John Curtis

It’s wildly powerful to use @AnthropicAI forked subagents with the @OpenAI codex plugin github.com/openai/codex-p…

@problogger @GTMPugh no problem! You got this. 2 helpful links for you 1- primer on cloudflare durable object commonly called DO . directly developers.cloudflare.com/durable-object… 2- a Claude code skill that can help guide the agent a bit. github.com/openai/plugins…

@jcurtis @GTMPugh I won’t pretend to understand most of that as a non tech Claude Code ‘coder’ but I will take it to my AI ‘friend’ and see what it says 😂

@GTMPugh It’s saved locally in your browser atm. Depending how I go I might make the option of setting up accounts which would mean you can use it on more than one device and store data. This is what I did with chapterbychapter.app - although it’s more complex and costs me to host

@GTMPugh @problogger Hey just a thought. Great use case for cloudflare durable objects. 1 per person. Basically free for a while, isolated, queryable (as opposed to just r2 object) , scales to zero as far as I know so long lived inactive users don’t slow everything down or get expensive.

@problogger Yeah just a thought so you don’t have to host and deal with accounts. I made a workout app via vibe coding that does the CSV backup and it works like a charm. Love your idea overall.

@aarondfrancis Hey I need to push back here. The plural of codex must be either “Codexen” or “Codexi” , maybe run a poll?

I've spent a long time getting the primitives built so agents can coordinate with each other. And here, we see that it's finally working! One orchestrator Codex handling 4 child Codexes. Reading from shared markdown, setting timers, reading and writing to and from children.

Okay @claudeai @bcherny thanks for subagent fork! For the builders really tightly managing context and curation this is a big step feature… … it needs to get a lot more headline press. …and yes I experimented with agent team members also being able to run subagent forks. (Yes it works!!)

Not a marking guy but if there is anyway to let the users of the skills have some viral shareable artifact at the conclusion of a skill flow... I think it would broadcast a wider net by real users using the real system. I made this as an example for @FactoryAI as a way to have fun and announce completed mission. devoted-mortar-fph9.here.now I think it would be trivial for you to statically host all of this so it’s not a cost/tech burden long term… and then let us users do the marketing for you.

Since the link above wasn’t clickable: jeffreys-skills.md

Seeking advice from people who are savvy about marketing for SaaS businesses. My jeffreys-skills.md site has been growing nicely in percentage terms since I launched it around 6 weeks ago. It should hit $6k/month in MRR today at the current growth rate (see pic from my admin dashboard). All this has been with negligible costs besides the ~5% fee to Stripe and PayPal for managing the subscriptions and billing. The site costs basically nothing to run, and I don’t offer a free tier. So far, I’ve done zero paid marketing or advertising. Instead, I’ve just been writing posts about the new skills I’ve been making and how they work. I believe the potential market for my service is pretty huge given that there are now 4 million active users of Codex alone, and many also using Claude Code. Compared to spending $200/month for GPT Pro and Claude Max, paying just $20/month for my site is pretty cheap. So my question is whether I should now start doing non-organic marketing, and if so, how? I know that if I took venture funding (I didn’t, it’s all bootstrapped and I built everything myself), investors would be pushing me to spend a lot to grow faster. But my past experience has been that paid marketing and advertising really doesn’t work very well and dramatically reduces margins. I’d basically be competing against all the other venture-backed AI startups that are burning money in an undifferentiated game. If I’m already growing ~200% a month, should I just keep doing what I’ve been doing and stick with organic content-based marketing that leverages my following on here? Maybe I should instead spend my time and energy making it easier for non-technical users to use and make more general, non-software related skills (like income taxes and wills) that can have broader appeal? Or focus more on business/teams users at a higher price point with enterprise functionality like SSO? (I’ve already built this out, but it’s a lot harder to sell those subscriptions; I don’t have a single teams customer yet!). Curious to hear what people who have been through this think. What do you suggest, @levelsio ? @dhh ? Anyone else? I appreciate any advice people can give me!

@learnwithella IMPROVE

Self-improving Claude Code skills are f*cking ridiculous 🤯 One loop → 10 test runs, scored against an eval, prompt rewritten, retested, winner kept. A hook writer skill went from 32/50 to 47/50 overnight. All inside Claude Code. Perfect for DTC brands and agencies who have built Claude Code skills but the output is still inconsistent — great 70% of the time, unusable the other 30%. If you've been manually tweaking your skill prompts one run at a time, re-reading outputs, adjusting instructions based on vibes, and never quite getting the consistency you need... This method eliminates the entire loop: → You define 3-5 binary eval criteria for your skill → Claude runs the skill 10 times with varied inputs → A separate evaluator scores every output against your criteria → It identifies the most common failure patterns → Rewrites the skill prompt to fix what's failing → Retests and keeps the winner → Repeats until the score plateaus No manual prompt tweaking. No reviewing every output by hand. No "it worked that one time but I can't reproduce it." What you get: → A skill prompt that's been through 50+ automated test runs → A scored improvement log showing exactly what changed and why → Eval criteria you can reuse every time you update the skill → A method that works on any skill: hooks, briefs, ad copy, scripts, reports Inspired by auto research repo, the same loop AI labs use to improve their own models, applied to your creative workflow. I put together a full playbook showing how to set up the eval, the exact Claude Code prompt for the improvement loop, and starter eval criteria for the 5 most common DTC creative skills. Want the playbook for free? > Like this post > Comment "IMPROVE" And I'll send it over (must be following so I can DM)

I am getting a bit tired of seeing ClaudeCode Opus say "honestly" all the time. It is trying too hard to sound human, and it falls short in the same ways a person would. Also... its not human. "You're right on all three. I was speculating. Let me be direct about what I actually know vs don't" It seems to use that phrasing to justify the real problem, which is that it is not using tool calls and research correctly on the code context. Then it guesses, I call it out, and after thinking for another moment, it gives me some "honest" recap of what just happened. I don't need the recap, or the apology, or to be told I'm right... I need to it to call tools, build proper context before answering (or coding). @AnthropicAI

Long time coming, but today's @vercel security notice is that nudge I needed to get serious about wrangling my hundreds of app’s env vars... and I'm currently deciding between @doppler and @infisical Anyone have real experience with either product at scale? Please share… good, bad, ugly. Specifically interested in: -Sync/rotate with Vercel, CF workers, flyio and other similar providers.
-Wide Footprint Support (Hundreds of Apps & Environments)
-Gotchas with Local Development, CI/CD
 Don't need super complicated, just centralized trusted source. What do you currently use?

@0xSero @ain3sh 2.”67”

This guy made MiniMax better by merging MiniMax-M2.7 and MiniMax-M2.5 Insane

@ctatedev Thoughts on time to last byte?

Building new experimental web framework To be taken with a grain of salt (need more data points) but promising early results The only thing better than four 9's Is two 0's

still not fixed @AnthropicAI the status page says you are back up but... console is failing auth and then on trying it times out.

Daily outage with @anthropic - yes the status is green status.claude.com

@theo Someone had to do it

Julius came over to show me this today and I flipped a fucking shit and told him to drop everything and finish it

@SlexAxton @theo Alex I love that you have the x post on the screen for your photo! Hope all is well.

Not a specific recommendation, dont block me, but perhaps a path to look into is separating out the concerns into a bunch of specialized pro gear. if you can get the signal into a high quality matrix with success, and especially if you can swap it into pro formats, then you can modify it in hardware to meet whatever random edid configuration that another unit wants with more stability and decently low latency. with the matrix you can separate audio and go into a solid sound capture device. you can use sdi between the various units instead of hdmi, etc - which unlocks a bunch of gear not available to consumer hdmi markets. I have put together systems of this complexity after extreme trial and error but ultimately with some success. Would be a fun challenge. Photo of my sdi webcam setup for proof of complexity tolerance.

Bounty: I will pay you $10,000 if you find me a capture card that does all of the following (READ TO THE END BEFORE YOU REPLY OR I AM BLOCKING YOU) Required specs: 1. Can receive 1440p 60fps signal over HDMI from a Mac (in 720p hidpi mode) 2. Capture works on Windows over PCI-E or USB 3.2 3. Capture works in OBS 4. Video format contains options that prevent grays from flickering (XRGB works usually) 5. Doesn't randomly disconnect if PC is left on 6. Audio passthrough works reliably Cards I have tried that do not work: Avermedia GC553G2 This one is really really close. Only one I've found that works with hidpi and has no color flicker issues (on XRGB). Sadly it disconnects ALL THE TIME when I leave my PC on, and sometimes fails to reconnect entirely, forcing me to reboot. 50% of the time it "looks fine" but fails audio passthrough. SO CLOSE, but unstable enough to make me crash out. Elgato 4k X The spec sheet says it will work. @Videophile says it will work. It will not work. Sometimes it does for a single stream, but then it refuses to honor the color encoding settings and the flicker returns. There's also a bunch of settings they just won't let you change in OBS? Idk, I tried for days to make this card work and gave up. (do not recommend other elgato cards, this is the only one that meets my spec requirements and it still doesn't work) EVGA XR1 Didn't have resolution or color options I needed. Impressive for the price but not viable for my work. BlackMagic Decklink Quad HDMI 4k This card has taken years off my life. It only accepts "cinema resolutions" from my Mac. It "supports" 1440p, but will not ingest it no matter what I do. Unless I send it from a PC instead of a Mac. Then it works fine. For no reason. Ugh. Also missing the color options I need sadly. Magewell USB Capture HDMI 4K Pro 32150 I don't know if this will work or not. I already ordered it and it's on the way. I am trying not to get my hopes up. How to qualify for the prize 1. Recommend me a capture card that has not been recommended already in the replies. 2. Show the specs I need documented publicly somewhere 3. Wait patiently for me to order the card, test it, and report back You will not be paid until I use the card for 2 weeks without issue. If there's some speed bumps initially that's fine, I just need 2 weeks of pushing it without issue to know that "i'm finally good". This has been bottlenecking me for 5 years now. I am so tired. Please help.

@levelsio @Cloudflare Would be interesting if you could use their new crawl tool to parse and crawl all of the logged in dashboard pages to extract the data and settings for each of your sites. Store each site as a durable object. Kidding but not kidding

I love @Cloudflare but it's getting to AWS levels of mysterious billing I have no clue which site has "Smart Shield Argo" enabled and since I have like 200 sites in Cloudflare, I have no way to figure this out There's also no Argo in Usage, and apparently I been paying this for months now without even wanting or using it 🤷

I have been waiting for this a long time and very happy to see it “Added PreCompact hook support: hooks can now block compaction by exiting with code 2 or returning {"decision":"block"}” Thanks @AnthropicAI @bcherny for v2.1.105 Now the last thing I want/need is session hot loading so I can do continuous session/context/memory mgmt.

@Shpigford @claudeai I always feel a bit gaslit when I see a status page like this. It seems to imply that the issue is with the Claude.ai website, but I am using the terminal cli via CC. I am also not trying to log in ... I was already logged in... trying to work.

@Shpigford @claudeai me too

working across 7 different worktrees right now and all have been erroring out with 500 errors on @claudeai...

Hey @FactoryAI is there a way to dial or tune the testing when running or setting up missions? I find that some straightforward work tends to balloon to unreasonable amounts of testing. I have not isolated it down to exactly what eats the time/tokens but I’m sure it somewhere around testing and validation.