Skip to content

docs: update to v0.6.0 status #50

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
tryweb merged 1 commit into from
Mar 31, 2026
Merged

docs: update to v0.6.0 status #50

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 4 additions & 2 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -103,6 +103,7 @@ That's it! OpenCode will now automatically capture key decisions and inject them
| `memory_forget` | Remove or disable a memory | [Doc](docs/ADVANCED_CONFIG.md#memory_forget) |
| `memory_what_did_you_learn` | Show recent learning summaries | [Doc](docs/ADVANCED_CONFIG.md#memory_what_did_you_learn) |
| `memory_dashboard` | Weekly learning dashboard with trends | [Doc](docs/ADVANCED_CONFIG.md#memory_dashboard) |
| `memory_kpi` | Learning effectiveness KPIs (retry-to-success, memory lift) | [Doc](docs/ADVANCED_CONFIG.md#memory_kpi) |

*(For full details on tools like **Effectiveness Feedback**, **Cross-Project Sharing**, **Deduplication**, **Citations**, and **Episodic Learning**, please refer to the [Advanced Configuration](docs/ADVANCED_CONFIG.md).)*

Expand Down Expand Up @@ -165,6 +166,7 @@ Alternatively, install via `.tgz` release asset or build from source. See [Insta

## 🗺️ Version History

- **v0.6.0**: Learning Dashboard, KPI Pipeline, Feedback-Driven Ranking, Task-Type Aware Injection
- **v0.5.0**: Memory Explanation Tools (`memory_why`, `memory_explain_recall`)
- **v0.4.0**: Citation Model (`memory_citation`, `memory_validate_citation`)
- **v0.3.0**: Episodic Learning Hooks
Expand All @@ -190,5 +192,5 @@ See [CHANGELOG.md](CHANGELOG.md) for all changes.
- **Issues**: Submit errors or requests on [GitHub Issues](https://github.com/tryweb/lancedb-opencode-pro/issues).
- **License**: MIT License - see [LICENSE](LICENSE).

**Last Updated**: 2026-03-29
**Latest Version**: v0.5.0
**Last Updated**: 2026-03-31
**Latest Version**: v0.6.0
6 changes: 4 additions & 2 deletions README_zh.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -103,6 +103,7 @@ curl http://127.0.0.1:11434/api/tags
| `memory_remember` | 手動儲存記憶 | [進階設定](docs/ADVANCED_CONFIG.zh.md#memory_remember) |
| `memory_forget` | 移除或停用記憶 | [進階設定](docs/ADVANCED_CONFIG.zh.md#memory_forget) |
| `memory_what_did_you_learn` | 顯示近期學習摘要 | [進階設定](docs/ADVANCED_CONFIG.zh.md#memory_what_did_you_learn) |
| `memory_kpi` | 學習效能 KPI (retry-to-success, memory lift) | [進階設定](docs/ADVANCED_CONFIG.zh.md#memory_kpi) |

*(關於 **記憶效果回饋**、**跨專案記憶共享**、**去重複機制**、**引用模型** 以及 **事件式學習** 等更詳細的功能說明，請參見 [進階設定](docs/ADVANCED_CONFIG.zh.md)。)*

Expand Down Expand Up @@ -165,6 +166,7 @@ docker compose exec opencode-dev npm run verify:full

## 🗺️ 版本歷史

- **v0.6.0**: 新增學習儀表板、KPI 管線、反饋驅動排序、任務類型感知注入
- **v0.5.0**: 新增記憶解釋工具 (`memory_why`, `memory_explain_recall`)
- **v0.4.0**: 新增引用模型追蹤與驗證 (`memory_citation`, `memory_validate_citation`)
- **v0.3.0**: 引入事件式學習 Hooks
Expand All @@ -190,5 +192,5 @@ docker compose exec opencode-dev npm run verify:full
- **報告問題**: 軟體錯誤回報或功能請求，請至 [GitHub Issues](https://github.com/tryweb/lancedb-opencode-pro/issues) 提交。
- **授權協議**: MIT License - 詳見 [LICENSE](LICENSE)。

**最後更新**: 2026-03-29
**最新版本**: v0.5.0
**最後更新**: 2026-03-31
**最新版本**: v0.6.0
14 changes: 7 additions & 7 deletions docs/backlog.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -84,12 +84,12 @@

| BL-ID | Title | Priority | Status | OpenSpec Change ID | Spec Path | Notes |
|---|---|---|---|---|---|---|
| BL-030 | Learning dashboard summary | P0 | planned | TBD | TBD | 週摘要可視化 [Surface: Plugin；新增/擴充 plugin summary tool 進行聚合與格式化呈現] |
| BL-031 | Learning KPI pipeline | P0 | planned | TBD | TBD | retry-to-success / lift 指標 [Surface: Docs + optional plugin tool] |
| BL-030 | Learning dashboard summary | P0 | **done** | 2026-03-30-learning-dashboard-summary | openspec/specs/learning-dashboard-summary/ | 週摘要可視化 [Surface: Plugin] v0.6.0 |
| BL-031 | Learning KPI pipeline | P0 | **done** | 2026-03-30-learning-kpi-pipeline | openspec/specs/learning-kpi-pipeline/ | retry-to-success / lift 指標 [Surface: Plugin] v0.6.0 |
| BL-032 | Eval harness for learning quality | P1 | planned | TBD | TBD | 固定資料集回歸驗證 [Surface: Test-infra] |
| BL-033 | A/B testing framework | P1 | planned | TBD | TBD | learning feature 效益驗證 [Surface: Test-infra + docs] |
| BL-038 | Feedback-driven ranking / routing weights | P0 | planned | TBD | TBD | 根據 feedback 動態調整記憶排序與注入權重 [Surface: Plugin] |
| BL-039 | Task-type aware injection policy | P0 | planned | TBD | TBD | 依任務類型（coding/docs/review/release）調整注入策略 [Surface: Plugin] |
| BL-038 | Feedback-driven ranking / routing weights | P0 | **done** | 2026-03-30-feedback-driven-ranking | openspec/specs/feedback-factor/ | 根據 feedback 動態調整記憶排序與注入權重 [Surface: Plugin] v0.6.0 |
| BL-039 | Task-type aware injection policy | P0 | **done** | 2026-03-31-task-type-injection-policy | openspec/specs/task-type-injection/ | 依任務類型（coding/docs/review/release）調整注入策略 [Surface: Plugin] v0.6.0 |
| BL-040 | Success pattern playbook surface 強化 | P1 | planned | TBD | TBD | 把已存的 episodic success patterns 產品化成可感知的跨任務 playbook surface [Surface: Plugin；以 playbook summary/recommendation tool 交付] |

## Epic 9 — 儲存引擎與規模韌性
Expand Down Expand Up @@ -121,10 +121,10 @@ BL-001, BL-002, BL-005, BL-006, BL-008, BL-010, BL-011, BL-012
### Release B（經驗學習閉環）— ✅ DONE
BL-003, BL-014, BL-015, BL-016, BL-017, BL-018, BL-019, BL-020

### Release C（治理與產品化）— ✅ DONE（部分已取消）
已完成：BL-022, BL-023, BL-024, BL-025, BL-026, BL-027, BL-034
### Release C（治理與產品化）— ✅ DONE（v0.6.0 完成）
已完成：BL-022, BL-023, BL-024, BL-025, BL-026, BL-027, BL-030, BL-031, BL-034, BL-038, BL-039
已取消（不值得做）：BL-028, BL-029, BL-035
待處理：BL-021（Plugin；視 upstream 事件）, BL-030（Plugin）, BL-031（Docs + optional plugin tool）, BL-032（Test-infra）, BL-033（Test-infra + docs）, BL-038（Plugin）, BL-039（Plugin）, BL-040（Plugin）
待處理：BL-021（Plugin；視 upstream 事件）, BL-032（Test-infra）, BL-033（Test-infra + docs）, BL-040（Plugin）

### Release D（儲存引擎與規模韌性）— 📝 PLANNED
BL-036, BL-037
Expand Down
24 changes: 12 additions & 12 deletions docs/roadmap.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -348,13 +348,13 @@ OpenCode 要從「有長期記憶的工具」進化成「會累積團隊工作

## Phase 4 — 自適應與產品化（4-8 週）

**目標**：讓 OpenCode 在不同專案與團隊中越跑越順。
**目標**：讓 OpenCode 在不同專案和團隊中越跑越順。

### 功能項目
- [ ] feedback-driven ranking / routing weights
- [ ] task-type aware injection policy
- [ ] learning dashboard / weekly summary（Surface: Plugin）
- [ ] KPI 定義與報表管線（Surface: Docs + optional plugin tool）
- [x] feedback-driven ranking / routing weights（v0.6.0）
- [x] task-type aware injection policy（v0.6.0）
- [x] learning dashboard / weekly summary（v0.6.0）
- [x] KPI 定義與報表管線（v0.6.0）
- [ ] regression evals for learning quality（Surface: Test-infra）
- [ ] A/B framework 驗證學習功能是否真的提高效率（Surface: Test-infra + Docs）
- [ ] large-scope retrieval 的 ANN fast-path（Surface: Plugin）
Expand Down Expand Up @@ -390,10 +390,10 @@ OpenCode 要從「有長期記憶的工具」進化成「會累積團隊工作
## 優先級清單（先做什麼）

### P0
1. learning dashboard / weekly summary（Surface: Plugin）→ BL-030
2. KPI 定義與報表管線（Surface: Docs + optional plugin tool）→ BL-031
3. feedback-driven ranking / injection adaptation（Surface: Plugin）→ BL-038
4. task-type aware injection policy（Surface: Plugin）→ BL-039
1. learning dashboard / weekly summary（Surface: Plugin）→ BL-030 ✅ DONE v0.6.0
2. KPI 定義與報表管線（Surface: Docs + optional plugin tool）→ BL-031 ✅ DONE v0.6.0
3. feedback-driven ranking / injection adaptation（Surface: Plugin）→ BL-038 ✅ DONE v0.6.0
4. task-type aware injection policy（Surface: Plugin）→ BL-039 ✅ DONE v0.6.0

### P1
5. eval harness for learning quality（Surface: Test-infra）→ BL-032
Expand Down Expand Up @@ -421,11 +421,11 @@ OpenCode 要從「有長期記憶的工具」進化成「會累積團隊工作

如果只能選最有感的前三件事，建議依序做：

1. **Learning dashboard / weekly summary（BL-030，Plugin）**
1. **Learning dashboard / weekly summary（BL-030，Plugin）** ✅ DONE v0.6.0
- 最快把既有學習成果轉成使用者真正看得見的產品體感。
2. **KPI + eval harness（BL-031 + BL-032，Docs / Test-infra）**
2. **KPI + eval harness（BL-031 + BL-032，Docs / Test-infra）** ✅ DONE v0.6.0
- 先把「是否真的越做越會成功」量化，避免只靠感覺判斷。
3. **Feedback-driven ranking + task-type aware injection（BL-038 + BL-039，Plugin）**
3. **Feedback-driven ranking + task-type aware injection（BL-038 + BL-039，Plugin）** ✅ DONE v0.6.0
- 讓既有記憶與偏好不只被存下來，還能更直接影響下一輪任務成敗。

若要平行壓低技術風險，建議以「低風險可驗證」切片穿插進行：
Expand Down