Could your next Claude Code skill be a diagram? Not a diagram just describing the skill — a diagram that is the skill, executed like code. The answer is YES. Once you've tried it, you may not want to go back. antonymarcano.substack.com/p/claude-cod... #claudecode #ai #aiaugmented #claudeskills

Your Coding Agents Just Broke – Yet Nothing in Your Repo Changed?! This has just made another compelling case for a test-driven approach to your CLAUDE/SKILL.md changes when using coding agents at scale... Read on → open.substack.com/pub/antonyma... #aicoding #aiaugmented #claudecode

Here’s the latest on the evolution of Stagentic-AI’s genesis project… Enabling Test-Driven Development of Claude Skills… open.substack.com/pub/antonyma... Now, every test-step is carried out by a separate sub-agent. Given – Haiku When – Opus (agent under test) Then – Sonnet Finally – Haiku Enjoy…

Just for fun, I pointed NotebookLM's AI Podcast Hosts at my latest article "Taming Claude Code, One Agentic-Test at a Time"... It's metaphors, and insights – especially at the end – were a genuine surprise! Listen here (article linked in shownotes) open.substack.com/pub/antonyma...

I was stuck in a “prompt and pray” loop, changing CLAUDE-md and SKILL-md files. Fix one behaviour, break another without even knowing. Until I realised this new problem… has an old solution. "Taming Claude Code,—One Agentic Test at a Time" open.substack.com/pub/antonyma... #claudecode #vibecoding

Better late than never... antonymarcano.substack.com

When people say "guardrails", they really mean "guidance"... But there are true guardrails available. Combine both with gateways and not only do you have greater control over your coding agent, you improve token efficiency... → ideas.riverglide.com/what-people-... #AIcoding #agenticcoding

What if you could develop your SKILL . md files in a test-driven way, just like your code? That was the question I asked myself, and the result was Test-Driven Agentic Behaviours (TDAB)... Let me show you → youtu.be/i3g9ZXiaqh0 (Re-uploaded) #ai #aiaugmentedcoding #aiassistedcoding #claudecode

What if you could develop your SKILL . md files in a test-driven way, just like your code? That was the question I asked myself, and the result was Test-Driven Agentic Behaviours (TDAB)... Let me show you → youtu.be/uisjVdEvrCY #ai #aiaugmentedcoding #aiassistedcoding #claudecode

If the collective comprehension and articulation of a concept is flawed, and our individual comprehension is flawed, then we cannot expect AI to be any less flawed than the material it has to work with. This seems to be at the heart of much of the debate in AI augmented coding.

Mutation testing is a method that's becoming increasingly important with AI-augmented coding. Without it, we could risk future bugs in our agent-written code, even if what we release today seems to work. Read on: ideas.riverglide.com/mutants-in-t...

Thanks to everyone that joined us for @xtclondon.bsky.social on Tuesday! Loved all the discussion sessions on enjoying programming, effectiveness of AI, security and "best vibe engineering practices" 💪 Looking forward to future meetups: www.meetup.com/extreme-tues...

I recently joined Jack Hannah for a chat on @tuple's "Distributed" podcast. We talk about what distinguishes teams at the 'Elite' benchmark of DORA's Software Delivery Performance metrics, how AI is reshaping software delivery and more. Watch here: youtu.be/y7KZd6x7-lo (55mins)

Is Your AI-assisted Coding Strategy Quietly Backfiring In 2026? Most AI-assisted coding strategies lean on one simple assumption: If developers can write code faster, delivery will be faster too. The latest DORA research suggests that assumption is wrong, more often than not. youtu.be/dQ1apxgvOa0

"Product managers were frustrated by how long even small features took to release". By making it possible to slice their backlog differently an additional £2.6M in revenue was found, representing "repeated revenue they were likely missing out on for some time". ideas.riverglide.com/your-product...

I need to be more active here 🤔

Great catching up with @antonymarcano.bsky.social at The Rake in Borough Market tonight. Great conversation and some spiffy guest ales. Opinion was divided on the lemony one.

Think about it...

Why the story of the viral skateboard-to-car, agile in a nutshell, image needs an unhappy ending... You’ve probably seen Henrik Kniberg's skateboard-to-car sketch a thousand times. But, not with this version of the story behind it... youtu.be/UR4Cq42BtbU #agile #agilesoftwaredevelopment

I've been living over on LinkedIn all this time and forgot that BlueSky was a thing... Going to see what it's like here... And if anyone is listening.

It feels odd to me when people say they code solo by default, and only pair when it makes sense. My default is to pair, and only code solo when that makes sense... which is generally exploring, when it's about discovery of new ways rather than executing on one of them. But only for short periods.

Hello! Late to the party... But better late than never.