The v3 major marks Xeref's transition from a reactive chat tool to a proactive, autonomous agent platform. These two releases add the building blocks for agents that work without you.
v3.0 — Keyboard Shortcuts and Profile Avatar
A small but polished release: a Keyboard Shortcuts panel (⌨ icon in the right rail) lists every wired shortcut grouped by category, with Mod key auto-rendering as ⌘ on macOS. The bottom-left profile avatar now mirrors the photo from Settings → General instead of always showing initials.
v3.1 — Agents That Run Themselves
This is the substantive v3 release:
- Scheduled Agents: Set a cron schedule for any agent to run automatically in the background — no manual trigger required.
- Dynamic Subagents: Agents can spawn specialized sub-agents mid-task (e.g.,
/browser,/search) for parallelized workloads. - Agent Artifacts: View and download structured deliverables and outputs directly from agent conversation histories.
- Skills Library: Package and reuse custom instruction packs across agents, with full shareability for the Skool community.
- Browser & Dispatch Settings: Configure custom browser automation parameters, runtime permissions, and MCP tool access from a dedicated settings surface.
- Strict Permission Mode: A new toggle requires human approval before sensitive actions are executed.
- Usage Dashboard: Per-model rate limits, credit tracking, and refresh timers visible in Settings.
- Run History & Diff View: Inspect exactly what files, data, or filesystems each agent touched in a session.
v3 makes it practical to hand a task off to an agent, close the laptop, and come back to a completed result — the core promise of agentic AI.