TaoOfDev

Coder since 11.

Building 🚀

[TaoOfDev @taoofdev]

"If you can develop technology that's simply too hard for competitors to duplicate, you don't need to rely on other defenses. Start by picking a hard problem, and then at every decision point, take the harder choice." - @paulg

[TaoOfDev @taoofdev]

I particularly agree with this point. For an agent harness I'm building I've got an Engineer agent and a QA agent who checks the Engineer's work. Thinking about it, more teams should have dedicated QAs and Testers but often don't because the ROI doesn't seem worth it. But with agents we don't have the cost issue as much so we can do things the right way!

[TaoOfDev @taoofdev]

Why is Claude Code referring to me as Daisy???

[TaoOfDev @taoofdev]

Shifted from Matomo to Umami for open-source self-hosted analytics. Far superior UX and it includes session replays out of the box. Great stuff!

https://t.co/qZHJShXW3V

[TaoOfDev @taoofdev]

14/ And you can read the Nipper skill file at https://t.co/4FO31DQLzj

[TaoOfDev @taoofdev]

1/ If your AI agent skill file is just an API reference, agents will fail on it in confident, plausible-looking ways.

The fix is decision rules, named anti-patterns, rationale, and code blocks treated as contracts.

28 patterns I learned shipping Nipper:

[TaoOfDev @taoofdev]

13/ The trajectory of any good skill file: start as API reference, end as operations manual.

Treat it as executable spec for a new hire, not schema docs for a human reader.

All 28 rules with worked examples:

https://t.co/LTAlbsLWlA

[TaoOfDev @taoofdev]

@jabsborja @thisistamhn @bcherny Fixed by disabling Posthog MCP.

[TaoOfDev @taoofdev]

@jabsborja @thisistamhn @bcherny When I expand the command it says "Prompt is too long"