---
title: "We built an org-wide AI agent in 4 days. Here's what broke in the weeks after."
url: https://daily.dev/posts/we-built-an-org-wide-ai-agent-in-4-days-here-s-what-broke-in-the-weeks-after--tdngi1q4d
source_url: https://daily.dev/blog/we-built-an-org-wide-ai-agent-in-4-days-heres-what-broke-in-the-weeks-after
type: article
source: "daily.dev"
author: "Ido Shamun"
published: 2026-03-29T17:38:46.863Z
updated: 2026-03-30T14:00:41.150Z
tags: ["security", "nodejs", "typescript", "docker", "ai-agents"]
reading_time: 13
upvotes: 98
comments: 21
language: en
---

> ## Documentation Index
> Fetch the complete documentation index at: https://daily.dev/llms.txt
> Use this file to discover all available pages before exploring further.

# We built an org-wide AI agent in 4 days. Here's what broke in the weeks after.

**[daily.dev](https://daily.dev/sources/daily)** · [@idoshamun](https://daily.dev/idoshamun) · 13 min read · 98 upvotes · 21 comments

## Summary

daily.dev built 'Smith', a 29K-line TypeScript AI agent integrated into their Slack workspace in just 4 days using Codex. The post covers the production incidents and security challenges that followed: credential leaks in a shared runtime requiring a growing command sanitizer, GitHub token bleeding between user sessions, a Node.js event-loop hang that systemd couldn't detect (fixed with a watchdog + health checks), memory exhaustion from a power user's long conversations (fixed with cgroup limits), and a progressive tool disclosure system to manage 60+ tools. Smith self-authors its own reusable skills via a git-backed 'brain' repo and now runs autonomous nightly tasks like spam sweeps and A/B experiment audits. Known remaining issues include an unaudited skill brain, incomplete command sanitizer, and an unsolved crash pattern from one heavy user.

## Full article

daily.dev links to this article rather than hosting it. Read it at the original source: <https://daily.dev/blog/we-built-an-org-wide-ai-agent-in-4-days-heres-what-broke-in-the-weeks-after>

## Community discussion

Top comments from developers on daily.dev.

**@rsteadman** · 7 upvotes

> Appreciate the honesty and openness, good read.

**@grantkennedy** · 3 upvotes

> I misread the title and thought this person built an "orgy-wide AI agent."

**@webmasterccesinc** · 2 upvotes

> Thanks for sharing your experience with implementing AI into your system.
> We hear so much hype about AI productivity, but never enough about what it really takes to make it productive, stable, and implemented correctly.
> I am a project manager / senior software engineer working with embedded systems using FPGAs, and I've been wanting to integrate AI into our CI/CD pipeline
> But in developing a context management system, I've run into many " gotchas" and "oh by the way" in trying to configure a proper repeatable workflow.
> Thanks again for sharing your journey

**@hypnoticdata777** · 2 upvotes

> Hmmm.. tell me more

**@ezpie** · 2 upvotes

> well what did we learn from this kids? Never let AI agents on the lose, those things are scawy

## Similar posts on daily.dev

- [OpenAI Codex adds SDK, admin tools, Slack integration](https://daily.dev/posts/openai-codex-adds-sdk-admin-tools-slack-integration-odlz3t0wo) · InfoWorld · 1 upvotes · 0 comments

---

Tags: [#security](https://daily.dev/tags/security), [#nodejs](https://daily.dev/tags/nodejs), [#typescript](https://daily.dev/tags/typescript), [#docker](https://daily.dev/tags/docker), [#ai-agents](https://daily.dev/tags/ai-agents)

[View this post on daily.dev](https://daily.dev/posts/we-built-an-org-wide-ai-agent-in-4-days-here-s-what-broke-in-the-weeks-after--tdngi1q4d)