Drew Breunig

A couple notes on Claude’s Fable 5.1 system prompt.

Models Are Designed to Persist, Reason, and Coordinate Over the last few weeks I’ve been increasingly annoyed by the media coverage of the OpenAI’s accidental attack on Hugging Face and other similar incidents.

There’s some talk today about how agentic coders are balking at Anthropic’s pricing and adopting alternatives.

The best way to define a harness is as a “situated agent” Harrison Chase once excitedly shared an insight that agents are comprised of 4 things: a system prompt, a planning tool, a file system, and subagents.

Keep Your Agent’s Skills and Tools in Check When I wrote “How to Fix Your Context”, it was really hard to find real-world examples of the Tool Loadout pattern.

When a sector goes too fast, it loses support Without a doubt, the pace of the AI ecosystem is dizzying.

You can’t be model agnostic if you’re hand-tuning prompts Thanks to natural language interfaces, AI applications can be prototyped quickly.

When an agent can explain anything, should we keep writing docs for humans?

OpenAI winding down fine tuning is an interesting development and one to watch.

What should we do when code is cheap?

Is security spending more tokens than your attacker?

I like reading system prompts, either when they’re published as part of open-source software, exfiltrated via crafty prompting, explicitly shared, or (in the case of last week) accidentally leaked.

Yesterday we talked about how cheap code is fueling an era of idiosyncratic tooling, and previously we’ve talked about the rise of spec driven development.

Our era of sprawling, idiosyncratic tooling In 1998, Eric S.

The following is a write up of a talk I delivered at MLOps Community’s “Coding Agents” conference, on March 3rd.

The biggest buyers will want to audit and influence post-training Beneath the Anthropic and Department of War fracas, there is a legitimate & essential conversation to be had about how much control any organization has when deeply adopting an AI model they

There’s a lot of noise about how AI is changing programming these days.

If code is free, why aren’t all apps native?

This post was co-authored with Srihari Sriraman Coding agents are fascinating to study.

Handling Your Long Context Today & Designing Your Agent Tomorrow Context Rot is the Worst Context Failure “Context Rot” is a common problem agent designers must avoid and mitigate.