v0.17.0: Manual verification workflow, Stats TUI, and Shared section viewer across TUIs
v0.17.0 is a workflow-and-UX release: a brand-new manual-verification loop, a full interactive stats TUI, section-aware navigation across three TUIs, and a ground-up rewrite of the website and landing page.
Manual verification workflow
Some things you just can’t test with a bash assertion — you have to load the TUI, click through it, and see whether the row reorders. v0.17.0 gives those checks a proper home: mark a task with issue_type: manual_verification, list the items to verify, and /aitask-pick walks you through a Pass / Fail / Skip / Defer loop for each one. Failures become linked bug-fix follow-ups automatically, deferred items carry over into a fresh task on archival, and the archival gate makes sure you can’t forget half-finished verification runs.
Stats TUI
ait stats --plot is gone. In its place, ait stats-tui (or just t from the TUI switcher) gives you twelve live stats panes across Overview, Labels, Agents, and Velocity categories — counters, charts, heatmaps, and ranked tables of your most-run operations. An inline layout picker lets you swap between presets or build your own, and layout choices persist to your user-level config without ever touching the shared project config.
Shared section viewer across TUIs
The structured section markers that shipped in v0.16.1 for brainstorming are now a first-class navigation tool everywhere. Open a plan in the codebrowser, the Brainstorm node-detail modal, or the board’s task-detail screen, and you get a minimap of sections you can click to jump to — or press V to pop the whole thing fullscreen with keyboard navigation. Long plans stop being a scroll-wall.
Docs and website overhaul
The landing page, the overview, the README, and a new 12-page Concepts section have all been rewritten around a single framing: aitasks is an agentic IDE that lives in your terminal. On top of that, a systemic consistency sweep caught drift in the TUIs, Skills, Workflows, Concepts, and Commands sections, and every docs page now carries maturity and depth badges so you can see at a glance whether a feature is experimental or stable and whether a page is main-concept, intermediate, or advanced.
Full changelog: v0.17.0 on GitHub