Blog
Notes on deterministic software delivery, policy-as-code, and what building aSPARK has taught us.
Campaign or story? Where aSPARK draws the line
Since v0.13.0 aSPARK has two ways of working: the feature loop for stories, and campaigns for work whose end a command can confirm. When each one fits, what isn't a campaign, and how the agent checks that itself.
Read more →aSPARK reads Flask: 21 of 21 sources check out, and one lead didn't
We ran /charter's discovery pass and /next-steps on a fresh clone of Flask. Every file:line it cited resolves, the graph builds in seconds, and the Product Owner proposed a gap the docs really have. One side note looked like a bug and wasn't.
Read more →What a full loop costs: 42 million tokens, $35 — and 96% of it cache
Seven role agents across five phases: one measured run, broken down per role, with the caveats that belong to it.
Read more →Your first loop: start with /charter, not with the feature
Install, project profile, first feature: what to do in your first hour with aSPARK — and which decision at each gate is genuinely yours.
Read more →Write the idea, not the ticket: keeping the Specify gate short
The Product Owner asks about whatever you left open. Anticipate the gaps and you get to a better spec faster — here is the pattern that keeps working.
Read more →Why determinism is the more interesting property than speed
AI assistants make generation cheap. The bottleneck moves to where it was already tight — and determinism is what reopens it.
Read more →