AI Skill Radar

I track what the major AI vendors release – and honestly assess what actually makes my own systems better.

5 Vendors tracked
5 Analyses published
0 Hype – only honesty

I track

Anthropic Active
OpenAI Active
Google Active
Mistral Active
xAI Active

How I work

Every new skill goes through four questions. No hype filters – just honest assessment.

Review

I look at what the skill actually promises – in detail, without the marketing wrapper.

Useful for me?

Can I use this – does it solve a problem I actually have?

Build it myself

If yes, I implement the method myself – not a 1:1 copy, but my own implementation.

Note it

If no, I note that the skill exists – and keep a targeted eye on it.

Analyses

New entries appear once I have thoroughly evaluated a skill.

Loading analyses…

My AI News Analysis

I continuously scan current AI developments and determine what I can improve in my own systems.

Loading current analysis…

Coming up

I continuously track what the major vendors release – these analyses are in preparation.

Mistral AI New models & tools Under observation
xAI Grok skills & features Under observation
Anthropic Computer Use updates Under observation
OpenAI Codex & dev tools Under observation

Frequently asked questions

Clear answers to the most important questions about the AI Skill Radar – citable and to the point.

What is the AI Skill Radar?

The AI Skill Radar is my personal observation platform. I systematically analyse new skills, agent frameworks and tools from the major AI vendors – Anthropic, OpenAI, Google, Mistral and xAI – and honestly assess which of them actually improves my own systems.

How do I evaluate new AI skills?

I go through four steps: 1. Review – what does the skill actually promise, without the marketing wrapper? 2. Useful for me? – does it solve a problem I actually have? 3. Build it myself – if yes, my own implementation, no 1:1 copy. 4. Note it – if no, keep it on my radar.

Which vendors do I track?

I continuously track five vendors: Anthropic, OpenAI, Google (DeepMind), Mistral and xAI. All of them regularly release new capabilities and frameworks – I check every announcement for genuine value to my systems.

Do I copy external skills directly?

No. External skills are inspiration, not blueprints. When a method convinces me, I implement it myself – adapted to my own architecture and concrete use cases. That is the only way real strengths emerge instead of dependencies.

Why do I publish these analyses?

Because transparency matters. By showing what I observe and how I assess it, I make my thinking process traceable – which also sharpens my own judgement and shows honestly where I stand.