CoderLuii

I built HolyCode. OpenCode in Docker. Use your Claude subscription instead of an API key. One container, every tool, any provider. @CoderLuii holycode.coderluii.dev

@sama switched to codex and the biggest difference wasn’t quality it was finally being able to stay in flow without hitting limits that changes everything

To celebrate 3 million weekly codex users, we are resetting usage limits. We will do this every million users up to 10 million. Happy building!

the biggest advantage in ai coding right now is not talent. it’s having a setup that lets you move from idea to execution with almost no resistance.

Life after switching from Claude Code to OpenCode.

models are improving faster than most people can rebuild their habits around them that’s kind of crazy when you think about it what’s your current ai workflow looking like?

most people are still using ai like a chatbot. builders are using it like infrastructure.

I’m the developer behind HolyClaude and this is honestly incredible to see. Thank you for sharing it like this, seriously means a lot. The whole thing started because I was tired of wasting hours on setup and environment issues before writing real code. I wanted one command that just works so builders can start shipping immediately.

🚨BREAKING: A developer just built what AI coding infrastructure should have been from day one. It's called HolyClaude. Claude Code + web UI + 5 AI CLIs + headless browser + 50+ dev tools one command to launch. No. Manual installs. No. Config hell. No. Permission nightmares.

@AIDailyGems Appreciate the feature 🙌 HolyCode is built for one goal: remove setup friction so you can open a browser and start building immediately. If anyone here tries it, I’d love your feedback (especially what you want added next).

AI coding workstation with OpenCode, Claude support, and 30+ tools. Features a headless browser and multi-agent orchestration. github.com/CoderLuii/Holy…

Fast feedback > perfect architecture. You can recover from ugly code. You can’t recover from months of silence. Ship, measure, adjust, repeat. What are you shipping this week, specifically?

repo: github.com/coderluii/Holy…

most people are benchmarking models. i’m benchmarking recovery time. how fast can i break, rebuild, and get back to shipping? repo in first reply.

@Kynlix @ErickSky jajaja gracias bro, de verdad 🙏 si me van a dar premio que sea en forma de bugs menos y deploys más rápidos 😂 yo solo voy compartiendo todo lo que me habría ahorrado semanas cuando empecé.

@ErickSky @CoderLuii q locura lo q hace este chico compartiendo todo mereces un premio bro 🫡

🚨A los que les gustó HolyClaude, el autor nos acaba de dar la versión para OpenCode. Presentando [HolyCode] el nuevo repo de @CoderLuii El win más grande: PERSISTENCIA REAL. Tus sessions, settings, plugins y config de MCP viven en un bind mount → rebuilds, updates o lo que sea... tu progreso NO se pierde nunca. → Modo Claude-Auth opcional (usa tu propia suscripción de Claude, ya sabes los riesgos). → Modo oh-my-openagent opcional (flujos multi-agent potentes). Open source, transparente, estable y pensado para workflow diario serio. REPOOO👇

@supapa_0 @ErickSky no bro, no es solo Linux 👌 HolyCode corre en: → Linux (amd64 + arm64) → macOS con Docker Desktop → Windows con WSL2/Docker Desktop la experiencia más estable sigue siendo Linux, pero no estás limitado a Linux.

@ErickSky @CoderLuii sólo Linux?

BROOO esto está durísimo 😂🔥 gracias por el apoyo de verdad. esa parte de la PERSISTENCIA era justo el dolor que quería matar: rebuild, update, cambio de máquina… y tu progreso sigue vivo como cucaracha nuclear 🪳 HolyCode va en modo “daily driver serio”: default limpio y estable, y si quieres ponerte en modo turbo: → ENABLE_OH_MY_OPENAGENT=true → ENABLE_CLAUDE_AUTH=true (opcional, cada quien con su riesgo) si tu gente quiere, me meto al hilo y les respondo TODO en modo técnico + sin humo.

new model drops every week. your environment still breaks the same 4 ways. fix that once and stop restarting your life. repo in first reply.

@Pranto39 HolyCode - my open-source AI coding workstation in Docker. Built around OpenCode with persistent state, browser tooling prewired, and multi-provider flexibility so you can ship faster without rebuilding setup every machine switch. github.com/coderluii/holy…

Drop your project url Let's drive some traffic

@CiprianiRanieri HolyCode — my open-source AI coding workstation in Docker. Built around OpenCode with persistent state, browser tooling prewired, and multi-provider flexibility so you can ship faster without rebuilding setup every machine switch. holycode.coderluii.dev

Share your product. Let's get some traffic for you!

you need a review budget. i use: max 2 plan-review passes only fix blockers (security/data loss/core architecture) everything else goes into v2 backlog then start shipping and let real execution reveal what matters without a review budget, agents optimize for more findings, not progress.

My biggest challenge with vibe coding / agentic engineering lately has been getting stuck in what I call a "plan doom loop" - have AI write a plan - review myself, seems good - have AI review plan, it always finds something - repeat It drains my time and energy to determine how important the "findings" really are Who has solved this

repo: github.com/CoderLuii/Holy…

"which ai coding tool should i use?" the one you can run tomorrow, on a different machine, with the same state, without drama. that’s the bar. repo in first reply.