Open the Delphi RTL source and you will find this pattern in many places: if Assigned(X) then FreeAndNil(X); FreeAndNil already copes with nil, so the “if” looks redundant. But there is a classic counter-argument: when X is mostly nil, the explicit test is faster, because it skips a procedure call. Opinions are cheap, so I […]
if Assigned(X) then FreeAndNil(X): is the check faster? I measured it Read More »
Anthropic’s prompting guide for Opus 5 says: “Claude Opus 5 has a 1M token context window as both the default and the maximum, and its instruction following, tool calling, and reasoning stay consistent throughout the window.” No number backs the second half. So I checked it. Context Arena, an independent site, runs a test called
Opus 5 and the 1M Promise: does Claude stay good over its whole context window? Read More »
This article is one chapter of my book Delphi in all its glory – For more than mere programmers, the first book of the series. Want to see what the compiler really does with a property? Ask for its address. Below, Obj.Prop is a read-only property that either reads the field FProp directly or calls
The address of a property: what the Delphi compiler does with @, var and FreeAndNil Read More »
This article is one chapter of my book Delphi in all its glory – AI-assisted development, the fifth book of the series. Claude Code has a voice dictation feature that lets you speak your prompts instead of typing them. You hold down a key, talk into your microphone, and your speech appears in the input
Voice Dictation in Claude Code: /voice on Windows Read More »
This article is one chapter of my book Delphi in all its glory – AI-assisted development, the fifth book of the series. Claude Code lets us choose between different models (Opus, Sonnet, Haiku) and effort levels (low, medium, high, xhigh, max). The core idea is simple: not every task deserves the same horsepower. Renaming a
Switching Models and Effort on the Fly: Claude Code’s /model and /effort Read More »
This article is one chapter of my book Delphi in all its glory – AI-assisted development, the fifth book of the series. In the newer versions of Claude, a tip that pops up in the terminal says: “Run tasks in the cloud while you keep coding locally.” At first glance, this sounds like marketing fluff.
Running Tasks in the Cloud: Claude Code on the web for Delphi Read More »
This article is one chapter of my book Delphi in all its glory – AI-assisted development, the fifth book of the series. According to Anthropic (therefore, take these numbers as a direction, not as a promise), Claude Opus 5.5 works at the level of Fable 5.1 on most jobs, and costs 40% less to run
Opus 5.5: cheaper than Opus 5, and on a cybersecurity leash Read More »
This article is one chapter of my book Delphi in all its glory – AI-assisted development, the fifth book of the series. Claude has amnesia. Not the poetic kind. The official documentation says it plainly: “Each Claude Code session begins with a fresh context window.” You spent four hours with it yesterday. You explained why
HandOver.md – The File That Outlives The Session Read More »
This article is one chapter of my book Delphi in all its glory – AI-assisted development, the fifth book of the series. Anthropic’s figure above, $150-250 per month, is the average across enterprise deployments. A heavy user is a different story. Some Delphi users investigated what their subscription would cost at API list prices. One
This article is one chapter of my book Delphi in all its glory – AI-assisted development, the fifth book of the series. In Windows, Ctrl+C means Copy. In a terminal, it means Stop. In Claude Code, it means whichever one you did not want. I press Ctrl+C by reflex. In Claude Code, one stray press
Ctrl+C: Copy or Kill? – Stop it from cancelling Claude Code Read More »
This article is one chapter of my book Delphi in all its glory – AI-assisted development, the fifth book of the series. Claude is verbose. Ask it to fix a one-line bug and you get three polite paragraphs before the fix arrives. Every one of those words is an output token, and output tokens cost
This article is one chapter of my book Delphi in all its glory – AI-assisted development, the fifth book of the series. Skills and sub-agents are both ways to extend Claude’s behavior, and they’re easy to confuse at first. They look similar on the surface but work very differently under the hood. A skill is
Skills vs Agents: when to use each in Claude Code Read More »
This article is one chapter of my book Delphi in all its glory – AI-assisted development, the fifth book of the series. Skills are addictive. Each one solves a real annoyance, each one takes ten minutes to write, and none of them feels expensive. Then one day you run /doctor and it tells you that
What 47 Skills Actually Cost You: Claude Code’s skill budget Read More »
This article is one chapter of my book Delphi in all its glory – AI-assisted development, the fifth book of the series. Somewhere around Opus 5 a new complaint started showing up everywhere: the model got smarter and the answers got harder to use. Not wrong. Harder to use. You ask a small question and
Everything, from voidtools, finds any file on my disks while I am still typing its name. Claude Code has its own file finder, Glob, and next to Everything it feels slow. So I told Claude to switch. It took one small download, one line in CLAUDE.md, and learning a few traps that return nothing without
Teaching Claude Code to find files with Everything Read More »
I have madExcept wired into everything I ship. It catches the crash, builds a report with the call stack, and mails it to me. It works beautifully, and that is exactly what makes the hole in it so easy to miss. madExcept only ever sees an exception that nobody caught first. So this: try …
Your try..except is hiding your bugs from madExcept Read More »
Opus 5 will not be shortened by lowering the effort level. A Claude Code skill that re-says the last answer in plain words, without re-deriving it and without dropping a single caveat.
Opus 5 talks too much. So I made it say it again, in plain English Read More »
Anthropic publishes three “agentic coding by effort level” charts for Opus 5 but prints no numbers on them. I measured the dots. On two of the three, max effort scores lower than xhigh while costing more.
Claude Opus 5: xhigh or max effort? The charts say xhigh Read More »
I ran my AI code-review pipeline twice on the same Delphi unit: once with Claude Opus 4.8 agents, once with Claude Fable 5 agents. Same byte-for-byte prompts, same file state (restored between runs), models verified from the agents’ own transcripts. The result was not what the hype predicts. The setup The target: frmTestMain.pas (482 lines)
Opus 4.8 vs Fable 5 – which one reviews Delphi code better? Read More »
Finally, a tiny bit of transparency Since April 2026, Google Maps shows German users something the rest of Europe doesn’t get: a counter on every business profile telling you how many reviews were deleted in the last 12 months after a defamation complaint. Click into the Reviews tab and the number is right there. It’s
Google Maps now shows how many reviews German businesses got deleted Read More »
Anthropic’s Opus 4.7 ships with better benchmarks but persistent hallucinations. Here’s what’s real, what’s hype, and how to roll back to 4.6.
Shipped alongside Claude Opus 4.7, /ultrareview is the heavyweight cousin of the /review command we already covered. Where /review is a quick sanity check you run between keystrokes, /ultrareview is what you fire up before pushing that big refactor to the main branch – the one that touched half your DPR, three packages, and the
On April 16, 2026, Anthropic dropped Claude Opus 4.7. Same day this chapter was written, which means by the time you read it some of the rough edges may have been sanded off and a few new ones added. That is the nature of writing a book about a moving target. I am going to
If you have been building FireMonkey applications for any length of time, you have probably noticed that certain parts of the framework feel like they were never quite finished. Menus are one of those parts. After spending too long debugging a strange visual glitch in our app, I decided to investigate properly — and what
Why I Stopped Using TPopupMenu in Delphi FMX (And What I Use Instead) Read More »
I just published a new GitHub repository: Awesome-AI-For-Delphi. It is a curated list of tools, libraries, MCP servers, and resources for using AI with Delphi. Think of it as a single place to discover everything that exists at the intersection of AI and Pascal/Delphi development. The list covers: MCP Servers — let Claude Code and
When you signed up for the Max 5x plan at $100 (€90+VAT), you didn’t just buy intelligence — you bought a static expectation of capacity. In the early days, “usage limits” felt like a solid wall: you knew where they were, and when you hit them, the rules of engagement were clear. Then Anthropic shifted
The 5-Hour Mirage – Anthropic’s Diabolical “Moving Goalposts” Subscription Read More »
I’ve been using this trick for months and honestly thought everyone else was doing it too. But after mentioning it to a friend who looked at me like I’d just discovered fire, I realized: it’s time to share the secret. The Problem: You sit down to code at 10:00 AM, burn through your messages by
Incredibly cheap trick to “shorten” your Claude 5-hour window Read More »
“AI-Assisted Development for Delphi” is the fifth book in the “Delphi in All Its Glory” series. Released April 2026. Why This Book Exists There are a gazillion books about AI out there. This is the only one written for Delphi developers. Your LinkedIn feed says programmers will be obsolete by next Thursday. Meanwhile, you’re
The “Delphi in all its glory” book [5] – AI-Assisted Development for Delphi Read More »
Smart CodeInsight stopped appearing in my Delphi IDE after an upgrade. Here is the fix. Uncheck the “Enable Smart CodeInsight” checkbox in Options, close Options, then reopen Options and check the box again. A EULA prompt should now appear (Embarcadero Supplemental AI Terms). Accept it, and Smart CodeInsight will work again. Apparently, during the upgrade
The Controversy In early 2026, two independent research teams published the first empirical studies on whether CLAUDE.md-style instruction files actually improve AI coding performance. Their headlines seemed to tell different stories — one modest, one dramatic. Tech blogs promptly cherry-picked the modest one and ran with it. An article on XDA Developers declared that CLAUDE.md
The Science Behind CLAUDE.md: What the Research Says Read More »
