890903)  2026-01-30T20:30Z 71.5K followers, 1M engagements


"claude code tastefully integrated two emerging orchestration tools: ralph and beads. easy to use and understand for everyone. [--]. ralph: clear context and send plan as initial prompt [--]. beads: todos tasks"  
[X Link](https://x.com/anyuser/status/2014649798901821578)  2026-01-23T10:41Z [---] followers, [----] engagements


"contrary to what others say this is a tough problem. seems like codex aborted their TUI2 experiment recently. 80k lines of code gone. "Why stop: making that experience feel fully native across the environment matrix (terminal emulator OS input modality multiplexer font/theme alt-screen behavior) creates a combinatorial explosion of edge cases." https://github.com/openai/codex/pull/9640 Our fix to remove flickering is rolled out again to everyone on Claude Code. Why did we roll it back If Claude is so good why were there bugs to begin with Why was it so complex 🧵"  
[X Link](https://x.com/anyuser/status/2014422174531256422)  2026-01-22T19:37Z [---] followers, [---] engagements


"Our fix to remove flickering is rolled out again to everyone on Claude Code. Why did we roll it back If Claude is so good why were there bugs to begin with Why was it so complex 🧵 Weve rewritten Claude Codes terminal rendering system to reduce flickering by roughly 85%. We wanted to share more about why this was so difficult how the fix works and how we used Claude Code to fix it 🧵 Weve rewritten Claude Codes terminal rendering system to reduce flickering by roughly 85%. We wanted to share more about why this was so difficult how the fix works and how we used Claude Code to fix it 🧵"  
[X Link](https://x.com/anyuser/status/2014051499798831291)  2026-01-21T19:04Z 70.3K followers, 477.2K engagements


"ive seen enough. @DevinAI is going to win code review"  
[X Link](https://x.com/anyuser/status/2014391632876163155)  2026-01-22T17:36Z [---] followers, [---] engagements


"sneak peak of codex subagents UI"  
[X Link](https://x.com/anyuser/status/2014163863407837242)  2026-01-22T02:31Z [---] followers, [---] engagements


"https://github.com/openai/codex/pull/9551 https://github.com/openai/codex/pull/9551"  
[X Link](https://x.com/anyuser/status/2014163961239966085)  2026-01-22T02:31Z [---] followers, [--] engagements

Limited data mode. Full metrics available with subscription: lunarcrush.com/pricing

@andyrewlee Avatar @andyrewlee andyrewlee

andyrewlee posts on X about codex, open ai, claude code, code the most. They currently have [---] followers and [---] posts still getting attention that total [------] engagements in the last [--] hours.

Engagements: [------] #

Engagements Line Chart

  • [--] Week [---------] +49%
  • [--] Month [---------] +20,387%
  • [--] Months [---------] +272,842%

Mentions: [--] #

Mentions Line Chart

  • [--] Week [--] +112%
  • [--] Month [--] +242%
  • [--] Months [---] +8,150%

Followers: [---] #

Followers Line Chart

  • [--] Week [---] +7.70%
  • [--] Month [---] +20%
  • [--] Months [---] +199%

CreatorRank: [---------] #

CreatorRank Line Chart

Social Influence

Social category influence technology brands stocks nba social networks finance cryptocurrencies celebrities countries gaming vc firms

Social topic influence codex #160, open ai, claude code, code, build, this is, openclaw, ai, agent, model

Top accounts mentioned or mentioned by @steipete @convex @openclaw @openai @markk @jackfriks @conductorbuild @aisdk @steveyegge @factoryai @customclaw @0xlalice @winigoat7 @joshbickett @llmwizard @catchphotons @ctodotnew @graphite @jshchnz @rayfernando1337

Top assets mentioned Alphabet Inc Class A (GOOGL) Cogito Finance (CGV) DoorDash Inc. (DASH)

Top Social Posts

Top posts by engagements in the last [--] hours

"should consider changing its name. since the name is the url posts will always be nerfed here. that being said its vibe coding experience far exceeds devin v0 cursor and claude code. http://phoenix.new http://phoenix.new"
X Link 2025-07-05T20:38Z [--] followers, [----] engagements

"@nizzyabi @vercel vercel is eating javascript"
X Link 2025-07-08T14:26Z [--] followers, [---] engagements

"@theo still waiting for cogito"
X Link 2025-08-28T03:41Z [---] followers, [---] engagements

"4Cs of vibe coding - convex - codex - clerk - cursor"
X Link 2025-08-30T08:52Z [---] followers, [--] engagements

"@TheAhmadOsman next up: ollama gets faster ✅"
X Link 2025-09-05T02:19Z [---] followers, [---] engagements

"@HoopMixOnly lakers turning into weight loss camp"
X Link 2025-09-10T18:11Z [---] followers, [----] engagements

"@dyfrai @AlexReibman @talents_kids nano banana is pretty new will check them out"
X Link 2025-09-12T23:08Z [---] followers, [--] engagements

"@noticeUsual @winigoat7 luka getting better. nominated for best defensive player during eurobasket"
X Link 2025-09-15T07:06Z [---] followers, [---] engagements

"i miss game of zones @BleacherReport. nba offseason had so much material"
X Link 2025-09-22T21:42Z [---] followers, [--] engagements

"want to try alpha releases of codex on a mac [--]. [--]. download codex-aarch64-apple-darwin.tar.gz [--]. unpack for codex-aarch64-apple-darwin executable [--]. xattr -d ./codex-aarch64-apple-darwin [--]. ./codex-aarch64-apple-darwin http://com.apple https://github.com/openai/codex/releases http://com.apple https://github.com/openai/codex/releases"
X Link 2025-09-24T08:04Z [---] followers, [--] engagements

"@josh_bickett i don't see the appeal of CLI agents from providers without their own models. makes sense for codex claude code qwen and gemini. they optimize it for their models and dogfood internally. does it make sense for cursor and copilot"
X Link 2025-09-25T21:55Z [---] followers, [---] engagements

"context anxiety is real. a new term has entered the lexicon. i've experienced context anxiety with codex i've experienced context anxiety with codex"
X Link 2025-10-02T06:55Z [---] followers, [---] engagements

"who will capture the most attention this week OpenAI Anthropic Google OpenAI Anthropic Google"
X Link 2025-10-06T08:24Z [---] followers, [--] engagements

"openai has devday anthropic has hats and google hyping up the week https://x.com/osanseviero/status/1973437740210594119 Excited for next week :) https://x.com/osanseviero/status/1973437740210594119 Excited for next week :)"
X Link 2025-10-06T08:33Z [---] followers, [--] engagements

"openai really doesn't want you to leave chatgpt third time's the charm ❌ chatgpt plugins ❌ custom gpts ❔ chatgpt apps"
X Link 2025-10-06T17:24Z [---] followers, [--] engagements

"familiar faces as launch partners. devs will be able to publish their own end of year"
X Link 2025-10-06T17:28Z [---] followers, [--] engagements

"apps sdk examples now available on github https://github.com/openai/openai-apps-sdk-examples https://github.com/openai/openai-apps-sdk-examples"
X Link 2025-10-06T17:47Z [---] followers, [--] engagements

"oh snap all the openai language libs updated at the same time"
X Link 2025-10-06T17:52Z [---] followers, [--] engagements

"canva devs having to integrate with chatgpt a third time"
X Link 2025-10-06T18:13Z [---] followers, [--] engagements

"@boneGPT was hoping for more with codex. i could care less about slack sdk and enterprise controls"
X Link 2025-10-06T18:20Z [---] followers, [--] engagements

"@gregisenberg prompt engineering context engineering agent engineering"
X Link 2025-10-06T18:23Z [---] followers, [---] engagements

"@shauvikkumar isn't it better to launch first i think it's because meta vibes had the usual AI slop. sora [--] provided us with slop we haven't seen before with higher quality cameos and shocking disregard for copyright"
X Link 2025-10-07T06:11Z [---] followers, [--] engagements

"@llm_wizard days are limited. next generation of vibe coders are claiming the term "coding". https://x.com/a16z/status/1971259522670088514 The default way to code is vibecoding. OpenAI chief research officer Mark Chen (@markchen90) says high schoolers already see coding from scratch as weird. If coding is vibes what does great research look like https://t.co/xLSKCbtUxJ https://x.com/a16z/status/1971259522670088514 The default way to code is vibecoding. OpenAI chief research officer Mark Chen (@markchen90) says high schoolers already see coding from scratch as weird. If coding is vibes what"
X Link 2025-10-07T16:22Z [---] followers, [--] engagements

"@llm_wizard nothing wrong with that. can't blame a bad diff on vibes"
X Link 2025-10-07T16:24Z [---] followers, [--] engagements

"@tszzl openai is nothing without its people"
X Link 2025-10-08T18:34Z [---] followers, [---] engagements

"keeping up with codex has me learning rust through osmosis Codex is so good and is going to get so amazing. I am having a hard time imagining what creating software at the end of [----] is going to look like. Codex is so good and is going to get so amazing. I am having a hard time imagining what creating software at the end of [----] is going to look like"
X Link 2025-10-12T08:05Z [---] followers, [---] engagements

"@josh_bickett you can also select a previous session from a list with: codex resume"
X Link 2025-10-14T04:38Z [---] followers, [---] engagements

"plan mode coming soon to codex in openai's latest video on codex cli one of previous sessions via codex resume starts with: "Make a plan to implement a plan mode in plan-mode-planmd for a 'planning' mode that runs via codex plan similar to /review.""
X Link 2025-10-14T22:02Z [---] followers, [---] engagements

"@steipete are you maxing out [--] chatgpt pro accounts"
X Link 2025-10-15T06:10Z [---] followers, [---] engagements

"@steipete holy crap here i was thinking chatgpt pro limits were generous"
X Link 2025-10-15T07:03Z [---] followers, [---] engagements

"@AniC_dev looking forward to ariana"
X Link 2025-10-18T08:02Z [---] followers, [--] engagements

"@zeeg AGI delayed by [--] years"
X Link 2025-10-19T00:42Z [---] followers, [--] engagements

"@krishna_avi at least friend is explicit in nature. disguised alternatives are more sinister"
X Link 2025-10-19T09:39Z [---] followers, [---] engagements

"upgrading my AI workflow requires effort. in the short term i slow down but velocity compounds. here are my recent upgrades: - sectioned checklist to guide [--] hr+ tasks - git worktrees to parallelize ambitious features - delegating small tasks to codex cloud next up: tmux"
X Link 2025-10-20T19:27Z [---] followers, [--] engagements

"@fearedlaker @BallsackSports @hoodg3nius @Ar15Mvp @winigoat7 i'm sitting next to woj"
X Link 2025-10-21T03:35Z [---] followers, [--] engagements

"@steipete focus dogfooding and model expertise matters. hard to beat OpenAI at getting the most out of gpt-5-codex"
X Link 2025-10-21T04:46Z [---] followers, [---] engagements

"first class codex cli integration on chatgpt atlas will unlock vibe testing"
X Link 2025-10-21T18:26Z [---] followers, [--] engagements

"@AgentifySH i want codex atlas not chatgpt atlas"
X Link 2025-10-22T10:28Z [---] followers, [--] engagements

"codex CLI weekly limit with chatgpt pro sub: - cost: $1000 - total tokens: 5.2B hit the limit and verified with ccusage/codex"
X Link 2025-10-24T05:52Z [---] followers, [---] engagements

"@Thetruth8240 shaq needed someone to push him. kobe motivated him even when he was on the heat"
X Link 2025-10-25T02:44Z [---] followers, [---] engagements

"@catchphotons @mark_k @OpenAI good point. noticed droid auto compacts proactively"
X Link 2025-10-25T04:45Z [---] followers, [--] engagements

"@catchphotons @mark_k @OpenAI did codex auto compact before i ran into many situations where it asked me to /compact with 0% left. what's worse is it sometimes wouldn't even let me compact. will give that a try and see if quality improves"
X Link 2025-10-25T05:58Z [---] followers, [--] engagements

"@growing_daniel lets go san francisco"
X Link 2025-10-26T06:11Z [---] followers, [---] engagements

"@CompLikeDrew wemby has [--] blocks"
X Link 2025-10-26T07:46Z [---] followers, [--] engagements

"i'll take the other side to this argument with a twist. @claude_code and @OpenAI will always have the best coding agent for its own models. independent coding agents can get close but their main disadvantage is the lack of focus. jack of all trades master of none. however i am bullish on independent coding agent orchestrators. for example if @conductor_build supported codex and gemini CLI as well. trying to build a better coding agent than the creators of the models is a losing game. instead orchestrators can focus on the following: - allow engineers to choose the best coding agent(s) for the"
X Link 2025-10-28T10:40Z [---] followers, [---] engagements

"@TheDunkCentral @hoopinghollerin white chocolate might be better at a pickup game but john stockton is indubitably a better nba player"
X Link 2025-10-28T19:34Z [---] followers, 14.6K engagements

"list of free coding agents that doesn't require waitlist [--]. gemini CLI"
X Link 2025-10-30T07:12Z [---] followers, [--] engagements

"2. qwen CLI which is a fork of gemini CLI"
X Link 2025-10-30T07:12Z [---] followers, [--] engagements

"@Rome_Beast lakers ain't the knicks"
X Link 2025-10-30T17:13Z [---] followers, [---] engagements

"@conductor_build here's the script not sure if user should worry about script at all. i just want to specify the commands. this is cursor's. "setup-worktree": "pnpm i" "cp $ROOT_WORKTREE_PATH/.env .env" "cp -r $ROOT_WORKTREE_PATH/.clerk" https://gist.github.com/andyrewlee/b5c9ad77e969ed801f2a7ab06f606238 https://gist.github.com/andyrewlee/b5c9ad77e969ed801f2a7ab06f606238"
X Link 2025-10-31T19:22Z [---] followers, [--] engagements

"been testing @ctodotnew and it works as advertised. cursor and windsurf are able to launch quality models due to their access to data. cto .new provides access for free and makes money by selling anonymized training data to others. here's my observations:"
X Link 2025-11-01T04:37Z [---] followers, [---] engagements

"@LakeShowYo put nico's jersey up in the rafters"
X Link 2025-11-01T04:45Z [---] followers, [----] engagements

"@ctodotnew oh wow i just gave it a try and it works thanks for the tip. it's not clear from the UI/UX. i think the chat and task interface can be better unified into a single agent experience"
X Link 2025-11-01T11:56Z [---] followers, [--] engagements

"@bwinterrose pomelli seems tasteful https://blog.google/technology/google-labs/pomelli/ https://blog.google/technology/google-labs/pomelli/"
X Link 2025-11-03T12:00Z [---] followers, [---] engagements

"nice i can stop wasting tokens testing my UI mock your @aisdk responses with loremllm. create ui demos write tests build your frontend without having to worry about the llm costs npm i @loremllm/transport https://t.co/9VTepX788E mock your @aisdk responses with loremllm. create ui demos write tests build your frontend without having to worry about the llm costs npm i @loremllm/transport https://t.co/9VTepX788E"
X Link 2025-11-03T18:41Z [---] followers, [--] engagements

"big PRs are common. @graphite solves this with stacked diffs but it's hard to stack retroactively. it's a mindset shift that takes time to build. a stack agent that breaks a large PR into stacked diffs would be a big unlock"
X Link 2025-11-04T15:45Z [---] followers, [---] engagements

"@Polymarket call it"
X Link 2025-11-05T04:55Z [---] followers, [----] engagements

"@DiscussingFilm gpt [--] before gta 6"
X Link 2025-11-06T21:34Z [---] followers, [----] engagements

"i'm so glad i built on top of nano banana. nano banana [--] outputs look amazing. it's peanut butter jelly time"
X Link 2025-11-09T17:33Z [---] followers, [---] engagements

"trying 4o to see what the fuss is about"
X Link 2025-11-09T22:30Z [---] followers, [--] engagements

"@a16z AI might write 90% of the code but we all know the last 10% takes 90% of the time"
X Link 2025-11-11T03:19Z [---] followers, [---] engagements

"@jshchnz @Porsche depends on the sound it makes"
X Link 2025-11-11T17:23Z [---] followers, [---] engagements

"@NBAPR @NBA @NBAonNBC @NBAAllStar @NBCSportsPR top [--] players are all in world"
X Link 2025-11-12T05:10Z [---] followers, [--] engagements

"@RayFernando1337 4o lives on"
X Link 2025-11-12T19:55Z [---] followers, [---] engagements

"@siddhantparadox i do all my coding on gpt-5-codex high through the codex CLI. for harder brainstorming/exploration i've been relying on gpt-5-pro. i'll have to compare gpt-5.1 thinking with gpt-5-pro to see which is better"
X Link 2025-11-12T19:58Z [---] followers, [--] engagements

"@chatgpt21 couldve been gpt 6-7"
X Link 2025-11-13T20:29Z [---] followers, [---] engagements

"@Kalshi is this what happens when you go all-in on AI"
X Link 2025-11-13T22:10Z [---] followers, [--] engagements

"if y'all are in sf check out doordash's going out tab. big discounts if you upload a receipt from the spot. that receipt is the holy grail of marketing: attribution. bullish on doordash as it expands beyond delivery"
X Link 2025-11-14T18:51Z [---] followers, [---] engagements

"@Angaisb_ logan taught googl3 how to build hyp3 on twitt3r 🍌🍌"
X Link 2025-11-15T17:46Z [---] followers, [---] engagements

"@EnoReyes @bentossell columns for ccusage/codex are: Input Output Reasoning Cache Read Total Tokens Cost (USD) i added up Total Tokens across the [--] days. ah i see will run some experiments on my end. @ryoppippi ccusage/droid soon"
X Link 2025-11-15T20:16Z [---] followers, [--] engagements

"@KobeissiLetter rumor has it buffett decided to invest after a private demo of gemini [--] and nano banana 2"
X Link 2025-11-15T20:48Z [---] followers, [---] engagements

"@vikhyatk nano banana is the other onewhats the third"
X Link 2025-11-16T00:23Z [---] followers, [----] engagements

"@vettmoney @SpursReporter for now yes but skys the limit for this duo"
X Link 2025-11-16T01:01Z [---] followers, [---] engagements

"@RobinhoodApp may i merch you"
X Link 2025-11-18T19:10Z [---] followers, [--] engagements

"of every frontier model open shot the fate of the universe on the line the martians have the death beam pointed at earth you better hit it I WANT GPT-5.1-CODEX"
X Link 2025-11-18T23:09Z [---] followers, [--] engagements

"codex is down time to touch grass https://github.com/openai/codex/issues/7156 https://github.com/openai/codex/issues/7156"
X Link 2025-11-22T19:01Z [---] followers, [---] engagements

"antigravity is anti serious without openai models"
X Link 2025-11-22T22:06Z [---] followers, [--] engagements

"3. handoff mode # Handoff guidelines You are producing a concise handoff for the next coding agent. Save it to handoffs/file_name.md. These are default rules for handoffs. They are not the final word: any higher-priority instruction in system/developer/user messages or task-specific guidance overrides them. What the handoff must cover 1) Current objective and status what we were doing and how far we got. 2) Decisions made key choices rationales and any reversals. 3) Code and assets files/paths touched branches feature flags and important diffs or todos. 4) Running context commands run (and"
X Link 2025-11-24T06:03Z [---] followers, [--] engagements

"based off of codex's review prompt https://github.com/openai/codex/blob/main/codex-rs/core/review_prompt.md https://github.com/openai/codex/blob/main/codex-rs/core/review_prompt.md"
X Link 2025-11-24T06:03Z [---] followers, [--] engagements

"@zephyr_z9 gpt-5.1-codex-max gemini [--] nano banana pro on the other hand"
X Link 2025-11-24T08:46Z [---] followers, [---] engagements

"week [--] of building $10k biz last week - integrate nano banana pro - pivot to all types of worksheets - iterate with SAT center on math + verbal - add free manipulatives and demo editor - improve landing this week - improve onboarding - post on etsy youtube instagram week [--] of building $10k biz last week - tailored solution for SAT tutoring center - started posting on X and Teachers Pay Teachers - free worksheets page this week - get feedback on SAT feature and iterate - upgrade to nano banana [--] - start posting on additional channels https://t.co/uZIzcKeT4k week [--] of building $10k biz last"
X Link 2025-11-24T14:26Z [---] followers, [---] engagements

"@RealSkipBayless skip to my lakers"
X Link 2025-11-26T10:12Z [---] followers, 12.2K engagements

"week [--] of building $10k biz last week - optimize onboarding - start posting on youtube instagram etsy - notes and answer key with nano banana pro this week - keep experimenting with social media - teacher interviews - deeper AI integration week [--] of building $10k biz last week - integrate nano banana pro - pivot to all types of worksheets - iterate with SAT center on math + verbal - add free manipulatives and demo editor - improve landing this week - improve onboarding - post on etsy youtube instagram https://t.co/RmVk45oTcu week [--] of building $10k biz last week - integrate nano banana pro -"
X Link 2025-12-01T18:20Z [---] followers, [---] engagements

"how to add frontend-design skill to claude code: /plugin marketplace add anthropics/claude-code /plugin install frontend-design@claude-code-plugins"
X Link 2025-12-02T13:40Z [---] followers, [--] engagements

"skill aka markdown can be found here: https://github.com/anthropics/claude-code/blob/main/plugins/frontend-design/skills/frontend-design/SKILL.md https://github.com/anthropics/claude-code/blob/main/plugins/frontend-design/skills/frontend-design/SKILL.md"
X Link 2025-12-02T13:47Z [---] followers, [--] engagements

"@RayFernando1337 cursor deserves its flowers but you're underestimating how on device inference will play out. there's a reason why openai considers apple as the biggest long term competitor and not google"
X Link 2025-12-12T08:01Z [---] followers, [--] engagements

"codex wrapped - joined: [---] days ago - streak: [---] days - tokens: 39.85B"
X Link 2025-12-21T21:56Z [---] followers, [--] engagements

"@TomasReimers @graphite @conductor_build you can create worktree off any branch. just missing some scripts/UI to make branching/syncing easy no"
X Link 2025-12-26T17:09Z [---] followers, [--] engagements

"hot takes from @Steve_Yegge - engineers with 12-15 years of exp hate AI the most - if you're still using an IDE you are a bad engineer - claude code ain't it agent orchestration is next - "never rewrite your code" is no longer true - models will be [--] times smarter in [--] years"
X Link 2025-12-29T01:39Z [---] followers, [--] engagements

"building perch: UI for your gas town Happy New Year I've just launched my coding agent orchestrator Gas Town for anyone crazy enough to try it. https://t.co/xWJLZzmpZH Happy New Year I've just launched my coding agent orchestrator Gas Town for anyone crazy enough to try it. https://t.co/xWJLZzmpZH"
X Link 2026-01-03T01:07Z [---] followers, [---] engagements

"review of gas town - first time having 15+ agents work on same project and merge without babysitting - issue tracker to sling code + merge queue to figure out the conflicts will go mainstream - try it out on a new project. high learning curve and requires shift in thinking"
X Link 2026-01-06T17:49Z [---] followers, [---] engagements

"@LyalinDotCom @geminicli 90k congrats. thankful it's open source"
X Link 2026-01-07T16:52Z [---] followers, [--] engagements

"tried GLM [---] with claude code opencode and droid. peak experience is with droid"
X Link 2026-01-08T02:35Z [---] followers, [---] engagements

"happy claude code and codex usage reset day for those who celebrate"
X Link 2026-01-08T15:07Z [---] followers, [--] engagements

"@benvargas @clawdbot what about running claude code in cloud sandboxes"
X Link 2026-01-09T03:33Z [---] followers, [---] engagements

"DMing @beyang for amp free is a no brainer. $300 worth of frontier models a month for free on a fundamentally different take on coding agents. i am forever indebted because @ampcode shipped an important feature for me that claude code and codex both failed at. For those running into stricter limits and enforcement on coding agent subscriptions check out Amp Free which offers the most powerful free model inference: https://t.co/qHUrbnDq3o And DM me if youd like to get access to our trial of free ad-supported Opus [---] For those running into stricter limits and enforcement on coding agent"
X Link 2026-01-09T04:52Z [---] followers, 35.9K engagements

"@0thernet 🤔 i don't find it simple or boring"
X Link 2026-01-09T05:44Z [---] followers, [---] engagements

"getting GLM [---] via has enabled me to experiment freely with gas town and other coding agents. i can save my expensive tokens for important features and let it rip with GLM [---]. http://z.ai Worried about token limits since you're Ralph'ing with Claude Code GLM [---] via https://t.co/BErUakJ9V2 coder is on sale - $288/year for a LOT of tokens You can plug this straight into Claude Code via a handy tool Turbo time Get 10% off with my link: https://t.co/wzCzm9Sm4v http://z.ai Worried about token limits since you're Ralph'ing with Claude Code GLM [---] via https://t.co/BErUakJ9V2 coder is on sale -"
X Link 2026-01-09T23:04Z [---] followers, [---] engagements

"@rayzhudev @jadenitripp @opencode @FactoryAI access to all models under one subscription. opencode is open source. droid beats claude code in benchmarks using opus [---] but do your own tests and judge for yourself. benchmarks are imperfect"
X Link 2026-01-10T13:48Z [---] followers, [---] engagements

"@Yuchenj_UW bit misleading since cowork is a wrapper on top of claude SDK and a tab inside claude desktop which have both existed for some time"
X Link 2026-01-13T05:47Z [---] followers, [---] engagements

"@kmdrfx @douglascamata i use all coding agent CLIs pretty frequently. claude code is the worst but not the only one (i.e. gemini CLI). never saw a flicker on codex CLI. maybe a combination of a simpler feature set + optimizations in rust"
X Link 2026-01-14T01:21Z [---] followers, [--] engagements

"@mikehostetler long way to go but has a generous free tier (gemini 3) that recharges daily. make sure you specify "Login with Google" as your authentication method using /auth. it's also open source https://github.com/google-gemini/gemini-cli https://github.com/google-gemini/gemini-cli"
X Link 2026-01-14T01:53Z [---] followers, [--] engagements

"@willccbb ralphheads will end up in gas town"
X Link 2026-01-14T20:04Z [---] followers, [---] engagements

"@thekitze is this the first killer apple vision pro app"
X Link 2026-01-14T20:07Z [---] followers, [--] engagements

"@steipete what do you mean by multiple checkouts don't you need either a worktree or a sandbox if you want to work on multiple checkouts at once"
X Link 2026-01-15T19:45Z [---] followers, [---] engagements

"@schrockn it's understandable but impact is undeniable - codex adds first class integration with opencode - opencode black takes off - stifled positive claude code momentum"
X Link 2026-01-15T21:46Z [---] followers, [----] engagements

"codex multi-agent (subagents) coming soon - collab tools: - multi-agent prompt: - orchestrator prompt: https://github.com/openai/codex/blob/main/codex-rs/core/templates/agents/orchestrator.md https://github.com/openai/codex/blob/main/codex-rs/core/templates/collab/experimental_prompt.md https://github.com/openai/codex/blob/main/codex-rs/core/src/tools/handlers/collab.rs https://github.com/openai/codex/blob/main/codex-rs/core/templates/agents/orchestrator.md https://github.com/openai/codex/blob/main/codex-rs/core/templates/collab/experimental_prompt.md"
X Link 2026-01-16T14:38Z [---] followers, [--] engagements

"@OpenAI chatgpt go to the polls"
X Link 2026-01-16T18:29Z [---] followers, [---] engagements

"been loving this new claude code feature. creates a new session with the plan as the initial prompt. context engineering made transparent and easy"
X Link 2026-01-17T00:21Z [---] followers, [---] engagements

"@jshchnz claude code is ralph pilled"
X Link 2026-01-19T20:46Z [---] followers, [--] engagements

"for about an hour i was disappointed that @GeoffreyHuntley and @Steve_Yegge both pumped meme coins. didn't we crucify hawk tuah girl over this since then i've come down from my high horse and realized they are all crazy.and craziness is a requirement for reimagining the future. squint your eyes and look past the $RALPH heads and $GAS bros and you might see a glimpse of what's next in software engineering. https://twitter.com/i/web/status/2013353657815781851 https://twitter.com/i/web/status/2013353657815781851"
X Link 2026-01-19T20:51Z [---] followers, [---] engagements

"sneak peak of codex subagents UI"
X Link 2026-01-22T02:31Z [---] followers, [---] engagements

"contrary to what others say this is a tough problem. seems like codex aborted their TUI2 experiment recently. 80k lines of code gone. "Why stop: making that experience feel fully native across the environment matrix (terminal emulator OS input modality multiplexer font/theme alt-screen behavior) creates a combinatorial explosion of edge cases." https://github.com/openai/codex/pull/9640 Our fix to remove flickering is rolled out again to everyone on Claude Code. Why did we roll it back If Claude is so good why were there bugs to begin with Why was it so complex 🧵"
X Link 2026-01-22T19:37Z [---] followers, [---] engagements

"@brexHQ @CapitalOne what's in your wallet"
X Link 2026-01-22T22:10Z [---] followers, [----] engagements

"claude code tastefully integrated two emerging orchestration tools: ralph and beads. easy to use and understand for everyone. [--]. ralph: clear context and send plan as initial prompt [--]. beads: todos tasks"
X Link 2026-01-23T10:41Z [---] followers, [----] engagements

"@jamwt @convex - skills - kebab-case - collaboration"
X Link 2026-01-28T03:05Z [---] followers, [---] engagements

"@jamwt @convex @waynesutton oo will try it out. i have a canvas app similar to figma. getting cursors and collaboration working on convex requires a lot of hacks + not great experience"
X Link 2026-01-28T03:11Z [---] followers, [--] engagements

"no components fit. i might be asking too much from convex. to get the figma experience i have to turn off convex realtime and bring my own websocket server. for example doing cursor this way feels hacky and the experience isn't great. bunch of db writes for a ephemeral cursor effect. https://www.youtube.com/watchv=Tx_YIN-87_M https://www.youtube.com/watchv=Tx_YIN-87_M https://www.youtube.com/watchv=Tx_YIN-87_M https://www.youtube.com/watchv=Tx_YIN-87_M"
X Link 2026-01-28T04:37Z [---] followers, [--] engagements

"@chris_mccord skills are worth exploring for all library maintainers. i use vercel's skills for react ai-sdk and react native every day. skills docs"
X Link 2026-01-28T17:19Z [---] followers, [----] engagements

"@windsurf i'm scared"
X Link 2026-01-30T22:53Z [---] followers, [--] engagements

"@Altimor no code was vibe code all along"
X Link 2026-02-01T01:10Z [---] followers, [---] engagements

"@GergelyOrosz @onchainbeliever @Steve_Yegge understandable if you dont want to cover it. maybe youre not fully aware of what happened. it is a topic many are curious about even for those who didnt buy the memecoin"
X Link 2026-02-01T10:29Z [---] followers, [--] engagements

"there's less value comparing two models on the same coding agent vs. comparing coding agents themselves (i.e. codex vs. claude code). models and coding agents are so intertwined that instead of choosing the best model choose the best coding agent for the job. Introducing Arena Mode in Windsurf: One prompt. Two models. Your vote. Benchmarks don't reflect real-world coding quality. The best model for you depends on your codebase and stack. So we made real-world coding the benchmark. Free for the next week. May the best model win. https://t.co/qXgd2K4Yf6 Introducing Arena Mode in Windsurf: One"
X Link 2026-01-30T22:13Z [---] followers, [----] engagements

"claude code should stick to being a coding agent instead of being an orchestrator. agent teams should be spun off into another project called claude town. Claude Code now supports agent teams (in research preview) Instead of a single agent working through a task sequentially a lead agent can delegate to multiple teammates that work in parallel to research debug and build while coordinating with each other. Try it out today by https://t.co/vi7lUJDOTi Claude Code now supports agent teams (in research preview) Instead of a single agent working through a task sequentially a lead agent can"
X Link 2026-02-06T04:09Z [---] followers, [---] engagements

"@bentossell i used 5B tokens in [--] week with $200 chatgpt pro subscription. am i comparing it correctly (Total Tokens in ccusage) they seem to be tightening up the usage lately so i'll have to measure again but seems like it's still way cheaper for GPT tokens. https://x.com/andyrewlee/status/1981599674370023575s=20 codex CLI weekly limit with chatgpt pro sub: - cost: $1000 - total tokens: 5.2B hit the limit and verified with ccusage/codex https://t.co/8cAWGsqfSY https://x.com/andyrewlee/status/1981599674370023575s=20 codex CLI weekly limit with chatgpt pro sub: - cost: $1000 - total tokens:"
X Link 2025-11-15T17:38Z [---] followers, [---] engagements

"@waynesutton @openclaw @convex @steipete convex is the best. just applied 🦞"
X Link 2026-02-06T02:45Z [---] followers, [---] engagements

"@ImSh4yy not sure why OP doesnt mention the main benefit which is your data is never sent to the cloud"
X Link 2026-02-08T20:27Z [---] followers, [----] engagements

"@alexpark @customclaw @convex @openclaw thanks. lobster movement is just getting started"
X Link 2026-02-11T22:34Z [---] followers, [--] engagements

"@Jason are your top team members using local models for privacy and security reasons local models are good but there's still a noticeable difference between the frontier models (for now). wondering how you decided on this tradeoff"
X Link 2026-02-14T18:19Z [---] followers, [----] engagements

"@jackfriks it has to be chatgpt models via subscription [--]. API prices are 10x more expensive than tokens subsidized via subscription [--]. openai is the only frontier lab that supports using its subscription on other harnesses [--]. frontier models are best prepared for prompt injection"
X Link 2026-02-14T19:32Z [---] followers, [----] engagements

"@0xLalice @jackfriks positive https://x.com/thsottiaux/status/2010064438033104966 Codex ❤ OSS. Over the coming days we are prioritizing working with open source coding agents and tools to support them in the same way as OpenCode so that codex users can benefit from their account and usage in those combined with using our models in codex directly. We are https://x.com/thsottiaux/status/2010064438033104966 Codex ❤ OSS. Over the coming days we are prioritizing working with open source coding agents and tools to support them in the same way as OpenCode so that codex users can benefit from their"
X Link 2026-02-14T19:55Z [---] followers, [---] engagements

"@0xLalice @jackfriks i'm using gpt-5.3-codex"
X Link 2026-02-14T20:49Z [---] followers, [--] engagements

"got accepted to Convex for 🦞Claw with @customclaw was going to use @convex anyway. for now i'm installing @openclaw for those in SF. i'll help you set up and customize your lobster (remote or in-person). lower fee for first handful of installs. Convex for 🦞Claw. If you're building with @openclaw & @convex this is for you We're backing developers and founders building with OpenClaw. Apply here https://t.co/JS4aLpsB1f If @steipete uses Convex you should too. https://t.co/3uM97p0e3e Convex for 🦞Claw. If you're building with @openclaw & @convex this is for you We're backing developers and"
X Link 2026-02-11T20:06Z [---] followers, [---] engagements

"access gpt-5.3-codex-spark on codex cli codex -m gpt-5.3-codex-spark if you want to set reasoning codex -m gpt-5.3-codex-spark --config model_reasoning_effort="high" The first time we demo'd GPT-5.3-Codex-Spark internally people thought it was fake or sped up. As someone commented on the team the infra for this is coming in screaming hot directly from the surface of the sun but it is a cool glimpse into what will become mainstream in the https://t.co/XOnbY8EQ9q The first time we demo'd GPT-5.3-Codex-Spark internally people thought it was fake or sped up. As someone commented on the team the"
X Link 2026-02-12T19:39Z [---] followers, [---] engagements

"separate usage limit for spark"
X Link 2026-02-12T19:40Z [---] followers, [--] engagements

"needed to run the CLI command for gpt-5.3-codex-spark to show up on the model selector in my CLI. just got the update on codex app and now i have access there too"
X Link 2026-02-12T20:02Z [---] followers, [--] engagements

"gpt-5.3-codex-spark is NOT gpt-5.3-codex but faster. difference in quality is obvious. use it for the right task. GPT-5.3-Codex-Spark is now in research preview. You can just build thingsfaster. https://t.co/85LzDOgcQj GPT-5.3-Codex-Spark is now in research preview. You can just build thingsfaster. https://t.co/85LzDOgcQj"
X Link 2026-02-12T20:59Z [---] followers, [----] engagements

"@benvargas i was confused by the output at first thinking it was going to be like opus [---] fast"
X Link 2026-02-13T00:32Z [---] followers, [--] engagements

"@juscallmevyom you think so wondering if inference from cerebras will always be degraded"
X Link 2026-02-13T02:45Z [---] followers, [--] engagements

"@leviath666 @0xSero bengalfox was a mystery model added to the codex repo in december. i just got this error when i hit the [--] hour limit for gpt-5.3-codex-spark. now we know what bengalfox was referring to"
X Link 2026-02-13T02:46Z [---] followers, [--] engagements

"@Hckytwn53 for some tasks yes. but for difficult tasks you still have to bring out the big guns with gpt-5.3-codex. i had initially thought this was like opus [---] fast which is just as good as opus [---] but faster. not the case here"
X Link 2026-02-13T04:36Z [---] followers, [--] engagements

"@Austen frontier and codex app are pretty close"
X Link 2026-02-13T07:05Z [---] followers, [---] engagements

"@zbruhnke @sarahfim @Scobleizer @openclaw only the composio api key is exposed to openclaw. now getting that leaked is still a concern but a bit better than your actual tokens/api keys"
X Link 2026-02-14T06:32Z [---] followers, [---] engagements

"@h_rrish @hermann_ai @JJismusic @jackfriks use anthropic subscription at your own risk. it is against their TOS and many people have been banned. https://x.com/trq212/status/2009689809875591565 Yesterday we tightened our safeguards against spoofing the Claude Code harness after accounts were banned for triggering abuse filters from third-party harnesses using Claude subscriptions. https://x.com/trq212/status/2009689809875591565 Yesterday we tightened our safeguards against spoofing the Claude Code harness after accounts were banned for triggering abuse filters from third-party harnesses using"
X Link 2026-02-14T21:16Z [---] followers, [---] engagements

"gpt-5.3-codex-spark is NOT gpt-5.3-codex but faster. difference in quality is obvious. use it for the right task. GPT-5.3-Codex-Spark is now in research preview. You can just build thingsfaster. https://t.co/85LzDOgcQj GPT-5.3-Codex-Spark is now in research preview. You can just build thingsfaster. https://t.co/85LzDOgcQj"
X Link 2026-02-12T20:59Z [---] followers, [----] engagements

"GPT-5.3-Codex-Spark is now in research preview. You can just build thingsfaster"
X Link 2026-02-12T18:07Z 4.6M followers, 1.4M engagements

"access gpt-5.3-codex-spark on codex cli codex -m gpt-5.3-codex-spark if you want to set reasoning codex -m gpt-5.3-codex-spark --config model_reasoning_effort="high" The first time we demo'd GPT-5.3-Codex-Spark internally people thought it was fake or sped up. As someone commented on the team the infra for this is coming in screaming hot directly from the surface of the sun but it is a cool glimpse into what will become mainstream in the https://t.co/XOnbY8EQ9q The first time we demo'd GPT-5.3-Codex-Spark internally people thought it was fake or sped up. As someone commented on the team the"
X Link 2026-02-12T19:39Z [---] followers, [---] engagements

"The first time we demo'd GPT-5.3-Codex-Spark internally people thought it was fake or sped up. As someone commented on the team the infra for this is coming in screaming hot directly from the surface of the sun but it is a cool glimpse into what will become mainstream in the next months. https://twitter.com/i/web/status/2022023219738554773 https://twitter.com/i/web/status/2022023219738554773"
X Link 2026-02-12T19:01Z 30.6K followers, 92.6K engagements

"separate usage limit for spark"
X Link 2026-02-12T19:40Z [---] followers, [--] engagements

"needed to run the CLI command for gpt-5.3-codex-spark to show up on the model selector in my CLI. just got the update on codex app and now i have access there too"
X Link 2026-02-12T20:02Z [---] followers, [--] engagements

"got accepted to Convex for 🦞Claw with @customclaw was going to use @convex anyway. for now i'm installing @openclaw for those in SF. i'll help you set up and customize your lobster (remote or in-person). lower fee for first handful of installs. Convex for 🦞Claw. If you're building with @openclaw & @convex this is for you We're backing developers and founders building with OpenClaw. Apply here https://t.co/JS4aLpsB1f If @steipete uses Convex you should too. https://t.co/3uM97p0e3e Convex for 🦞Claw. If you're building with @openclaw & @convex this is for you We're backing developers and"
X Link 2026-02-11T20:06Z [---] followers, [---] engagements

"Convex for 🦞Claw. If you're building with @openclaw & @convex this is for you We're backing developers and founders building with OpenClaw. Apply here If @steipete uses Convex you should too. https://www.convex.dev/claw https://www.convex.dev/claw"
X Link 2026-02-06T02:14Z 66K followers, 33.3K engagements

"open source AI spreadsheet finally got around to finishing this up free template to build your own ai spreadsheet with @shadcn ui @sadmann17 dice ui @tan_stack @aisdk. link in thread https://t.co/8rq937iBoK finally got around to finishing this up free template to build your own ai spreadsheet with @shadcn ui @sadmann17 dice ui @tan_stack @aisdk. link in thread https://t.co/8rq937iBoK"
X Link 2026-02-11T05:29Z [---] followers, [---] engagements

"finally got around to finishing this up free template to build your own ai spreadsheet with @shadcn ui @sadmann17 dice ui @tan_stack @aisdk. link in thread spreadsheet + data enrichment built with @shadcn @tan_stack table @aisdk. thinking about publishing it as a custom artifact. anyone interested https://t.co/e8iUG3EzdP spreadsheet + data enrichment built with @shadcn @tan_stack table @aisdk. thinking about publishing it as a custom artifact. anyone interested https://t.co/e8iUG3EzdP"
X Link 2026-02-10T18:04Z [----] followers, [---] engagements

"CLI (not MCP) is table stakes in the age of openclaw Anything you can do in Obsidian you can do from the command line. Obsidian CLI is now available in [----] (early access). https://t.co/B8ed2zrWHe Anything you can do in Obsidian you can do from the command line. Obsidian CLI is now available in [----] (early access). https://t.co/B8ed2zrWHe"
X Link 2026-02-10T20:06Z [---] followers, [---] engagements

"Anything you can do in Obsidian you can do from the command line. Obsidian CLI is now available in [----] (early access)"
X Link 2026-02-10T15:14Z 171.5K followers, 2.4M engagements

"glimpse of how @openclaw is about to shake up businesses"
X Link 2026-02-08T19:38Z [---] followers, [---] engagements

"setting this up is a no brainer. thanks for sharing @FactoryAI. one prompt and [--] hours later my repo is healthier and better set up for success. Our internal lint rules are now open source featuring 23+ custom rules we use to guide droids at @FactoryAI. These rules cover file organization React patterns testing error handling API conventions and more. Every rule is 100% droid-generated and includes detailed https://t.co/wEeoNXd666 Our internal lint rules are now open source featuring 23+ custom rules we use to guide droids at @FactoryAI. These rules cover file organization React patterns"
X Link 2026-02-06T20:20Z [---] followers, [---] engagements

"Our internal lint rules are now open source featuring 23+ custom rules we use to guide droids at @FactoryAI. These rules cover file organization React patterns testing error handling API conventions and more. Every rule is 100% droid-generated and includes detailed Markdown docs. While this isn't meant to be imported as-is we hope it inspires you to build custom linting rules tailored to your own codebase. Take what works ignore what doesnt and tweak as needed. You can leverage the docs as building blocks to suit your specific framework or language. Check it out:"
X Link 2026-02-03T18:30Z [----] followers, 58.7K engagements

"claude code should stick to being a coding agent instead of being an orchestrator. agent teams should be spun off into another project called claude town. Claude Code now supports agent teams (in research preview) Instead of a single agent working through a task sequentially a lead agent can delegate to multiple teammates that work in parallel to research debug and build while coordinating with each other. Try it out today by https://t.co/vi7lUJDOTi Claude Code now supports agent teams (in research preview) Instead of a single agent working through a task sequentially a lead agent can"
X Link 2026-02-06T04:09Z [---] followers, [---] engagements

"Claude Code now supports agent teams (in research preview) Instead of a single agent working through a task sequentially a lead agent can delegate to multiple teammates that work in parallel to research debug and build while coordinating with each other. Try it out today by enabling agent teams in your settings.json https://twitter.com/i/web/status/2019469032844587505 https://twitter.com/i/web/status/2019469032844587505"
X Link 2026-02-05T17:51Z 61.3K followers, 1.4M engagements

"wow so everyone who was using gpt-5.2-codex outside of codex had a much slower experience this whole time no wonder gpt-5.2-codex is underrated. 40% faster in cursor feels really nice 40% faster in cursor feels really nice"
X Link 2026-02-04T20:06Z [---] followers, [---] engagements

"40% faster in cursor feels really nice GPT-5.2 and GPT-5.2-Codex are now 40% faster. We have optimized our inference stack for all API customers. Same model. Same weights. Lower latency. GPT-5.2 and GPT-5.2-Codex are now 40% faster. We have optimized our inference stack for all API customers. Same model. Same weights. Lower latency"
X Link 2026-02-04T17:52Z 52.3K followers, 28.6K engagements

"i applaud codex app for imagining something new but workspaces is a better abstraction than threads for an agent orchestrator. parallelizing many agents gets way too chaotic with threads. i want to work on multiple workspaces with multiple agents in each one"
X Link 2026-02-03T06:30Z [---] followers, [---] engagements

"codex usage reset and vague posting. it can only mean one thing. @mark_k @OpenAI 👀 @mark_k @OpenAI 👀"
X Link 2026-02-02T18:14Z [---] followers, [---] engagements

"@mark_k @OpenAI 👀"
X Link 2026-02-02T17:00Z 37K followers, [----] engagements

"GUI lovers rejoice codex app is here https://x.com/thsottiaux/status/2018385183171592464s=20 I am Tibo and I have an incredible team. Codex would not exist without them and they cooked. Enjoy the new Codex app access through your free/go ChatGPT plan and 2X rate limits on other plans. Can't wait to hear what you do with it. https://t.co/Lwg13vEJDn https://t.co/c7AaRCenoQ https://x.com/thsottiaux/status/2018385183171592464s=20 I am Tibo and I have an incredible team. Codex would not exist without them and they cooked. Enjoy the new Codex app access through your free/go ChatGPT plan and 2X rate"
X Link 2026-02-02T18:17Z [---] followers, [---] engagements

"I am Tibo and I have an incredible team. Codex would not exist without them and they cooked. Enjoy the new Codex app access through your free/go ChatGPT plan and 2X rate limits on other plans. Can't wait to hear what you do with it. https://openai.com/codex/ https://openai.com/codex/"
X Link 2026-02-02T18:05Z 30.6K followers, 1.1M engagements

"there's less value comparing two models on the same coding agent vs. comparing coding agents themselves (i.e. codex vs. claude code). models and coding agents are so intertwined that instead of choosing the best model choose the best coding agent for the job. Introducing Arena Mode in Windsurf: One prompt. Two models. Your vote. Benchmarks don't reflect real-world coding quality. The best model for you depends on your codebase and stack. So we made real-world coding the benchmark. Free for the next week. May the best model win. https://t.co/qXgd2K4Yf6 Introducing Arena Mode in Windsurf: One"
X Link 2026-01-30T22:13Z [---] followers, [----] engagements

"Introducing Arena Mode in Windsurf: One prompt. Two models. Your vote. Benchmarks don't reflect real-world coding quality. The best model for you depends on your codebase and stack. So we made real-world coding the benchmark. Free for the next week. May the best model win"
X Link 2026-01-30T20:30Z 71.5K followers, 1M engagements

"claude code tastefully integrated two emerging orchestration tools: ralph and beads. easy to use and understand for everyone. [--]. ralph: clear context and send plan as initial prompt [--]. beads: todos tasks"
X Link 2026-01-23T10:41Z [---] followers, [----] engagements

"contrary to what others say this is a tough problem. seems like codex aborted their TUI2 experiment recently. 80k lines of code gone. "Why stop: making that experience feel fully native across the environment matrix (terminal emulator OS input modality multiplexer font/theme alt-screen behavior) creates a combinatorial explosion of edge cases." https://github.com/openai/codex/pull/9640 Our fix to remove flickering is rolled out again to everyone on Claude Code. Why did we roll it back If Claude is so good why were there bugs to begin with Why was it so complex 🧵"
X Link 2026-01-22T19:37Z [---] followers, [---] engagements

"Our fix to remove flickering is rolled out again to everyone on Claude Code. Why did we roll it back If Claude is so good why were there bugs to begin with Why was it so complex 🧵 Weve rewritten Claude Codes terminal rendering system to reduce flickering by roughly 85%. We wanted to share more about why this was so difficult how the fix works and how we used Claude Code to fix it 🧵 Weve rewritten Claude Codes terminal rendering system to reduce flickering by roughly 85%. We wanted to share more about why this was so difficult how the fix works and how we used Claude Code to fix it 🧵"
X Link 2026-01-21T19:04Z 70.3K followers, 477.2K engagements

"ive seen enough. @DevinAI is going to win code review"
X Link 2026-01-22T17:36Z [---] followers, [---] engagements

"sneak peak of codex subagents UI"
X Link 2026-01-22T02:31Z [---] followers, [---] engagements

"https://github.com/openai/codex/pull/9551 https://github.com/openai/codex/pull/9551"
X Link 2026-01-22T02:31Z [---] followers, [--] engagements

Limited data mode. Full metrics available with subscription: lunarcrush.com/pricing

creator/x::andyrewlee
/creator/x::andyrewlee