MIRA v0.336.0
MIRA v0.336.0 — public beta. "Send a briefing now" no longer times out. No breaking changes; safe to upgrade in place.
Self-hostable single binary, AGPL-3.0-or-later, signed with the vexillon.ai release key.
Install
# Linux / macOS
curl -fsSL https://get.vexillon.ai/install.sh | sh
# Windows (PowerShell)
irm https://get.vexillon.ai/install.ps1 | iex
Fixed
- "Send a briefing now" is now asynchronous. It used to block while MIRA generated the briefing — a full model turn that often ran past a mobile client's ~30s read timeout, so the request never returned. The server now accepts the request and responds immediately, generates the briefing in the background, and delivers it on your companion channel when it's ready — just like a scheduled briefing. The button now tells you it'll arrive shortly.
For integrators
POST /api/me/briefing/send-now now returns 202 Accepted with {status:"accepted", …} immediately, instead of 200 with the delivered channel/size. The briefing arrives on your companion channel a moment later.
Verify (optional)
Every archive is signed with the vexillon.ai minisign key committed at verification/release-pubkey.minisign:
minisign -Vm mira-0.336.0-<target>.tar.gz -p verification/release-pubkey.minisign