iiinova — where your digital self keeps evolving

this wants to be an MCP tool.

the stackoverflow chronicles.

020

coming for atproto. npmx is a bonus.

020

ul>li*5>a was a life hack before ai.

000

Devarno @devarno.com · 01/06/2026

used to feel like cheating.

100

next: agents on the fleet.

000

Devarno @devarno.com · 01/06/2026

time-to-first-runner-pickup, mostly under 5 seconds. all bursts absorbed cleanly. #buildinpublic

HERMES dashboard, THROUGHPUT tab, 24 hour window. Three stacked
panels. Top: throughput stacked by conclusion, showing success in
green and failure in red over time, with a large success peak in
the back half of the day. Middle: concurrency over time, a sawtooth
pattern of ephemeral runner bursts escalating in amplitude across
the window. Bottom: queue wait distribution (time-to-first-runner-
pickup), with a tall 0-5s bar, a smaller 5-10s bar, a tiny 10-20s
bar, and all buckets from 20s to 1800s+ empty.

140

Devarno @devarno.com · 31/05/2026

this is the only correct response to an empty house.

000

Reposted by Devarno

daniel wraith @danielroe.dev · 31/05/2026

I would like to stop having Thoughts about LLM usage

9811

Reposted by Devarno

npmx @npmx.dev · 17/04/2026

the npmx.dev team is 250 humans strong now 🤍

211212

Devarno @devarno.com · 24/05/2026

mostly as a ci runner host. ephemeral, bare-metal, 24/7. actions compounding fast across orgs so centralising onto one box just made sense. this is what i watch it through. what do you use yours for?

HERMES dashboard health overview. Top row shows 216 runs in the
last 24 hours, 117 failures, 0 currently in flight, and a 54%
failure rate. A throughput chart below shows success and failure
volume over time. At the bottom, a failure heatmap maps failures
by day of week and hour of day.

servers, networks, firewalls, load balancers, DNS, object storage. one tab. the control plane as a single surface.

000

bridging a dedicated server into one of these networks. the dialog gives you the actual shell commands. ip link, ip addr, ip route. the docs link is a footnote.

Hetzner Console modal titled "Configuration of your dedicated Server"
explaining how to bridge a dedicated server into a cloud private network. Three requirements listed: create a private network interface with the correct VLAN ID and MTU 1400, use an IP from the vSwitch subnet 10.0.1.0/24, and add a route to the broader cloud network 10.0.0.0/16 via gateway 10.0.1.1. Below, a worked example shows five shell commands: three ip link commands to create and bring up the VLAN interface, an ip addr command to set the IP, and an ip route command to add the network route. A footer notes the configuration will not persist across reboots and links to docs.

100

private network and subnet exposed as separate resources. 10.0.0.0/16 outer, /24 carved inside, server attached at 10.0.0.2. all visible in one panel. other providers treat VPC like a configuration project. here it's a resource you make.

Hetzner network configuration page showing a private network named hermes with IP range 10.0.0.0/16, a 10.0.0.0/24 subnet, and one attached server at private IP 10.0.0.2.

100

cpu, disk throughput, IOPS. graphed in the provider console. the dashboard is the dashboard.

Hetzner Console graphs panel for server panel-1, showing three stacked time-series charts: CPU usage hovering near 1% with periodic spikes, disk throughput peaking around 40 KBps in regular bursts, and disk IOPS in matching 4-6 IOPS bursts.

100

daniel wraith @danielroe.dev · 22/05/2026

hetzner, week one. the surface area is small enough to enjoy.

Hetzner Console dashboard showing a tile grid with resource counts across servers, load balancers, primary IPs, floating IPs, volumes, networks, firewalls, buckets, storage boxes, and DNS zones, alongside a map of European datacenter locations and a recent activities feed.

200

Reposted by Devarno

💥 just released a github action: `uppt` it aims to be a very simple, secure release workflow for maintainers that adheres to best security practices ⚪ trusted, staged publishing on npmjs.com ⚪ github environment protection ⚪ generated release PR + changelog ⚪ automatic release + publish on merge ✨

1116228

the way I think about problems has changed more than the tools themselves. the abstraction layer has lifted. you'll see it in what I ship.

i am not the same developer i was 2 years ago, and honestly not even the same one i was 6 months ago. the way i build, ship, and think has shifted completely, and i wouldn't have it any other way. #buildinpublic

270

the attacker's lifeline goes silent for the longest stretch. that's the attack.

A sequence diagram showing five actors across the top: the attacker, the prep room, the assembly line, the shop shelf, and the filing cabinet.
Four phases, separated by labelled bars.
Planting: the attacker offers to contribute to the project. the contribution is granted write access to the prep room. the attacker poisons the cache.
Eight hours — the attacker does nothing. the prep room is marked: armed and waiting.
Trigger: a legitimate merge starts the build. the prep room returns the poisoned cache. the assembly line ships to the shop shelf with a valid attestation. a note sits on the shop shelf: the certificate was genuine. the build was not.
Spread — 45 minutes: the assembly line harvests credentials from its own memory. injects into every reachable package on the shelf. accesses the filing cabinet. writes itself into .claude and .vscode — survives uninstall.
Dead man switch: the assembly line sends a dashed arrow back to the attacker — token rotated, home directory wiped.
The attacker's lifeline is silent for the longest stretch in the diagram. Everything that matters happens without them.

010

6/ eight hours of silence. that wasn't patience. that was the whole attack.

A quote card. Frank Herbert photographed in black and white, alongside his words from Dune: "The Fremen were supreme in that quality the ancients called spannungsbogen — which is the self-imposed delay between desire for a thing and the act of reaching out to grasp that thing." The worm waited eight hours. Herbert named it in 1965.

000

5b/ the partnership with Vect (ransomware infrastructure) was announced before this wave shipped. the next wave won't be credential theft. it won't be a worm either.

A quote card from the Akamai Security Intelligence Group report, May 2026. Text reads: "we noted that the ransomware group Vect had announced a partnership with TeamPCP, hinting at a possible shift from credential theft toward large-scale extortion and ransomware operations. This new campaign offers the first concrete signal that this shift may already be underway."

5a/ so why didn't they move faster? they had credentials. they had access. they had a live worm inside a major ecosystem. but they waited. they ran the spread methodically. they armed a dead man switch before touching anything irreversible. someone wrote a spec. the worm had a project manager.

100

3/ remediation was the attack. if the chef changes the locks to stop it, the worm destroys the entire kitchen first. rotating your credentials was already in the plan. npm uninstall didn't touch it. it was already living in the walls.

100

2/ enter shai-hulud. it reads credentials from its own memory. it injects itself into every package it can reach. it writes itself into the walls and arms a dead man switch against whoever tries to stop it. the worm that swallows everything. named deliberately.

100

1/ the attacker did nothing for 8 hours. a real chef merged code (like they do every day) and the assembly line pulled the bad ingredients automatically. the health certificate was applied and the meal shipped. every verification layer said yes.

100