Intent’s core argument is one we agree with completely: a spec that only exists for the length of one coding session isn’t really a spec, it’s a prompt with extra steps. Intent positions itself directly against AWS Kiro’s session-scoped model, making the case that a spec should be a shared artifact both humans and agents read from and write back to, persisting across the life of the project rather than resetting every time someone opens a new session.
That’s the right instinct. It’s also, philosophically, the same instinct behind Argus — we just apply it one layer earlier in the process.
Where Intent’s living spec lives
Intent’s version of “living” is scoped to the engineering artifact: the spec that describes what the code should do, kept current as the code evolves, readable by both the team and the coding agents building against it. That’s a real and useful thing to fix — a huge amount of AI-assisted engineering still runs on specs nobody bothers to update after the first sprint.
Where Argus’s living context lives
Argus starts a step earlier, with the product decision the spec is supposed to represent. Before there’s a spec to keep alive, there’s a question: why are we building this at all, and what evidence says it’s the right call? Argus keeps that reasoning — the customer calls, the support tickets, the usage data, the codebase constraints — connected to the decision permanently, and reshapes it for whichever audience needs to see it: engineering gets implementation detail, Sales gets what changed and why, Support gets what’s now supported.
The two problems compound, they don’t compete
A team that adopts something like Intent’s philosophy on the engineering side and Argus’s approach on the product side isn’t running two competing systems — they’re closing the same kind of gap from both ends. The spec Intent keeps alive is more useful when the product decision feeding it is itself alive, grounded, and traceable, instead of a one-time PRD someone wrote three sprints ago and never revisited.
The takeaway
If your engineering team has already adopted a living-spec philosophy, that’s a good sign, not a competing purchase — it means they’re already primed for the mental model Argus is built around. The question worth asking isn’t “spec or decision layer,” it’s whether the reasoning behind what you’re building survives past the meeting where it was decided, on both sides of the handoff.