5 New Prompts & Today's Actionable News
By Robert MacKelfresh — August 12, 2026Check a claim before you repeat it
EASYI'm about to repeat this claim: [THE CLAIM, IN THE EXACT WORDS YOU'D USE]. Check it before I do. Use the web. 1. WHAT'S THE PRIMARY…Open full prompt →for ChatGPT, Claude or Gemini (with web access)What this announcement changed for you
EASYHere's a long announcement. Here's me: [YOUR JOB, THE TOOLS YOU USE]. Don't summarise it. I can read. Tell me what changed for me…Open full prompt →for ChatGPT or ClaudeCompare two tools on your actual job
MEDIUMI'm choosing between [TOOL A] and [TOOL B] for [THE SPECIFIC JOB]. Don't give me a feature matrix. Both marketing pages have one and…Open full prompt →for ChatGPT, Claude or Gemini (with web access)Work out what AI is really costing you
ADVANCEDI'm going to give you our AI-related spend. Help me work out what's actually happening. 1. THE TOTAL, MONTHLY AND ANNUALISED. Say…Open full prompt →for Claude or ChatGPT (file upload)Build a test for a task you keep rerunning
EXPERTI keep running this task through AI. Every time a model changes I have no idea whether it got better or worse, so I want a test I…Open full prompt →for Claude or ChatGPT
The Context Coda
One more thing: use anything you already know about me from our past chats and anything you're connected to. If you can find it yourself, don't make me look it up.
Paste this at the end of any prompt above. It tells your AI to use what it already knows about you instead of asking you to repeat it — which is the difference between a generic answer and a personal one.
Actionable News
What just happened, and the prompt, skill file, or setup that turns it into something you can use.
OpenAI's enterprise data shows Codex use grew 108× in legal since February, against 5× in engineering
here's the full adoption table, and the finding underneath it — what separates leading firms is whether people connected anything, not how much they chat
PRICING TABLEGrok 4.6 has a 500K context window, but crossing 200K rebills every token in the request at double
here's both pricing tiers and the agent-loop trap, because accumulated context crosses that line without anyone deciding it should
Lovable raised $400M at $13.3B and reports presence in nearly two-thirds of the Fortune 500
→here's what the company reports, plus the question it actually raises — whether anyone at your company has already shipped something on a card
GET THE NUMBERS TABLEGemini Enterprise's GitHub connector went GA with write actions — it can merge pull requests now
→here's the permissions checklist to run before an agent has commit rights on your main branch
GET THE PERMISSIONS CHECKLISTAn outlet confirmed a 50% Claude price rise two days after Anthropic cancelled it
→here's the sixty-second checklist for verifying any AI pricing claim, because most wrong AI news is stale rather than invented
GET THE VERIFICATION CHECKLIST
The AI Five is published by Automation Squad. Every prompt is written from scratch and every claim links to a primary source. Corrections: [email protected].