@joelhooks I stepped away overnight and found it stuck after 9 hrs. It kept failing because of a codex sandbox restriction it couldn't figure out.
I nudged it: "That's a sandbox thing."
It gave its version of "you're absolutely right," fixed the approach on the next try, and finished in 25
15 months later and I would probably now describe the Effect docs as 'exemplary'.
Really well laid out, with a great intro, lots of progressive disclosure.
They even have an awesome description of what an Effect is.
An Effect is a value that fully describes a program - what it produces, how it errors, and what it requires.
Great job, pals.
@zkochan Before moving and should you use AI written code and contributions you should reach out to the Codeberg team and ask what the following article exactly means. I'm not sure how they determine what is allowed or not
https://t.co/xOM1GtDmse
I'm sure both OpenAI and Anthropic are sitting on some absurd models right now, but at least until those drops everyone else really seems to be catching up.
Grok 4.6 is really good, so was Kimi K3, and this seems like it will get there as well
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
😂😂 Google 😂😂
Some improvements to @t3dotcodes themes shipping to nightly momentarily:
🌟 Built-in themes now use OKLCH color scheme to be more vivid on modern displays
🎨 Themed sidebar art for built-in themes!
🔍 Search and import any theme from VSX Marketplace
@theo Peter Steinberger's codexbar shows way lower usage. Can't talk about Claude usage but for Codex usage it's like half of what is in t3 code for me.
Please help me spread the word about this. People still think Claude Code subs won’t work in apps like T3 Code.
Anthropic’s ban is ONLY harnesses that aren’t Claude Code (see: Pi, OpenCode etc)
mattpocock/skills v1.2 is out!
We're now the 19th most-starred repo of all time. 13.5m downloads on skills.sh. Thanks for your support!
Here's what's new:
- Docs: the community's biggest ask. Every skill documented, with explanations of the main flows + troubleshooting
- Claude Plugin: install via Claude's official marketplace
- Codex Support: full Codex support via agents/openai.yaml files
Updated Skills:
- /grilling now asks you questions in rounds, not one-by-one
- /prototype now uses HTML instead of a TUI for building logic prototypes - easier to share and far richer
- /writing-for-agents renamed from /writing-great-skills, use it for ANYTHING your agents read (AGENTS.md, system prompts, docs)
New Skills:
- /wizard: tired of provisioning infra? Get your agent to build you a TUI to walk you through it
- /to-questionnaire: hit a grilling question you can't answer? Turn the session into a doc you can walk through on a call with a colleague
- /wait-what: no idea what the model said? Refocus it in your domain language and simplify with ASD-STE100
Full changelog + docs below. Video soon!
@faadingmare@theo@emanueledpt@maria_rcks@trySynara Everyone’s entitled to their opinion on Theo's behavior, but switching from a bad legal take to 'everyone is embarrassing' just proves you didn't have a point. The delivery says more about you than them.
@faadingmare@theo@emanueledpt@maria_rcks@trySynara I don't see your point. There are licenses and copyright to almost every codebase. They regulate how people are able and allowed to reuse code. They are legally binding.
@dexhorthy If possible, that would be awesome, but we're not there yet. I find that 5.6 compared to 5.5 does a way better job with the skill. Out of the box as you said it's slightly better.