Mug Club Boutique

♻️ ⬆️❤️

@ziwenxu_ Make it verify checksum manifests on every commit as well

I'm cooked. Codex just nuked my entire repo and everything inside it is gone 2 weeks of work he built for me that I never pushed to GitHub is now dust. Lessons 101: Always tell your AI to push to GitHub for every small win they complete

@ziwenxu_ @justgrm Every actionable checkpoint. Make it verify clean local workspace to local git to github a strict, frequent change management policy. You get the additional bonus of clean versioning on a long goal effort. You want to know where things changed, then, codex can revert and restore

@justgrm Lesson learned... I wasn't even checking to be honest just /goal and see the updates. Now I will just let him to git commit every once awhile

@ziwenxu_ Dang that sucks, yup. Every step. Goal has to tell it to keep that regular cadence and establish gitignore EARLY. otherwise all its gonna use your repo as a sandbox and all that runtime and tests get mingled in with production code.

@Kappaemme1926 Gonna publish one 🤜🤛

Meet builtwithcodex✨ A marketplace for Codex skills. Discover workflows created by other builders, install skills that extend what Codex can do, and publish your own skills directly from Codex. Each skill gets a dedicated listing with a clear description, install link, and project details, so people can understand what it does and start using it faster. If you built a Codex skill, builtwithcodex gives it a place to be found. Built with Codex. Published from Codex. Link in Bio.

@Kappaemme1926 Heck yeah

@ZippyNetworks @sharbel Ohh my gosh. You run a network?? Good luck.

You don't know what you're talking about. Bun is a very established thing. Go read before you act like an ai psychopath know it all without any clue about what you're writing. I generally don't comment on posts from bobs and noobs but this one made me laugh and you really need to research before making accusations about the OP and the list. It's just a simple list of the most active public repos. There is good reason for most of them.

The 10 fastest growing GitHub repos this week: 1. codegraph (+14.1K stars) Pre-indexed code knowledge graph for Claude Code, Codex, Cursor, OpenCode, and Hermes Agent — fewer tokens, fewer tool calls, 100% local github.com/colbymchenry/c… 2. openhuman (+17.1K stars) Your Personal AI super intelligence. Private, Simple and extremely powerful. github.com/tinyhumansai/o… 3. academic-research-skills (+11.6K stars) Academic Research Skills for Claude Code: research → write → review → revise → finalize github.com/Imbad0202/acad… 4. RuView (+6.8K stars) π RuView turns commodity WiFi signals into real-time spatial intelligence, vital sign monitoring, and presence detection — all without a single pixel of video. github.com/ruvnet/RuView 5. agentmemory (+6.9K stars) #1 Persistent memory for AI coding agents based on real-world benchmarks github.com/rohitg00/agent… 6. supertonic (+3.6K stars) Lightning-Fast, On-Device, Multilingual TTS — running natively via ONNX. github.com/supertone-inc/… 7. CloakBrowser (+7.0K stars) Stealth Chromium that passes every bot detection test. Drop-in Playwright replacement with source-level fingerprint patches. 30/30 tests passed. github.com/CloakHQ/CloakB… 8. ViMax (+2.7K stars) "ViMax: Agentic Video Generation (Director, Screenwriter, Producer, and Video Generator All-in-One)" github.com/HKUDS/ViMax 9. 12-factor-agents (+1.9K stars) What are the principles we can use to build LLM-powered software that is actually good enough to put in the hands of production customers? github.com/humanlayer/12-… 10. bun (+2.0K stars) Incredibly fast JavaScript runtime, bundler, test runner, and package manager – all in one github.com/oven-sh/bun The theme this week: agent memory, context efficiency, and on-device intelligence are making AI infrastructure the hottest build category. Bookmark this. Next week's list will look completely different.

@ZippyNetworks @sharbel Take a single moment of your day to research toxic injection

@ZippyNetworks @sharbel The absolute best attack vector is a library, sometimes libraries that are trusted are linked to other trusted libraries,it only takes one in the supply chain and everything downstream is polluted. Didn't you see Github was literally just fucking hacked!???

@ZippyNetworks @sharbel You act very naive when it comes to threat vectors

@ZippyNetworks @sharbel Repos get hacked, contributions get hacked, libraries get hacked and injected. Popular trusted repos are the target.

@ZippyNetworks @sharbel Think whatever you want. That repo introduced that trojan.

@ZippyNetworks @sharbel Ok. Well, I posted what windows defender said. Thats the repo it came from.

@spencerpratt @RealAlexJones @HarrisonHSmith LFG!!!

They not like us

People always ask me, “how can we fix LA? We don’t have the money!” We have the money…it’s just being stolen! You wanna fix the budget? END THE FRAUD. I will bust the scammers and put your tax dollars back to work for YOU.

@RealAlexJones You HAVE to get Spencer Pratt on. Running against Mayor Bass. @HarrisonHSmith

Watch LIVE: Trump Extends Offer To Iran To Join Abraham Accords, Touts New Unparalleled World Coalition! Plus, Netanyahu Tells Confidants That Israel Is Having Difficulty Influencing Trump Decisions On Iran! twitter.com/i/broadcasts/1…

@FCB_Cartel @sama Its all about the PLAN and GOAL prompt. First plan, then tell it plan every line item, then make a goal that excludes exactly zero PLAN line items and executes them sequentially. Verify clean git push after all slices. Verify, and chain down bro.

Codex in the last 3 days has been a nightmare. Hallucination after hallucination, writing so much unwanted code, creating things that are not at all intended, taking too much time and burning tokens unnecessarily. Especially /goal. @sama something is wrong. It burned 50% of my Pro plan tokens for the week in 1 night creating a PR of size 104k lines and when asked about the PR : it said it’s useless and cleared the whole PR and closed it.

@FCB_Cartel @sama It has to be single minimal slice, and yiu basically have to code like yoir problem solving. Do not use containment, use countermeasures, verify countermeasures, next minimal slice. Fist PLAN, then GOAL, and then-execute-plan will be allowed in superpowers.

@edenarox @KorduGG @argofowl Same pain 😄

@UsernameAndStuf @KorduGG @argofowl same pain? xD

not enjoying working with codex right now so much resistance would fix one thing and another thing breaks i audited every agents.md file i have, all the codex settings to make sure it's not some dumb user side issue feels very similar to how opus 4.6-4.7 was i was debugging sluggish scrolling in my swift app, after 15 min and a lot of "found the issue", "fixed the root cause" it completely broke the scrolling, there is no scrolling at all now my prompts are pretty detailed too, using the prompt guidance resource that openai updated for 5.5 using xcode bridge mcp, logging, build macos apps plugin, etc. i need to try using codex cli and see if it's a codex app issue feelsverybadman, killed my mood, don't feel like doing anything right now contemplating trying composer 2.5 even though i told myself to stick to 1 tool for a few weeks to waste less time i'm almost certain there is some nasty bug that is causing this output quality regression but take this with a grain of salt

@ziwenxu_ Yeah buddy!