5 New Prompts & Today's Actionable News
By Robert MacKelfresh — August 11, 2026Write the decision down so it stays decided
EASYWe decided something. Turn it into a note that means nobody has to reopen it in four months. 1. THE DECISION, in one sentence, in…Open full prompt →for ChatGPT or ClaudeRehearse the conversation, don't plan it
EASYYou're going to play them. Be realistic, not difficult for the sake of it. Don't concede quickly. If I make a weak argument, push on…Open full prompt →for ChatGPT or ClaudeFind the questions your policy will trigger
MEDIUMWe're about to announce this policy. Before we send it, tell me what happens when they read it. 1. THE FIRST QUESTION THEY'LL ASK…Open full prompt →for Claude or ChatGPTWrite the handover before you go away
ADVANCEDI'm away from [DATE] to [DATE]. Interview me and then write the handover. Ask me ONE question at a time and wait for my answer…Open full prompt →for Claude or ChatGPTStress-test it before you automate it
EXPERTI'm about to automate this. Before I build anything, find what breaks. 1. THE HUMAN JUDGEMENT I HAVEN'T NOTICED. Go through my steps…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.
Anthropic's Compliance API now returns transcripts of Claude Code sessions run on employees' own machines
here's the access map — three endpoints, an existing key, no new provisioning, and no announcement
DATA TABLEGemini reached 1 billion monthly users, and 63% of them now talk to it rather than type
here's the data table and why the 63% matters more than the billion if you write anything you want found
NVIDIA's Nemotron 3.5 Lightning holds 30B parameters but runs only 3B per token, on one consumer GPU
→here's the head-to-head against Meta's 30B from the day before, including the licence difference that matters commercially
GET THE COMPARISON TABLEChatGPT Ads launched in the UK, Mexico, Brazil, Japan and South Korea, on Free and Go plans only
→here's the exposure checklist and the five-minute test to see what now appears beside answers about your category
GET THE EXPOSURE CHECKLISTMistral began hosting other companies' open models on its own infrastructure, starting with Z.ai's GLM-5.2
→here's the options table for running open models under a residency constraint, and the three requirements people keep conflating
GET THE OPTIONS TABLE
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].