AI Agents vs AI Skills: Which One Should You Create?
Building a fleet of AI Agents sounds nice until you realize that you need to maintain your fleet for things that could have been a few skills.
Read more
Learn. Build.
Ship. Repeat.
This is my build-in-public AI skunk works blog where I post help articles, as well as, explore and discover the limits and benefits of Agentic AI for small to medium sized businesses.
What is an AI OS?
Every time you open a new chat with an AI assistant, you start over. You have to explain your business again, remind it who your clients are and what happened…
Why AI Assistants Forget Your Instructions
You’ve set up your AI assistants carefully, then the next chat they forget who you are, and what it’s supposed to. This is by design.
Agentic AI Terms You Need to Know
Confused by agentic AI terms like Skills, Plugins, MCP, and APIs? Here’s what each one actually means for your small business, in plain language.
What is Agentic AI?
The AI you’ve used so far is reactive. Every time you want a result, you prompt it, review it, and decide what to do next.
Why your AI Agents aren’t working: 3 tips you need to know
𝗧𝗵𝗲 𝗺𝗼𝘀𝘁 𝗰𝗼𝗺𝗺𝗼𝗻 𝘀𝘁𝗿𝘂𝗴𝗴𝗹𝗲 𝗜 𝘀𝗲𝗲 𝘄𝗶𝘁𝗵 𝗔𝗜 agents 𝗶𝘀 𝘀𝗲𝘁𝘁𝗶𝗻𝗴 𝘂𝗽 𝘁𝗵𝗲 𝘁𝗼𝗼𝗹, 𝘁𝗵𝗲𝗻 𝗯𝗲𝗶𝗻𝗴 𝗰𝗼𝗻𝗳𝘂𝘀𝗲𝗱 𝘄𝗵𝗲𝗻 𝗶𝘁 𝗱𝗼𝗲𝘀𝗻’𝘁 𝗱𝗼 𝘄𝗵𝗮𝘁 𝘆𝗼𝘂 𝘄𝗮𝗻𝘁. Read a LinkedIn post or watch…
AI Degradation: What to do when your AI tool loses it’s mind
AI degradation is one of those things nobody warns you about until it has already cost you hours. I learned that the hard way trying to finish a task that…
The Samaritan Project
Vibe Coding The Samaritan Project Agentic OSINT platform built with 🦞 OpenClaw and running on ParrotOS. The Samaritan Project The Problem: Public record research is slow by design. Every time…
Agentic AI for Small Business: Inside the Samaritan Project
If you’ve been wondering what Agentic AI can actually do for your small business not in theory, but in practice, this is worth your time. Most small business owners are…
Build Log #10: The Choosing the Right AI Model Problem
One of the hardest early lessons in building Samaritan was that not all local models are created equal, and the gap between “runs on your hardware” and “actually useful for…
Build Log #9: The Moment I Realized all my OpenClaw Agents were in the Wrong Place
At a certain point in any build you have to stop adding and start verifying. Things seemed to be going too well and based on previous experience, I started getting…
Build Log #8: The OpenClaw Update War
OpenClaw updates were coming in fast and furious with patches and security fixes and I had to keep things up to date. Upgrading software on a production system you’ve heavily…
Build Log #7: AI Development – Working Smarter
At some point I looked up and realized I was wasting time. The build was deep enough that managing it through chat and the terminal was frustrating and slowing me…
Build Log #6: Giving My AI Agent Its Own World
One of the core goals from the beginning was that Samaritan would have its own everything. Its own hardware, its own accounts, its own email, its own communication channels. Not…
Build Log #5: Creating the Seer Orchestration Agent
Having a fleet of specialist agents is only useful if something can coordinate them. Multiple investigative agents sitting in separate rooms with no one managing the case is not an…
Build Log #4: Building the OpenClaw Agent Fleet
Once I had the methodology, the next question was how to actually set it up. A single general-purpose OSINT agent sounds good, but in reality it produces jumbled results. Ask…
Build Log #3: Building the OSINT Methodology
Getting Samaritan to hold an identity was only half the problem. The other half was making sure that once it responded, what it produced was actually useful.
Build Log #2: OpenClaw Kept Forgetting Its Name
The first real problem with Samaritan wasn’t a missing feature.\. It was more basic than that. Every time OpenClaw started up, it forgot who it was.
Build Log #1: Why I built a cool OSINT Platform on OpenClaw
Years of collecting OSINT resources, a sci-fi vision of what an intelligence platform could be, and a hard stance on data privacy led to building Samaritan on OpenClaw.
Why generic AI prompts produce garbage results, and what to do instead
The problem with generic prompts You’ve probably seen them all over LinkedIn. The endless carousel posts and viral lists promising ChatGPT prompts that do something magical without needing to put…
Writing with AI: 5 Essential Tips to Keep Your Voice
Writing with AI is the subject of a never ended string of LinkedIn posts whose opinions range from warning that you’re “damaging your brand” to claims that it will result…
AI Prompts vs AI Workflows: What’s the difference?
If you’ve dabbled in AI tools recently, you’ve likely heard the terms AI prompts vs AI workflows used interchangeably, but they are not the same. This article discusses the real…
7 Simple Tips for Better AI Prompts
You’ve probably seen lists of generic AI prompts all over Linkedin. The ones that promise to change your life and 10x your business. So you give them a try and…
