Arthur's Blog

Claude Code - The Missing Manual

Sat, 10 Jan 2026 00:00:00 +0000

A guide on getting productive with Claude Code, Anthropic’s agentic coding tool. This is an update to my original post from mid-2025, incorporating newer patterns around plugins, issue tracking, and skills.

Installation

You’ll need an either an API key or a Pro/Max subscription. On first start, claude will ask you which you want - I recommend the latter to avoid unmanaged costs. To use a subscription, be logged in to https://claude.ai in a browser and follow the instructions.

Synthetic Publics Will Fail Confidently

Fri, 02 Jan 2026 00:00:00 +0000

By chance I read two articles at a similar time that join up nicely.

First an academic paper The potential existential threat of large language models to online survey research, by Sean Westwood.¹

Westwood demonstrates that autonomous AI agents can complete surveys with a 99.8% pass rate on attention checks, logic puzzles, and other standard bot detection methods. His bot maintains a consistent persona, can remembers its prior answers and calibrates its vocabulary and spelling errors to match its assigned education level. It strategically refuses superhuman tasks to avoid revealing itself.² and only a small number of synthetic respondents can be sufficient to flip a poll’s predicted winner.

Context Engineering for Claude Code

Thu, 27 Nov 2025 00:00:00 +0000

On Tuesday I did a talk to our internal AI group, along with talks by excellent colleagues, on Context Engineering. The very next day, Anthropic released a blog post on Effective harnesses for long-running agents and updated their prompting best practices guide.

Note to Anthropic: please can you release relevant material the day before I do a talk rather than afterwards!

I talked about managing context size, how LLM accuracy declines as the context window fills past 50% and my mental model of LLMs as amnesiac pedants, and how this drives Claude Code features like it’s heavy use of todo lists, summarising context into markdown files etc.

Creating Claude Code Plugins

Sun, 12 Oct 2025 00:00:00 +0000

On Friday Simon Willison published a look at Jesse Vincent’s¹ Superpowers plugin for Claude

You can read Simon’s and Jesse’s posts to see what this super cool plugin does, but I wanted to understand how to create a plugin and this is quick summary of my notes.

First up, Jesse uses a complex structure with three repos:

It’s setup this way to allow for contributions, but if you just want to play with creating a plugin, it’s not needed.

Shorts, June 2026

Thu, 26 Jun 2025 00:00:00 +0000

Trying out a new format this week. Instead of links, here’s some shorts. Too little for a post, too much for a link. This week we have

Power variation in AI data centres

(Link requires registration and full article isn’t free, but there’s a massive preview which is all I have access to)

A new to me issue with AI workloads - training runs are very, very spiky in their power use.

Getting started with the Anthropic Cookbook and Marimo

Tue, 10 Jun 2025 00:00:00 +0000

The Anthropic Cookbook is an excellent resource for learning how to write LLM code and it’s continually being updated, most recently with example code for agents.

The Cookbook comes as a set of Jupyter notebooks. I prefer Marimo to Jupyter, mainly because it supports reactive cell reloading i.e. if you change an earlier notebook cell that a later cell depends on, the chain of dependencies is calculated and everything is re-run correctly which saves a lot of mental energy and stupid mistakes when working with notebooks.

HEPI Report on AI in Higher Education

Wed, 04 Jun 2025 00:00:00 +0000

HEPI, a UK higher ed thinktank, released the results from their 2025 survey of AI use by students. There’s a lot in the report, so here’s a few quick highlights.

The big headline is that basically all students are using genAI:

In 2025, we find that the student use of AI has surged in the last year, with almost all students (92%) now using AI in some form, up from 66% in 2024, and some 88% having used GenAI for assessments, up from 53% in 2024. The main uses of GenAI are explaining concepts, summarising articles and suggesting research ideas, but a significant number of students – 18% – have included AI-generated text directly in their work.

Links, 4th June 2025

Wed, 04 Jun 2025 00:00:00 +0000

This week we have drones in Ukraine and SpaceX’s totally-not-a-cock rocket.

The use of drones in Ukraine is advancing the state of the art very, very rapidly. First we had drones trailing fibre optic cables to avoid jamming, now there’s drones that navigate via camera and landmarks and, very shortly, autonomous kill decisions: IEEE Spectrum - Ukraine’s Killer Drones
A very hot take on Starship, SpaceX’s big “go to Mars” rocket: Starship was doomed from the start. I’m not informed enough to assess the post’s validity, but the thesis is that Starship is just too heavy, and that making it lighter makes it too fragile in ways that just can’t be fixed. There’s updates as well - 1 and 2

Another look at per token energy costs

Thu, 29 May 2025 00:00:00 +0000

Piotr Mazurek and Felix Gabriel have an amazing post up on LLM Inference Economics from First Principles, which I found on Bluesky.

They go into a huge amount of detail about how inference works and how that affects processing speed. But I saw the graph above and thought “we can get energy from that”. And so I asked chatgpt-o3: “Looking at these graphs of throughput at different batch sizes on a 4xH100 80gb cluster, what ranges of power per token do they equate to?”

Claude Code - The Missing Manual

Tue, 27 May 2025 00:00:00 +0000

Note: This post has been replaced by Claude Code - The Missing Manual, which includes updated patterns around plugins, issue tracking, and skills.

Notes that I wrote for work on how to get started with Claude Code, Anthropic’s agentic code tool. Anthropic’s own documentation is good, but I wanted a single page that I could share with people to help them get rapidly up to speed.

Once past this stage, their Claude code best practices page is excellent.

Links, 16 May 2025

Sat, 10 May 2025 00:00:00 +0000

A very late set of weekly links! This time it’s heavy on AI and it’s impact on research and teaching. Plus cyborg cicadas.

First up: A NYMag article on LLM use at Universities went viral (at least in my education and AI heavy feeds!) last week. So that’s not the link as you’ve probably already read it. Instead, here’s Benjamin Breen on how AI makes the Humanities more important and also wierder and Will the Humanities Survive Artificial Intelligence? from the New Yorker.

Tokens, tokens everywhere

Sat, 10 May 2025 00:00:00 +0000

We have some actual numbers on tokens! In their commentary on the latest earnings report, Microsoft report:

We processed over 100 trillion tokens this quarter, up 5X year-over-year – including a record 50 trillion tokens last month alone

So we’ve got a baseline, and can estimate a growth rate. Assuming exponential growth and a three month token split of 19/31/50 we get 62% month on month growth, or, if it keeps going 100 T → 424 T → 1.8 P → 7.6 P, which is lower than the 5x yoy growth MS quote.

Environmental Impact of AI

Tue, 06 May 2025 00:00:00 +0000

As Andy Massick explains well, if you are making a few requests per day to a simple non-reasoning text based AI model the carbon impact is, to use a technical term, bugger all ¹.

I probably ask ChatGPT and Claude around 8 questions a day on average. Over the course of a year of using them, this uses up the same energy as running a single space heater in my room for 2 hours in total. Not 2 hours per day, just a one-off use of a single space heater for 2 hours […] By being vegan, I have as much climate impact as not prompting ChatGPT 400,000 times each year

Links, 30 April 2025

Wed, 30 Apr 2025 00:00:00 +0000

Finally a week with more than two links!

Beatrice de Graaf is a Professor at Utrecht University. Her talk at Stanford on The Quest for a Third Rome describes using text mining techniques to analyse Putin’s speeches since 1999. The paper, written with her student Neils Drost, is open access: Putin and the Third Rome:: Imperial-Eschatological Motives as a Usable Past

“In this corpus, we ask how Putin used the combination of history and Russian Orthodox theological, biblical tropes as a usable past, to create his powerbase, to understand how he built up to defending and legitimizing the invasion in Ukraine”

A similar area was covered by Teitelbaum in his book War for Eternity about the influence of Alexsandr Dugin and mysticism on both Putin and Steve Bannon in the US.

Does Putin believe it? Maybe. Is he using mysticism and religion for his own ends? For sure.

Links, 22 April 2025

Tue, 22 Apr 2025 00:00:00 +0000

Some AI development resources this week…

How to Build an AI Code Agent

Thorsten Ball has published how to build your own AI code agent. He creates a toy example in Rust that’s <400 lines but still includes tool use. Geoffrey Huntley does the same. It’s worth reading both

Claude Code Best Practices

Anthropic has released a detailed guide on Claude Code best practices.

Of note:

Claude can (and should!) use the architect/plan then code model that Aider makes explicit. Telling Claude to read code/docs, then plan, then code, then commit gives better results than just telling it to write code
Ask Claude to review things (e.g. the CLAUDE.md hint files) and suggest improvements. Repeat. (As an aside, the Anthropic workbench can improve general prompts)
Telling Claude to ’think’, ’think harder’ etc results in more thinking tokens being used
Clear context frequently (/clear) - usable context window is smaller than the advertised max

The model of asking Claude to review and improve its work (or human starting points) repeatedly was not obvious to me

Links, 15 April 2025

Tue, 15 Apr 2025 00:00:00 +0000

Two links this week…

AI Logos Are All Buttholes

In what might be the most important visual design analysis of 2025, Velvet Shark has compiled a gallery of AI company logos that look like buttholes. From Claude’s spiral to Midjourney’s abstract void, it seems AI branding has converged on a rather unfortunate design pattern.

If It’s Crypto, It’s Not Money Laundering

Money laundering is legal now, as long as it’s crypto. JP Koning summaries the Trump administration’s latest legal changes: If it’s crypto, it’s not money laundering.

Links, 11 April 2025

Fri, 11 Apr 2025 00:00:00 +0000

I’m trying not to be perfectionist about this blogging thing, and concentrate on getting stuff out so in that spirit so this week there’s just a few links, all on AI.

First up - how to use AI.

I see so many people type some random question into ChatGPT/Claude, get a mediocre answer back and move on. But how to get good results from these systems is not at all obvious. For a complex example, see my post on Experimenting with Deep Research or an even more extensive example of using one model to generate a prompt for another

Connecting Claude Desktop to Zapier

Thu, 03 Apr 2025 00:00:00 +0000

Today’s fun is setting up the new Zapier MCP integration. This creates a MCP server that allows Claude Desktop or Cursor or anything else that supports MCP to call any Zaps

Setting Up Zapier MCP

Step 1: Install mcp-remote

First, install the mcp-remote package using npm:

npm install mcp-remote

Step 2: Configure MCP Endpoint in Claude Desktop

In Claude Desktop, you’ll need to add the Zapier MCP endpoint to your settings. Add the following configuration to ~/Library/Application Support/Claude/claude_desktop_config.json

Links, 31 March 2025

Mon, 31 Mar 2025 00:00:00 +0000

Trying something new, with a weekly-ish list of links

1) OAPs are radicalised too

When Grandpa Turns Extremist: Digital Immigrants, Radicalisation and the German Reich Citizens Movement

the radicalisation of middle-aged and elderly people remains largely overlooked despite their active involvement in nearly all extremist movements. This Insight advocates for a more nuanced understanding of extremism and ageing, emphasising the need for radicalisation prevention strategies geared towards those born as Digital Immigrants, such as Baby Boomers and Generation X. These generations did not grow up in a digitised world, and their education largely did not include critical media studies. Among others, this factor makes them prone to misinformation consumption online. [This article looks..] at the example of the German Reich Citizens Movement, which has a very specific age structure and is characterised by the relatively late radicalisation of its adherents. The movement, which predates the digital age, now primarily relies on the internet for communication, spreading its ideology, and mobilising action against the state and governmental authorities

Model Context Protocol 101

Tue, 04 Mar 2025 10:00:00 +0000

Model Context Protocol (MCP) is a standardized approach for structuring context information when working with Large Language Models (LLMs). It was introduced by Anthropic but is standardised and an open protocol.

Yeah, great. But what does that even mean?

The problem that MCP is trying to solve is one of getting access to local/specialist information. When using Claude or ChatGPT etc, the model has access to its built in knowledge from training, but this only contains public information and will have a cut off date when the model was trained. To get around this, some tools have integrated web search that can add extra knowledge and/or the ability to import documents. But what if you want to talk to a database? Or files on disk? Or a knowledge base? Or …

Experimenting with OpenAI's Deep Research

Thu, 27 Feb 2025 00:00:00 +0000

Now Deep Research is available to ChatGPT users on the £20/mth tier, I’ve had a chance to experiment. The tl;dr is that it’s really very good, and the level of progress in just 6 months is mind-blowing.

I picked something I know about - financial pressures facing UK Universities. First, I asked ChatGPT to generate the prompt for Deep Research, using this template I found via a mailing list (and I can’t remember which one).

About

Mon, 26 Feb 2024 00:00:00 +0000

Welcome to my blog! I’m Arthur

This blog was started in 2000ish when the domain was registered, but over the years it’s moved around and been restarted many times, so there’s no back catalogue

Feel free to contact me on Bluesky or LinkedIn