Once the tickets are ready, the AI writes both the code and the tests. This post explains what implement, tdd, diagnosing-bugs, and prototype each rely on to tell whether the work is right, and walks through a real red-to-green test cycle using a per-seat billing example.

Once the requirements have been talked through, should you write a spec, break the work into tickets, or go ask the person who holds the answer? Using per-seat billing as the example, this post explains which kind of gap to-spec, to-tickets, triage, wayfinder, and to-questionnaire each fill.

A walkthrough of the grill-me and grill-with-docs interview flow: how to clarify requirements by asking questions in dependency order, and how to write confirmed terms and a few key decisions back into the project.

You don't need to memorize 25 skills to start using mattpocock/skills. This post covers how to install them, what setup-matt-pocock-skills leaves behind in your project, and how to use ask-matt to find your next step.

OpenWiki uses Git changes and Grounded Claims to find which repo wiki pages need updating, but an all-green source check only means the citations haven't gone stale. This post walks through the 0.5.2 source to show how far it actually goes and what adopting it costs.

Hacktron published an attack chain that started with an image-decoding bug in Discourse and reached OpenAI accounts and internal GitHub resources. This post breaks down the evidence, the layers of the fix, and the trust relationships involved, and ends with a list of questions you can use directly in operations and access reviews.

A walkthrough of Kevin Bai's framework for forward deployed engineering, covering when you need FDEs, how a shared platform absorbs custom work, and the maintenance costs you have to pay before you expand the team.

Parsing the question bank, building randomized quizzes, and using cross-session stats to find weak areas