📖About

Four improvements: (1) research shown as readable description not JSON, (2) Carol design language (gradient+glass), (3) proactive tab shows pa-c1 daily plans with 3 options + preferred choice — requires new proactive_plans table + planning worker in proactive.py, (4) droid attribution on each tab.

⚖️Decisions

• Research presented as readable description, not raw JSON. — v1 dump UX was unfriendly; users care about content not structure. (Ninad)
• Adopt Carol design language (gradient + glass). — Visual consistency across the Carol ecosystem. (Ninad)
• Proactive tab introduces proactive_plans table + planning worker in proactive.py. — Daily 3-option plans need persistence and a generator. Coupling them into proactive.py keeps the module cohesive. (Orion)
• requester rewritten ninad -> orion per CAROL-INI-744: orion is the only human-CLI requester — Backfill of historical rows after INI744 added API-level refusal of requester=ninad. Orion is Ninads CLI agent; all human-originated initiatives are filed with requester=orion. (orion)

✅Success criteria

• All 4 v1.1 improvements visible in the UI (must_have)
• proactive_plans table populated daily by the worker (must_have)