saravanakumardb1
14c7883d2a
docs(local-llm): mark all phases A-G complete in roadmap with commit links
2026-02-20 00:48:31 -08:00
saravanakumardb1
79cf42c8e3
docs(local-llm): mark Phase C complete with checkboxes and commit link
2026-02-20 00:27:33 -08:00
saravanakumardb1
f289099461
docs(local-llm): mark Phase B complete with checkboxes and commit link
2026-02-20 00:20:13 -08:00
saravanakumardb1
1335d47869
docs(local-llm): mark Phase A complete with checkboxes and commit link
2026-02-20 00:12:26 -08:00
saravanakumardb1
d7dc66eb92
docs(local-llm): Rich Features Roadmap — 45 tasks across 7 phases for coding agent
...
Detailed implementation roadmap for the Rich Features PRD with:
Phase A (Sprint 14-16, ~15hr): Foundation
A1: IndexedDB layer with idb — 9 object stores, compound indexes
A2: v4 TypeScript interfaces — all data models
A3: Route group (mission-control) — move existing dashboard
A4: Route group (workspace) — sidebar + content layout
A5: Sidebar — conversation list, time groups, search
A6: Conversation view — message thread, input bar, streaming
A7: Auto-title + context window usage bar
A8: v3 → v4 migration from localStorage
Phase B (Sprint 17-18, ~10hr): Quick Actions + Cmd+K
B1-B6: 30 built-in actions, fuse.js command palette, launcher,
custom editor, usage tracking, export/import
Phase C (Sprint 19-20, ~9hr): Custom Agents
C1-C5: 10 built-in agents, picker, full-screen editor,
conversation wiring (welcome msg, chips, temp), export
Phase D (Sprint 21-22, ~13hr): Model Router + Multi-Modal
D1-D7: regex classifier, model defaults, auto-routing UI,
rich input bar, file/voice/image processing, drag-drop
Phase E (Sprint 23, ~7hr): Response Enhancements
E1-E5: action bars, code-block copy, try-other-model,
live metrics, rating with aggregation
Phase F (Sprint 24-25, ~11hr): Scheduled Tasks
F1-F7: cron-parser, CRUD, editor, browser runner,
/api/system/exec with allowlist, notifications, templates
Phase G (Sprint 26-28, ~13hr): Projects + Orchestration
G1-G7: project CRUD, drag-to-project, system context,
Cmd+P switcher, chain/race/vote modes
Every task has: explicit file paths, step-by-step instructions,
pass/fail exit criteria, verification commands, and commit templates.
Dependency graph: A is foundation, B-F parallel after A, G needs A+B.
2026-02-19 23:54:07 -08:00
saravanakumardb1
7bd14054d4
docs(local-llm): Rich Features PRD rev 2 — comprehensive review + expansion
...
Review findings addressed (20+ issues):
Structure additions:
- Target Users section with 5 personas (solo dev, tinkerer, privacy pro, writer, power user)
- Non-Goals section (8 explicit out-of-scope items for v4)
- Risks & Mitigations table (10 risks with impact/likelihood/mitigation)
- New API Routes section (4 new routes with security notes)
- Settings Expansion section (full tree: General, Router, Models, Input, Tasks, Data, About)
- New Dependencies table (idb ~1KB, fuse.js ~6KB, cron-parser ~3KB)
- Error Handling appendix (12 edge cases with expected behavior)
Data model fixes:
- Conversation/Message split into separate IndexedDB stores (scalability)
- Message gets conversationId FK, promptTokens field, size/language on Attachment
- Design decision note explaining why messages are stored separately
Feature spec improvements:
- 3.1 Conversations: context window management (token bar, auto-summarize at 80/95%)
- 3.2 Quick Actions: expanded Cmd+K palette spec (5 result types, ranking)
- 3.3 Agents: tools marked v4 vs v5, duplicate-from-builtin, unlink on delete
- 3.4 Model Router: full resolveModel() with 4-level fallback chain + availability
- 3.5 Multi-Modal: attachment size limits, Whisper error handling
- 3.6 Response: hover-only action bars, rating aggregation per task type
- 3.7 Cron: built-in templates table, runtime constraints, security (execFile)
- 3.8 Orchestration: full data model, chain/race/vote UI specs, step limits
- 3.9 Projects: system context detail, project stats, unlink behavior
Acceptance criteria added to ALL 9 features (was missing on 5).
Competitive analysis expanded with local competitors (Open WebUI, LM Studio, Jan.ai).
Success metrics improved with measurement methodology and rationale.
Open questions restructured as decision table with recommendations.
IndexedDB schema with explicit indexes and compound keys.
Migration strategy: 7-step v3→v4 with safety (no delete until confirmed).
681 lines → 1149 lines (+69% content)
2026-02-19 23:47:59 -08:00
saravanakumardb1
1172dbb23e
docs(local-llm): Rich Features PRD — full local AI workspace spec
...
Comprehensive PRD evolving Mission Control into a ChatGPT-class local AI workspace:
- 3.1 Conversations: persistent, named, searchable, branching, IndexedDB
- 3.2 Quick Actions: 30 built-in 1-click launchers across 5 categories
(code, writing, analysis, creative, devops) + custom actions + Cmd+K palette
- 3.3 Custom Agents: 10 built-in local GPTs with system prompts, tools,
temperature, welcome messages, example prompts
- 3.4 Model Router: heuristic task classifier (<5ms, no LLM call),
auto-selects best model per task type, configurable defaults
- 3.5 Multi-Modal Input: file attach, voice (Whisper), images, drag-drop,
paste intelligence (code/image/error detection)
- 3.6 Response Enhancements: per-message actions, per-code-block copy,
branching with navigation, live metrics, rating/quality profiles
- 3.7 Scheduled Tasks: cron-based recurring prompts with shell/file input,
notification/file/conversation output, 5 built-in templates
- 3.8 Multi-Model Orchestration: chain, race, vote modes
- 3.9 Projects: conversation folders with system context + model defaults
7 implementation phases (~78hr), component architecture, storage migration,
competitive analysis, success metrics, open questions
2026-02-19 23:39:20 -08:00
saravanakumardb1
3dc0c441a9
docs(local-llm): mark all roadmap phases 1-6 complete with commit links
...
All 27 roadmap items + 5 bugs checked off across 6 phases:
- Phase 1 (040013e7f042976f6baf9588d21c44ad8a607d3911
2026-02-19 23:30:11 -08:00
saravanakumardb1
ae231d5aac
docs(local-llm): comprehensive roadmap review — 5 bugs, 6 phases, 31 items
...
Systematic code review of DASHBOARD_ROADMAP.md against actual codebase:
Bugs found (BN1-BN5):
- BN1: Compare buttons show unloaded models (can't generate)
- BN2: No AbortController on compare stream (leaks on close)
- BN3: Chat messages lost on modal close (no persistence)
- BN4: Logs panel has no refresh button
- BN5: Delete dialog missing reclaim size (partial impl exists)
Expanded from 4 phases to 6 + backlog (15 → 31 items):
- Phase 1: Pre-load intelligence + bug fixes (N1-N3, BN1-BN2, BN5)
- Phase 2: Rich metadata + persistence (N4-N5, BN3-BN4)
- Phase 3: Model intelligence badges + sort (N6-N10)
- Phase 4: Runtime metrics + UX polish (N11-N14)
- Phase 5 (NEW): Response quality — markdown, syntax highlight,
vision upload, think-block collapse, model library link
- Phase 6 (NEW): Data persistence — export/import, inference log,
factory reset
- Phase 7: Expanded backlog (F17-F38, +6 new ideas)
Improvements:
- Added checkboxes for all tasks and acceptance criteria
- Quant-aware RAM estimate multipliers (Q4/Q5/Q8/F16)
- Broader vision model regex (bakllava, moondream, llama-vision)
- DeepSeek R1 distill variant detection for think badge
- Conservative memory availability formula (free + cached*0.5)
- localStorage key registry with llm- prefix standardization
- Dependency graph between phases
- ~6 hrs total estimated effort
2026-02-19 23:02:25 -08:00
saravanakumardb1
cd6e561f1b
docs(local-llm): consolidate dashboard docs into dashboard/docs/
...
- Created DASHBOARD_PRD.md — full updated PRD with current 19-file
architecture, all 10 API routes, UI layout, data flow, localStorage
keys, security model, and v1-v3 changelog.
- Created DASHBOARD_ROADMAP.md — phased implementation plan for N1-N15
improvements across 4 phases: pre-load intelligence, rich metadata,
model intelligence badges, runtime metrics. Includes acceptance
criteria and implementation details per item.
- Updated DASHBOARD_REVIEW.md — refreshed file inventory to 19 files
(~2,930 lines), fixed broken Tier B markdown table, added cross-links.
- Replaced __LOCAL_LLMs/docs/05-mission-control-dashboard.md with
redirect pointer to new dashboard/docs/ location.
Dashboard docs are now co-located at __LOCAL_LLMs/dashboard/docs/:
- DASHBOARD_PRD.md (product requirements)
- DASHBOARD_REVIEW.md (audit + 39 completed items + N1-N15 proposals)
- DASHBOARD_ROADMAP.md (phased implementation plan)
2026-02-19 22:54:18 -08:00
