09/06/2026
I've been using AI for over a year now.
And for most of that time, I was using maybe 20% of what it could actually do.
Then I discovered something that changed how I use it entirely.
Most people type a question into Claude and wait for an answer.
What very few people know: there are commands that completely transform how it responds.
I use three of them every single day now — in my work at Provision, in my content, in my thinking.
Here's what they are and what they actually do:
→ /L99 — Expert mode, no filter
Type /L99 before any question and Claude stops simplifying. It responds like a domain specialist talking to a peer. No "it depends." No hedging. Just the committed, specific answer a real expert would give.
I use this when evaluating cybersecurity vendors, reviewing architecture decisions, or when I need a second opinion that doesn't pull punches.
Try it: /L99 Should I use zero-trust architecture for a 300-person manufacturing company?
The difference from a normal response is night and day.
→ /steelman — The best case for the opposite view
Before any big decision, I run this.
/steelman forces Claude to build the strongest possible argument for the position I disagree with. Not devil's advocate. The actual best version of the other side.
I used it last month before signing a major vendor contract. It found three risks I hadn't considered. We renegotiated.
Try it: /steelman Moving all enterprise data to public cloud is the right call for Indian SMEs
→ /uda — Military-grade root cause analysis
When something goes wrong — a security incident, a team failure, a missed target — I don't want opinions. I want the actual root cause, three layers deep.
/uda runs a structured breakdown: Understand → Diagnose → Analyse → Act.
It's the closest thing I've found to having a senior consultant in the room who isn't afraid to tell you the uncomfortable truth.
Try it: /uda Our IT team keeps missing security patch deadlines despite having a process in place
These aren't hacks. They're not tricks.
They're the difference between using AI as a search engine and using it as a thinking partner. I'm a Founder and CTO of a tech company. My job is to make better decisions faster than the competition. These three commands are part of how I do that.
If you're in IT leadership, security, or running a team — save this post.
You'll use it.
Which of these would you use first? Drop it below.
