Improve the AI wargaming post in light of the feedback from Claude.
Good post — solid practical content with real examples. Here's my honest feedback:
What's working well
The NotebookLM/Gemini Gem workflow is genuinely useful and specific. The actual AI response examples are the strongest part of the post — showing real outputs is far more persuasive than just describing what AI can do. The conclusion is also thoughtful and avoids being either breathlessly enthusiastic or dismissive.
Where it could be stronger
The intro list never gets revisited. You open with four use cases (rules, scenarios, list gen, strategy) but only really cover rules understanding and list/strategy via the examples. Scenario/map generation gets dropped entirely. Either cover it, or trim the list to what you actually discuss.
The examples need more of your voice. Right now the posts are quite heavy with the AI-generated text blocks and light on your commentary around them. After each example, you could add a sentence or two like "This matched what I found in the rulebook" or "The list it suggested was actually pretty playable — here's what I'd change." That turns it from a demo into a review.
"Interrogate" is misspelled — you have "interogate" in the Void Admiral section.
The Altar of Freedom example is underserved. It's an interesting use case (historical OOB generation) but gets almost no commentary. Why is that useful? Did you actually use this list in a game? That's a more interesting story than just showing the output.
The tags are oddly narrow. The post covers Altar of Freedom and Void Admiral but is tagged only sci-fi/Void Admiral. Worth adding tags for ACW, Bolt Action (even if you didn't use it here), NotebookLM, Gemini — anything that might bring in search traffic from people who play those systems.
SEO opportunity. The meta description is just "a description of how I use AI…" — quite generic. Something like "How to use NotebookLM and Gemini Gems as a wargaming rules expert — with real examples from Void Admiral and Altar of Freedom" would pull in more specific search traffic.
Minor tone thing. "pocket friend" in the conclusion is a bit odd — probably means to say something like "AI opponent" or "AI assistant."
The bones are solid. The main fix I'd prioritise is adding more of your own reaction and judgment around the AI outputs — right now the AI responses do most of the talking, and you're the one readers came to hear from.
Improve the AI wargaming post in light of the feedback from Claude.