Long-form on what actually ships.

The n8n Lead Stack: One Lead, Five Workflows, One Shared Ledger

An install script chains 5 published n8n workflows on one throwaway instance: install in 112 seconds, 5 of 5 inactive, a second install run refused on purpose.

Sep 28, 20268 min

The n8n Silent Lane Watchdog: Quiet Is Not Always Failure

An n8n workflow checked 63 lanes against their own expectation rows: 1 lane missed a test threshold, 0 false alarms on the lanes that were quiet by design.

Sep 25, 20267 min

The n8n Meta Ads Circuit Breaker: Our Own Breaker Missed Our Own Loss

An n8n workflow read our own ad set's insights on 2026-09-23 and found $70.15 spent, 0 leads, $10.15 over the $60 cap, a loss our older campaign-level breaker missed.

Sep 24, 20268 min

The n8n Shared Inbox Router: Rules Sort the Mail, a Model Reads What Is Left

An n8n workflow sorted 12 of 13 test messages with rules before any model saw them, 92.3%, clearing the 60% ship gate. Drafts only, no send node.

Sep 24, 20267 min

The n8n Clinic WhatsApp Booking Agent: It Refuses by Code, Not by Prompt

An n8n workflow handed off 10 of 10 medical and price questions by code before the model on 2026-09-22, then booked 4 of 12 simulated WhatsApp patients with 0 double bookings.

Sep 23, 202612 min

The n8n Freight Quote Email Parser: One Email In, Eight Fields Out

An n8n workflow read 20 freight quote emails on 2026-09-18, filling 6.75 of 8 fields each with Claude Haiku 4.5 and writing one Postgres row per email.

The n8n Shorts Factory Workflow: One Topic In, a Captioned 9:16 Short Out

The n8n Shorts Factory workflow turns one topic string into a captioned 9:16 MP4: 28 nodes wiring Groq, ElevenLabs and a self-hosted ffmpeg service.

The n8n Speed-to-Lead Workflow: an AI Reply in About Two Seconds

An n8n speed-to-lead workflow answered a web form in 2.4s and 1.9s on 2026-09-17, then qualified the reply and proposed a visit window. MIT JSON.

n8n Google Maps Lead Harvest: Trade and City In, a Deduped Lead Table Out

This n8n workflow wrote 100 Denver roofing businesses to Postgres in 31.9 seconds on 2026-09-18, deduped on place_id via the Apify Maps actor.

Playbook

Edit Your Videos Using Claude Code

The video-editing skills behind every SkynetLabs reel, vlog, and promo — each with the full master prompt you can paste into Claude Code and run. Free. One email unlocks the lot.

Aug 10, 202612 min

The AI Ops Command Center: How Small Businesses Stop Missing Leads Forever

An AI receptionist that answers, texts back, and books your calendar in under a minute, 24/7 — built in your own GoHighLevel and n8n accounts. How it works, what it costs to run, and who it's actually for.

Jul 13, 202611 min

Playbook

Claude Fable 5 Prompts: 40 Master Prompts to Run Before July 12

40 master prompts for Claude Fable 5 — Claude Code, n8n, GHL and agency templates that build reusable assets.

Jul 11, 202625 min

Stack

The week AI agents started running my shop while I slept

A mid-2026 field note on running a small AI agency where autonomous agents handle overnight builds, lead triage, and content drafting — what they own, the one mistake that cost a morning, and the guardrails that make it safe.

Jun 28, 20268 min

Field notes

Bali co-working economics — what shipping from Canggu actually costs

A line-by-line breakdown of the monthly cost of running SkynetLabs from Canggu, Bali — including the categories digital-nomad blogs leave out: visa runs, scooter rental, fast Wi-Fi tax, and food delivery.

May 19, 202611 min

Stack

A weekend with Claude Code

A weekend rebuild of the SkynetJoe theme using Claude Code as the primary tooling. What got faster, what got harder, and the five things I still do by hand.

May 17, 20269 min

Pricing

Why I price my AI builds publicly while every agency hides the number

An essay on the strategic case for public pricing in AI services — how SkynetLabs filtered out a year of wrong-fit briefs by publishing four flat-rate tiers, and what we still negotiate.

May 15, 20268 min

LLMO vs GEO vs AEO: The 2026 Acronym Guide for AI Search Optimization

Three acronyms, one job. LLMO, GEO and AEO are largely the same discipline with different emphasis. Here's which term to use when, what overlaps, and where the work actually diverges.

May 13, 20268 min

How to Track AI Citations Across ChatGPT, Claude, Gemini and Perplexity

The manual tracking method I use on every AEO engagement before touching a paid tool. 20 prompts, four engines, one spreadsheet, weekly cadence. Costs nothing, works for six months minimum.

May 13, 20269 min

How to Rank in ChatGPT: Direct Answers for Brands That Want to Be Cited

Five schema types, an llms.txt file, FAQ blocks in claim-plus-example pattern, explicit dateModified, and a 20-prompt tracking set. The playbook that actually moves ChatGPT citation rate.

May 13, 202610 min

How to Get Cited by Claude: Direct Answers for Anthropic AI Visibility

Claude is the conservative engine. Marketing copy that gets pulled by ChatGPT gets filtered by Claude. The Wikidata + Crunchbase + claim-with-source playbook that actually lifts Claude citation rate.

May 13, 20269 min

AEO vs SEO: How They Differ and Why You Need Both in 2026

AEO targets AI citation rate across four engines. SEO targets blue-link Google ranking. Shared technical foundation, genuinely different surfaces. Why most service businesses in 2026 need both, sequenced sensibly.

May 13, 20269 min

Tools

The 6 paid tools every small fleet uses (and which 4 to delete)

An audit of the typical $800/mo small-fleet SaaS stack — load board, ELD, dispatch, factoring portal, two CRMs, accounting — with specific recommendations on which to keep, which to consolidate, and which to delete.

May 13, 202610 min

The AEO Playbook for Service Businesses: Getting Cited by ChatGPT, Claude & Perplexity

Answer-engine optimization is SEO's bigger, weirder cousin. Here's the practical 7-step playbook we use to get clients cited in LLM answers — without keyword stuffing or fake authority signals.

May 12, 202611 min