All Posts

The new feature for custom script

Not sure if this "PAT" part will be the final solution, or i can make i better But the goal is to have a community to share custom scripts

You asked for it and you got the answer, very helpfull information / Tips you are getting, beside the /diff, but it good to be reminded on what you can do better, learning is the key here

That's a very strong feature, that can improve you agent in the repo you are working in, and also for you going forward..... and a reminder: Planning is everything, get the information out of your head and in to the agent knowlegede

Very useful feature if you forgot what you did yester day

## The Problem Every CISO Knows Your Azure Entra ID tenant is the front door to your entire cloud estate. Behind that door: hundreds of app registrations with over-permissioned secrets, service principals with stale credentials, conditional access policies with gaps you don't know about, and attack paths that chain these misconfigurations into real breaches. Microsoft provides tools. They're scattered across Defender, Entra admin center, Azure Portal, and Graph API queries you run manually. None of them give you a unified, scored, actionable view of your Entra security posture — and none of them let your team extend the scanning with organization-specific checks.

I am working on a Azure security scanner tool, and it gets new feature on as we go. My last idea i that i would make it possible to upload/make phyton script inside the apps dashboard, and make it run the scan, its must be with some templates etc, what you see below is a part of the Roadmap that is sent to Copilot that uses my Agent flow to build apps

Here is a structured, high‑quality set of prompt samples specifically designed for testing AI‑coder apps like Copilot, Cursor, Claude Code, Windsurf, Kiro, Gemini Code Assist, etc. These prompts are crafted to expose differences in: - multi‑file reasoning - refactoring quality - architecture planning - debugging - agentic workflows - spec‑driven development - UI generation - API design - database modeling They are ideal for benchmarking how well each tool performs in real development scenarios.