Linear published a bold claim: issue tracking is dead. Their argument is that the traditional handoff model, where PMs scope work and engineers pick it up later, created process overhead that "became the work." Agents compress the cycle, so the old system is obsolete.
I think they're half right. Issue tracking built for handoffs deserves to die. Issue tracking built for communication is more essential than ever.
The need for communication will never die. Right now, it's growing at a terrifying pace.
We built Cate to close that gap. Cate keeps you, your team, your agents, and even your teammates' agents on the same page, using the tools you already know: Jira, Linear, and GitHub.
We Can Do More. Can We Explain More?
We're undertaking work at a scale that would have been absurd two years ago. A single developer with a handful of agents can move a codebase forward at a pace that used to require a team. That's incredible. It's also terrifying, because our ability to communicate what we're doing hasn't kept up.
Right now, on teams across the industry, developers are running five, ten, twenty agents from their laptops, and nobody else on the team can see what those agents are working on, why, or what decisions are being made. A Slack thread isn't a plan. Someone's mental model of what their agents are up to isn't coordination. A PR with no context beyond "fixes #247" isn't documentation. The work is getting done. The communication isn't.
I was already bad at this when I was writing every line by hand. Now I'm moving ten times faster with the same communication habits. Scale that across a team, and it's not just my problem anymore. It's a management problem, a planning problem, and eventually a quality problem.
The problem was never that we tracked too many issues. It was that we tracked them poorly: status fields updated by rote, descriptions copied from a meeting that half the team missed, acceptance criteria that nobody revisited after the first draft. The answer isn't to stop tracking. It's to finally do it well.
Communication Is the Whole Game
Here's what I think Linear gets exactly right: agents need rich context. Customer feedback, strategy, decisions, code. Where I disagree is the implication that this context lives in some new abstraction that replaces the issue. The issue is the unit of communication. It's how we tell each other, and our future selves, what we intended, why we intended it, and what actually happened.
The real problem with traditional issue tracking is that writing good issues was always manual, always expensive, and always the first thing to get cut when the sprint got tight. I knew the business context. I knew why the work mattered. I just didn't write it down because I was too busy doing the work itself.
That's where AI changes the equation. Not by removing the need for communication, but by making it cheap enough that we actually do it.
Documentation at the Speed of Change
Here's the untapped power: when given concrete changes, like code, and a business context, AI can relate and document change at the same speed it makes it. That's not a small thing. That's the bottleneck that's been choking every engineering team since the first wiki page went stale.
We built Cate around this idea. At its heart, Cate is a communication tool. Agents research a problem and produce a spec. That spec becomes real issues on your board, in Jira, GitHub, or Linear, visible to your whole team, written with the kind of architectural context and acceptance criteria that I always meant to include but rarely had time for. After implementation, agents capture what they decided, what tradeoffs they made, and what surprised them, and that context lives on the PR and in shared research notes.
This isn't a new workflow. It's your existing workflow, done at both a scale and a level of detail that wasn't possible when every word had to be typed by a human who was also trying to ship the feature. The standup still works. The board still works. The PR review still works. They just carry real information now.
The Obituary Is Premature
The teams that thrive won't be the ones that abandon structure. They'll be the ones whose structure finally earns its keep, because AI made the cost of good communication low enough to actually pay it.
We don't need fewer issues. We need better ones. And for the first time, we have the tools to write them.
Let's add "just enough" planning to your AI workflow. Request early access to Cate, or get in touch. We'd love to hear how your team works.