Ready-to-use tools built during every edition's journey.

In Edition #2, I wrote about building an AI agent that monitors my Slack channels and delivers briefings before I even open a conversation. The artifact is that agent, the working tool, the agent configuration, the AI briefing that started the build, and the process decisions documented along the way.

That’s what artifacts are. Not templates. Not code snippets. The actual tools I build during my own work, packaged so you can use them, adapt them, and learn from how they were made.

The library grows with every new edition.

What’s inside:

Working tools: implementations you can run, not just read about. Each one solves a real problem that surfaced during the newsletter’s journey.

Agent configurations: the system prompts, instructions, and settings that define how each tool thinks and behaves. The decisions that shaped the tool matter more than the tool itself.

Process context: not just what was built, but why it was built that way. The trade-offs, the dead ends, the reasoning. The part tutorials never give you.

Currently available:

🤖 Slack Context Retrieval Agent (from Edition #2) Monitors Slack channels, identifies what needs your attention, and gathers relevant context before you respond. Includes the full working tool, the AI briefing that kicked off the build, and the documented decisions from two weeks of real-world iteration.

New artifacts ship with every edition.

Who gets access:

Every edition of the newsletter is free to read.

The artifacts: working tools, agent configurations, and the decisions behind them are exclusively for paid subscribers.

If an edition made you think differently, the artifact lets you work differently.