seyz

I build a governed infrastructure for internal tools and AI agents: https://rootcx.com

Everybody uses Claude Code or AI coding tools to build internal software, but they lack the governed infrastructure layer required for enterprise trust. RootCX provides that missing foundation. We offer the security, auditability, hosting and permissioning primitives necessary to move internal software from "cool demo" to prod

vs rootcx (https://rootcx.com) :D

The return of Arc browser?

I'm working on RootCX (https://github.com/RootCX/RootCX), a platform for building and shipping internal apps and AI agents in production.

Think of it like "Claude code on Supabase", but for internal apps and AI agents.

I got tired of choosing the deployment platform, wiring up Postgres, SSO (OIDC), RBAC, audit logs, secret vaults, integrations/tools/MCP, ... from scratch every time I needed an internal tool.

MCP versus Skills -> wrong debate. MCP versus CLI -> real debate.

Nice launch! It reminds me a lot RootCX (https://github.com/RootCX/RootCX)

toxic.

This is how you gift wrap the agentic era to the open source chinese LLMs. devs don't need the best model, they need one without lawyers attached.

the real insight here isn't that sizing is broken. Everyone knows that. It's that fixing it would require brands to admit their current customers don't match the label they've been selling them. "You're not a size 6, but size 10" is bad for business

Including 999 using Copilot.

No phd, no funding committee, no peer review anxiety. Just curoisity and paper. Sometimes the breakthrough comes from not knowing the problem was supposed to be difficult. This is what happens when you don't tell a kid something is too hard

This is why Ctrl+C is 0x03 and Ctrl+G is the bell. The columns aren't arbitrary. They're the control codes with bit 6 flipped. Once you see it, you can't unsee it. Best ASCII explainer I've read.

The moat here is local inference. Whisper.cpp + Metal gives you <500ms latency on M1 with the small model. no API costs + no privacy concerns. Ship that and you've got something the paid tools can't match. The UI is already solid, the edge is in going fully offline.

Side projects don't die from lack of time. They die from success anxiety. Shipping means facing judgment. An eternal WIP stays safe in the "potential" zone where it can't disappoint anyone including yourself

Debugging an LLM integration without seeing the reasoning is like debugging a microservice with no logs. You end up cargo-culting prompt changes until something works, with no idea why.

128k tokens sounds great until you see the bill

Classic antirez. Does more in 500 LOC than most frameworks in 50k.

ok

LLM failures go viral because they trigger a "Schadenfreude" response to automation anxiety. If the oracle can't do basic logic, our jobs feel safe for another quarter.

Wrong.

I built this because running OpenClaw locally felt a bit like running chmod 777 for the internet.

While Docker/VMs provide OS-level isolation, they don't solve the application-layer risk: if my agent hits a Prompt Injection on a malicious site, it can still use its authenticated tools to wipe my CRM or exfiltrate Slack data.

ClawBands is a middleware that hooks into the OpenClaw before_tool_call event loop. It enforces a synchronous blocking check before any dangerous operation (Shell Exec, File Write, etc.) is actually executed.

Ideally, this acts like sudo for your agent:

In Terminal (TTY): It pauses execution and renders an interactive prompt.

Headless: It can route approval requests via a dedicated tool (clawbands_respond) to Telegram/WhatsApp.

It's a 24h PoC written in TypeScript. I'm particularly interested in feedback on the blocking logic implementation, handling the async pause reliably in the event loop was the trickiest part.

Happy to answer any questions!

It's too late. I loved Heroku, but having 5+ major outages in a single year was unacceptable. The support was abysmal, and transparency was non-existent. After the Salesforce acquisition, we watched the company culture shift, getting worse and worse over the years. We were forced to migrate to AWS. Even though we were high-spend customers, the trust is gone. This post is a nice gesture of transparency, but it's way too fluffy for my taste. Heroku is dead to me, there’s no coming back from this.

I honestly couldn't tell if the GIF was lagging or if that's the actual typing speed. I give lessons to help reach double digit WPM if you're interested

I’m building RootCX (https://rootcx.com). A customer "operating system" replacing dozens of SaaS tools.

As a second-time founder, I've watched the SaaS boom create an ocean of best-of-breed tools. Each solving one slice of the problem. One solving it end-to-end.

Now every company runs on a patchwork of apps, APIs, and workflow hacks just to keep customer context alive. It's insane how normalized that's become.

RootCX starts from the opposite premise: the customer is the core, not the app. Everything: CRM, support, billing, workflows, AI, ... plugs into one shared customer base. Less juggling tools, more actually running the business.

Opening the page... eyes burned... closing the page. Ouch.

TL;DR: A good conference talk delivers value by informing (what you did and why it matters), educating (portable insights), and entertaining (keeping attention) all while aligning with the community’s shared values.

I don’t really have a good reason. It’s a hobby project. Like gardening, but with more segfaults.

Love it. You definitively deserve your +350 points!

[Show HN]?

Clearly, I've built dozens of admin panels and I was tired doing it… so I chose to dedicate all of my focus to resolving this issue.

Got it, thanks for the advice mtmail!

Where is Joel?