Are project management tools still being updated with coding agents doing all the work?

The old model was mostly built around humans picking up tickets, moving them across a board, and writing updates after the fact. That still matters, but the workflow is changing.

Now a single feature might involve a human writing the intent, one agent creating a branch, another reviewing logs, another opening a PR, and someone else deciding whether the result is safe to merge. In that world, the project management tool cannot just be a static task tracker.

It needs to become the coordination layer between people, agents, code, reviews, and decisions.

For me, the interesting questions are:

How do you track what an agent actually did, not just the final ticket status?

How do you show confidence, blockers, review notes, and handoff context without creating more manual work?

Should Jira, Linear, Notion, and similar tools become more execution-aware, or should agents live closer to GitHub and the IDE?

And when multiple agents work on the same project, who or what owns prioritization, accountability, and final context?

Are you updating your project management workflow for coding agents, or keeping agents separate from your core planning system?