docs(ai): mark Planner + Runner (Steps 4+5) as done in Workbench roadmap

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
This commit is contained in:
Till JS 2026-04-14 21:24:51 +02:00
parent 7535480007
commit b7a76ca24f

View file

@ -1760,10 +1760,17 @@ Code:
- [x] Schritt 1 — Actor-Attribution (Events + Records + Sync-Payload)
- [x] Schritt 2 — Policy-Config + `pendingProposals` + Propose-Path im Executor
- [x] Schritt 3 — Ghost-UI im Todo-Pilot (`<AiProposalInbox />`)
- [ ] Schritt 4 — Missions-Datenmodell + Planner-LLM-Task
- [ ] Schritt 5 — In-App MissionRunner (Foreground)
- [ ] Schritt 6 — Workbench-Timeline-Lens (cross-module AI-Aktivitaet)
- [ ] Schritt 7 — Server-side `mana-ai` / `mana-companion` Bun-Service
- [x] Schritt 4 — Missions-Datenmodell + Planner-LLM-Task
- Dexie `aiMissions` (v18), cross-device synct
- `aiPlanTask` als LlmTask (minTier browser, contentClass personal)
- Strikter JSON-Parser mit Tool-Allowlist + Rationale-Zwang
- [x] Schritt 5 — In-App MissionRunner (Foreground-Tick in `(app)/+layout.svelte`)
- `runMission(id, deps)` + `runDueMissions(now, deps)` — injiziert, testbar
- Default-Input-Resolver für Notes / Kontext / Goals
- 60-Sekunden-Tick, Overlap-Guard, idempotent
- [ ] Schritt 6 — Missions-UI (Create/Edit-Form + Iteration-History)
- [ ] Schritt 7 — Workbench-Timeline-Lens (cross-module AI-Aktivitaet)
- [ ] Schritt 8 — Server-side `mana-ai` Bun-Service (offline-of-tab Runs)
### 20.6 Offene Follow-ups