Tag: prompts
-
Relative to your interests, Friday
From: The adorable love story behind Wikipedia’s ‘high five’ photos ICYMI Original content published since last time. Shadow AI and Who Is Watching What the AI Bill Actually Looks Like People Actually Using It More Wastebook All things are possible, but this thing is more possible. AI Summaries I wanted to read these, but I…
-
What nobody tells you about writing agent skills
Say when to use them. 🔗 What nobody tells you about writing agent skills
-
London Productivity Library
Common prompts for local government in the UK. How lovely! There’s also small instructions for humans for common Office tasks like pivot tables and formatting in Word. 🔗 London Productivity Library
-
How to stop Claude from saying load-bearing
First Claude code hook prank? 🔗 How to stop Claude from saying load-bearing
-
Prompt to rank paintings, but explain how
How would you explain the differences between a ‘pretty good’ work by Mondrian, and an excellent work, referring to the abstract paintings here. 🔗 Prompt to rank paintings, but explain how
-
A system prompt to get AI to stop pretending to be human
A system prompt aimed at stopping models mimicking human social patterns, the “great question!”, “hmm, let me think” and “you’re absolutely right!” filler, so that output comes back directly. Summarized by AI. 🔗 A system prompt to get AI to stop pretending to be human
-
Related to your interests, Monday
Potsdamer Platz bei Nacht, Paul Paeschke, ca. 1929. Related to your interests Why your DIY Kubernetes stack won’t survive the era of agentic AI How tech chiefs gauge ROI on AI – AI uses: better search, preparing leases, and this: “When the system senses that ice is running low, it fires off an order to…
-
Relevant to your interests on Monday
Don’t let having an opinion and knowing what you like (being a critic in the classical sense) ruin your enjoyment of everything in life. Yet another cross-platform graphical process/system monitor, bottom. They travel with a knife-sharpener. Smart Tool Selection: Achieving 34-64% Token Savings with Spring AI’s Dynamic Tool Discovery People are often the largest part…
-
Even after 3 years of tuning ChatGPT to be my best dumb friend, I keep getting told to cut the shit with my rambling, e.g., recent faint praise from the robot for my attempt at re-stating an answer: “You’ve basically got it; the only thing to do now is sharpen the tool so it’s usable…
-
“your general intolerance for corporate euphemism.” // After almost three years of using ChatGPT, I feel like it really gets me. Related, here is how one of my AI friends described the persona I’ve asked it take on: “A well-maintained chainsaw in the shed, but a polite conversation on the patio.”
-
Nudging the AI
It’s a delight to find weirdly human and overly chatty prompts like this one on how to make PDFs: PDFs – Always use LibreOffice to create the PDF (it must be LibreOffice! If LibreOffice is not installed, you can install it yourself). Other libraries sometimes show weird artifacts on some computers I wonder if the…
-
My 2025 system prompt –
-
Is this your brain on ChatGPT? –
-
Edisum: Summarizing and Explaining Wikipedia Edits at Scale – A prompt for summarizing edits made to text.
-
MCP Is RSS for AI: More Use Cases for Model Context Protocol – I like this: “I’m inclined to think of MCP as RSS for AI.” // Once prompts are used as intended – to be user initiated activity in the UI, we’ll see if things evolve. Think of an MCP prompt as adding a…
-
It’s time we stopped asking for vases. – Most people cram their AI prompts with so many rules that they predetermine the answer. When the question is stuffed with constraints, the model can only spit back something obvious. A handful of well-chosen guidelines can help, but the real magic happens when you leave space for…
-
Diane, I wrote a lecture by talking about it (Interconnected) – Prompt for converting voice memo notes to a talk, along with out-of-band technique. // I’ve struggled over the years to make this kind of thing work for me.
-
3 prompts for better knowledge worker docs –
-
o1 isn’t a chat model (and that’s the point) – “o1 will just take lazy questions at face value and doesn’t try to pull the context from you. Instead, you need to push as much context as you can into o1.” // Instead, give it briefs, memos, reports to start with. As always: context.
You must be logged in to post a comment.