v0.4.4
multica-ai/multicav0.4.4Jul 17, 2026by github-actions[bot]
AI Summary
This release focuses on stability improvements, UI/UX enhancements for the editor and settings, and fixes for agents and Codex tasks. It includes a redesign of the autopilot schedule editor and a new Mermaid viewer.
Key Highlights
- Redesigned autopilot schedule editor
- New Mermaid viewer with pan/zoom and export capabilities
- Persistent daemon flags in config.json
- Harden Cursor terminal failures
- Batch issue reconciliation
New Features
- Persist daemon flags in config.json
- New Mermaid viewer for diagrams
- Batch issue reconciliation
- Move create-issue field config into a Settings Issue tab
- Improve create issue field controls
Full Release Notes
## Changelog * 910671b1853f8cea95c8c5ffdb4889701ab1dbfa Fix Codex task MULTICA_TOKEN passthrough (ZIC-82) * ce15300493b2d5c0f597ccc2697d980ed146d2dd MUL-4884: feat(issues): anchor the working chip on agents + real colour tiers (#5540) * ea8ccf31237b15c6cceeed346f5262e5dfbb85bd MUL-4903 fix(agent): harden Cursor terminal failures (#5559) * 8ca30e794e7d81cfcffa02a7325709939a9bdca8 docs(changelog): v0.4.4 (2026-07-17) release entry (#5577) * 465546b83baa3aaeeac791b51c1e9bd560d2f4df feat(autopilots): redesign the autopilot schedule editor (#5457) * e13eb6c21637109ad8c81ccecf43a84777986fcd feat(cli): persist daemon flags in config.json (#3824) * 17ee59ccc5b5770df62556feeeaa2ec4bf49c085 feat(editor): standard Mermaid viewer with pan/zoom, exports and a real dialog (MUL-4908) (#5564) * 18d41151eb16e697517c038eb7e3318a99bcd254 feat(gc): batch issue reconciliation (#5534) * 82906bd9e9d56752daa19239cae09d841d3d23fc feat(settings): move create-issue field config into a Settings Issue tab (#5556) * ed9adc2bbe408da386a00d96ce5424a2cd6742fa feat: improve create issue field controls (#5532) * 1507997272a133f2dbedbfda7df4a38f3eba5bf4 fix(agent): stop agents shipping local-path links, make Desktop 404 recoverable (MUL-4899) (#5557) * 28969a84938144f9902f8217089ad91d066bf1f6 fix(autopilots): clarify Chinese schedule day-pattern labels (MUL-4927) (#5576) * d1c7e1c56de78d234a715d97a584db2c90d8e1c4 fix(chat): always scroll to bottom when switching chat session (#5566) * 7d04b1d9a3e196c65db2613c89648d9c0cdf4139 fix(cli): fail fast with actionable daemon startup errors * 1e34dab67297d8fd50a337ef972a6e7dc6455f37 fix(cli): set agent type when updating autopilot (#5543) * 3547201bbe54f3f63bdd25ac02089c35a709eba9 fix(daemon): reuse managed workdirs for squad leaders (#5429) * d833ef520d092877ccf9d10ff23fd129039d519a fix(editor): preserve ordered list numbering on rich paste (#5538) * 07538e99281c089bc93e1f83837467d259e67537 fix(execenv): isolate Hermes SQLite state (#5560) * a18cc65b359a6bc066f23f8075305650b8e24623 fix(inbox): render archived rows as read (MUL-4893) (#5549) * 3ce25d16e76043c4c28e8ffc9502140efa2efd56 fix(migrate): auto-backfill attribution before migration 198 to unblock self-host upgrade (MUL-4897) (#5558) * 71141559c28562f68bbe2bfc3895ee3ce748e74c fix(shortcuts): ignore synthetic key events without a key * 6dba74c3c1b3369c4c13f2eeaa2c1398169e1c99 fix(task): auto-retry transient "Connection closed mid-response" like chat (MUL-4910) (#5565) * 6c846640891c19d3ca8c244dccd2630186b6efe6 test(agent): serialize thinking cache tests (#5551)