dsh-tdai-memory 0.4.5 → 0.4.6

package/README.en.md

@@ -0,0 +1,56 @@
+# DSH TDAI Memory
+
+[简体中文](README.md) | English
+
+Bring TencentDB Agent Memory's layered memory system to [DeepSeek Harness](https://github.com/deepseek-ai/deepseek-harness): capture conversations, extract structured memories and retrieve relevant context in later conversations. This project adapts the original OpenClaw plugin's core to DSH.
+
+## Features
+
+| Layer | Purpose |
+| --- | --- |
+| L0 | Capture raw conversations for later search |
+| L1 | Use an extraction model to organize facts, preferences and events |
+| L2 / L3 | Produce scene information and a user profile |
+| Recall and search | Retrieve relevant memory for the current message; expose `tdai_memory_search` and `tdai_conversation_search` |
+
+Data defaults to `~/.memory-tencentdb/memory-tdai`, allowing existing data in that directory to be reused. Capture, extraction and recall depend on enabled settings, model connections and retrieval results; saving or recalling every detail is not guaranteed.
+
+## Compatibility and installation
+
+Supports DSH Web / Desktop. The repository declares compatibility with `0.2.0-rc.2`, `0.2.1-rc.1` / `rc.2` / `rc.3` / `rc.4` / `rc.5` and versions matching `>=0.2.0-rc.2 <0.3.0`; see [package.json](package.json) for the complete prerelease range. Node.js `>=22.18` is required.
+
+For Desktop, install its `dsh` command through **Application → Manage dsh Command** first:
+
+```sh
+dsh plugin --profile desktop add dsh-tdai-memory
+```
+
+Replace `desktop` with `web` for Web. Restart DSH and open **Plugins → Installed → dsh-tdai-memory**. You can also install it through the [Enhancement Suite](https://github.com/Scorp1o117/dsh-enhancement-suite).
+
+## Quick setup
+
+1. Configure an **OpenAI-compatible LLM address, model and key** for memory extraction.
+2. Configure an **OpenAI-compatible embedding address, model and dimensions**. The default `http://127.0.0.1:8088/v1` is a configuration value, not an included service; provide a working embedding service yourself.
+3. Select **Automatic memory**, save and restart DSH to apply pipeline changes.
+4. After a conversation, ask the Agent to search memory or check the plugin logs for extraction and tool registration.
+
+| Mode | Capture | Extraction | Automatic recall | Search tools |
+| --- | --- | --- | --- | --- |
+| Automatic memory | On | On | On | On |
+| Search existing memory only | Off | Off | Off | On |
+| Pause memory features | Off | Off | Off | Off |
+
+Advanced settings expose individual switches, storage location and tuning. Existing mixed settings appear as a custom combination and are not overwritten automatically. The search-tool switch applies immediately; restart as prompted for other pipeline changes. The form writes edited fields only and retains unchanged keys.
+
+## Usage notes
+
+- The extraction model needs to produce reliable structured output. Extraction runs in the background; latency and quality depend on the model and service.
+- LLM deduplication is off by default. The default embedding configuration points to a local OpenAI-compatible endpoint; remote compatible endpoints do not require `node-llama-cpp`. For the fully local embedding backend, install and approve the optional `node-llama-cpp` build in the target profile separately.
+- If search tools are missing, check that they are enabled and look for `[tdai-memory] tools registered` in the logs. The interface follows the DSH language setting.
+- Captured conversations and memories are stored locally; extraction and embedding requests go to your configured services. Choose appropriate endpoints and back up your data.
+
+## Development, attribution and feedback
+
+- The [historical detailed documentation](https://github.com/Scorp1o117/dsh-tdai-memory/blob/7e1dd9c9b1214f5fc531f560296e492af204c4e1/README.md) preserves architecture, configuration examples and older notes; its install commands and compatibility information may be outdated.
+- This project ports TencentDB Agent Memory. See the corresponding directories for third-party code and license notices included with the repository.
+- [npm](https://www.npmjs.com/package/dsh-tdai-memory) · [Issues](https://github.com/Scorp1o117/dsh-tdai-memory/issues) · [MIT License](LICENSE)

package/README.md

@@ -1,181 +1,56 @@
-# dsh-tdai-memory
+# DSH TDAI Memory
 
-## DSH 0.2.1-rc.3
+简体中文 | [English](README.en.md)
 
-Supports DSH 0.2.1-rc.3, including explicit prerelease engines.dsh and peer ranges. This host update adds More navigation menus, keeps unsent drafts on New Session, and follows the official /models listing on the DeepSeek API-key route; the tool, system-prompt, settings and pi-ai interfaces used here remain unchanged, and no removed host API (`ctx.layout.entryVisibility`, `agentPresetNavigation`) was in use.
+将 TencentDB Agent Memory 的分层记忆系统接入 [DeepSeek Harness](https://github.com/deepseek-ai/deepseek-harness)：捕获对话、提取结构化记忆，并在后续对话中检索相关内容。项目基于原 OpenClaw 插件的核心进行 DSH 适配。
 
-## DSH 0.2.1-rc.2
+## 功能
 
-Supports DSH 0.2.1-rc.2, including explicit prerelease engines.dsh and peer ranges. This host update adds DeepSeek V4.1 Flash 1017 and Search/Fetch APIs; the plugin APIs used here remain unchanged. Existing reasoning settings and the opt-in augmentation policy are preserved.
+| 层级 | 用途 |
+| --- | --- |
+| L0 | 捕获原始对话，供后续检索 |
+| L1 | 使用提取模型整理事实、偏好和事件等结构化记忆 |
+| L2 / L3 | 生成场景信息与用户画像 |
+| 召回与搜索 | 按当前消息召回相关记忆；提供 `tdai_memory_search` 和 `tdai_conversation_search` 工具 |
 
-## DSH 0.2.1-rc.1
+数据默认保存在 `~/.memory-tencentdb/memory-tdai`，可复用已有同目录记忆。记忆写入、提取和召回依赖启用状态、模型连接及检索结果，不保证每条信息都会被保存或召回。
 
-Supports DSH 0.2.1-rc.1. Compatibility is declared through `engines.dsh` and matching peer ranges, including this exact prerelease. The existing 0.2.0-rc.2 support is retained.
+## 兼容性与安装
 
-The plugin follows the DSH language setting (Chinese and English in DSH 0.2.0-rc.2), including configuration, status messages and plugin-list metadata. Language-pack locales use the host fallback chain. Switching languages preserves unsaved settings; there is no separate plugin language selector.
+支持 DSH Web / Desktop；仓库声明兼容 `0.2.0-rc.2`、`0.2.1-rc.1` / `rc.2` / `rc.3` / `rc.4` / `rc.5` 和符合 `>=0.2.0-rc.2 <0.3.0` 的版本，预发布版本范围见 [package.json](package.json)。要求 Node.js `>=22.18`。
 
-## Configuration page (DSH 0.2.0-rc.2 and later)
+Desktop 先在 **应用 → 管理 dsh 命令** 安装桌面版 `dsh`：
 
-Open **Plugins → Installed → dsh-tdai-memory** from the homepage sidebar to configure and save this plugin. The page uses the official `plugins.bundle.config` interface, without a duplicate entry in global Settings. Web and Desktop share the page. This version requires DSH 0.2.0-rc.2 or a later 0.2.x host; existing configuration is retained.
-
-Choose **Automatic memory** for capture, extraction, recall and search together; **Search existing memory only** disables capture, extraction and automatic recall while keeping search tools; **Pause memory features** disables all four. The main page contains the two model connections. Individual switches, storage paths and tuning remain under **Advanced**. Existing mixed switch values appear as **Custom combination** and are preserved. Select a mode, Save, then restart DSH for pipeline changes (the search-tool switch applies immediately). Only edited fields are written; blank keys and untouched advanced options remain unchanged.
-
-[![中文文档](https://img.shields.io/badge/%E4%B8%AD%E6%96%87%E6%96%87%E6%A1%A3-blue)](README.zh.md)
-
-**GitHub**: [Scorp1o117/dsh-tdai-memory](https://github.com/Scorp1o117/dsh-tdai-memory) · **npm**: [dsh-tdai-memory](https://www.npmjs.com/package/dsh-tdai-memory)
-
-[![Enhancement Suite](https://img.shields.io/badge/part%20of-Enhancement%20Suite-3964fe)](https://github.com/Scorp1o117/dsh-enhancement-suite) [![npm](https://img.shields.io/npm/v/dsh-enhancement-suite)](https://www.npmjs.com/package/dsh-enhancement-suite)
-
-Part of the [DeepSeek Harness Enhancement Suite](https://github.com/Scorp1o117/dsh-enhancement-suite) — Vision · Soul/Persona · Long-term Memory · Plugin Marketplace.
-
-A port of **TencentDB Agent Memory** (Tencent Cloud's open-source four-layer
-memory system, originally an OpenClaw plugin) into DeepSeek Harness.
-
-## Compatibility (v0.4.1)
-
-Verified with DSH `0.1.7-rc.2` (Web) and `0.2.0-rc.2` (Desktop runtime) in isolated profiles. The Desktop app uses its own `desktop` profile. Other DSH prereleases remain unverified.
+```sh
+dsh plugin --profile desktop add dsh-tdai-memory
+```
 
-## Desktop install
+Web 将 `desktop` 改为 `web`。安装后重启 DSH，打开 **插件 → 已安装 → dsh-tdai-memory**。也可通过 [Enhancement Suite](https://github.com/Scorp1o117/dsh-enhancement-suite) 安装。
 
-Use the Desktop-installed `dsh` command (Application → Manage dsh Command), or the app’s Plugins page. Then install into the Desktop profile:
+## 快速配置
 
-```powershell
-dsh plugin --profile desktop add dsh-tdai-memory@0.3.6
-```
+1. 配置用于记忆提取的 **OpenAI 兼容 LLM 地址、模型和密钥**。
+2. 配置 **OpenAI 兼容 Embedding 地址、模型和维度**。默认地址 `http://127.0.0.1:8088/v1` 只是配置值，你需要自行提供可用的 Embedding 服务。
+3. 选择**自动记忆**并保存，重启 DSH 使记忆管线配置生效。
+4. 对话后可让 Agent 搜索记忆，或检查插件日志确认提取和工具注册情况。
 
-Restart the Desktop app to load the client bundle. Desktop keeps its profile under `$DSH_HOME/profiles/desktop`.
-
-
-## Features
-
-- **L0 conversation capture**: every turn (turn end, request boundary) is
-  written to raw conversation storage (JSONL + SQLite + FTS + vectors)
-- **L1 structured memory**: a background pipeline uses an LLM to extract
-  facts / preferences / events (persona / episodic / instruction) from
-  conversations, stored in `records/` + SQLite + FTS + vectors
-- **L2 scenes / L3 persona**: scene blocks and user profile generation
-  (pipeline-scheduled)
-- **Automatic recall injection**: on every prompt assembly, relevant memories
-  and the user profile are retrieved by the current user message and injected
-  as dynamic context (the model "just remembers")
-- **Tools**: `tdai_memory_search` (L1 structured search),
-  `tdai_conversation_search` (L0 raw-text search)
-
-The data directory reuses the existing `~/.memory-tencentdb/memory-tdai`, so
-**previously accumulated memories carry over seamlessly**.
-
-## Architecture (porting approach)
-
-| Layer | Content |
-|---|---|
-| Core | The host-neutral core of `tdai-memory-openclaw-plugin` (`src/core`, `src/utils`), tsc-compiled to ESM (`dist-dsh/`), zero changes |
-| Host adapter | `StandaloneHostAdapter` (official standalone mode, direct OpenAI-compatible calls) |
-| dsh shell | `index.js`: config mapping, `session/event` + `session/flush` capture, `system-prompt/assemble` recall injection on `agent.ctx`, tool registration, lifecycle |
-| Fallback | `recall-inject.js`: preset-row recall injection (used when mounted inside an agent preset) |
-
-Recall caches expire after 30 seconds and retain at most 128 sessions per
-injection instance. Failed or timed-out recalls are retried on the next assembly.
-The preset row also accepts `timeoutMs` (default 4000); reaching this deadline
-continues prompt assembly without memory context. Deadline timers are cleared
-when recall settles early. Underlying recall work may continue after the deadline.
-
-Hard-won wiring details:
-
-- **Capture**: `session/flush` listener (await semantics; must complete before
-  headless exits); `turn/start` timestamps as the L0 cursor floor; turn-id dedup
-- **Headless one-shot runs**: wait for `core.handleSessionEnd()` inside flush
-  (L1 extraction finishes before exit; otherwise the 5s shutdown timeout kills it)
-- **Recall injection**: must be registered on **`agent.ctx`** (assembly runs in
-  the agent scope; root listeners never see it); attach one tick after
-  `session/created` by resolving the agent from the `agents` service
-
-## Configuration (profile patch + settings)
-
-Configuration is stored in the `tdai-memory` entry of the active Profile patch.
-On first launch, DSH imports the old `$DSH_HOME/settings.yaml` values into that
-entry. The **Web UI Settings → 记忆**
-section edits every field (v0.2.0, write-only keys); TdaiCore is built at
-startup, so changes apply **after a restart**.
-
-```yaml
-# $DSH_HOME/profiles/web/cordis.patch.yml
-- id: tdai-memory
-  name: 'dsh-tdai-memory'
-  config:
-    extraction:
-      enabled: true
-      enableDedup: false      # dedup LLM output parsing is flaky; off by default
-    llm:                      # L1/L2/L3 extraction model (OpenAI-compatible)
-      baseUrl: 'https://opencode.ai/zen/go/v1'
-      model: 'mimo-v2.5'      # deepseek-v4-flash produces invalid extraction JSON
-      sendSessionHeader: true # send x-opencode-session on LLM requests (required by OpenCode Go & similar gateways)
-      sessionId: ''           # fixed session id; empty = persistent auto id under the data dir
-    embedding:                # vectors (OpenAI-compatible /v1/embeddings)
-      baseUrl: 'http://127.0.0.1:8088/v1'
-      model: 'Qwen3-Embedding-0.6B'
-      dimensions: 1024
-      sendDimensions: false
-```
+| 模式 | 捕获 | 提取 | 自动召回 | 搜索工具 |
+| --- | --- | --- | --- | --- |
+| 自动记忆 | 开 | 开 | 开 | 开 |
+| 仅搜索已有记忆 | 关 | 关 | 关 | 开 |
+| 暂停记忆功能 | 关 | 关 | 关 | 关 |
 
-## Install
+高级设置可单独调整开关、存储位置和调优参数；已有混合配置显示为自定义组合，不会自动覆盖。搜索工具开关可即时生效，其他管线配置按提示重启。表单只写入编辑过的字段，未修改的密钥保留。
 
-```bash
-dsh plugin --profile web add dsh-tdai-memory
-```
+## 使用说明
 
-then mount it in `$DSH_HOME/profiles/web/cordis.patch.yml`:
+- 提取模型需要稳定输出结构化结果；提取在后台进行，耗时和质量取决于模型与服务。
+- 默认关闭 LLM 去重。默认 Embedding 使用本地 OpenAI 兼容端点；远程兼容端点不需要 `node-llama-cpp`。若选择完全本地的 Embedding 后端，需要在目标 Profile 另行安装并批准可选的 `node-llama-cpp` 构建。
+- 搜索工具缺失时，确认工具开关已开启，并检查 `[tdai-memory] tools registered` 日志。语言跟随 DSH。
+- 捕获的对话和记忆会保存在本机；提取与 Embedding 请求会发送到你配置的服务，请自行选择合适的端点并备份数据。
 
-```yaml
-- insert:
-    - id: tdai-memory
-      name: 'dsh-tdai-memory'
-      config: {}          # fill through Settings → 记忆
-```
+## 开发、来源与反馈
 
-and restart `dsh web`. LLM/embedding API keys can be set in the Web UI
-settings page (记忆 / Memory) or in the `tdai-memory` Profile patch entry.
-
-> **Note for users**
-> - This plugin is a standard **profile bundle** (`dsh.bundle.patch`):
->   `dsh plugin --profile web add dsh-tdai-memory` installs and mounts it in
->   one step — no manual `cordis.patch.yml` edits needed.
-> - DSH exposes the registered `tdai-memory` settings namespace directly; the
->   plugin does not modify files in the host installation.
-> - Settings changes apply **after a restart** (TdaiCore is built at startup).
-> - Version 0.2.13 and newer require DSH `0.1.0-rc.7` or newer and are tested
->   against `0.1.0-rc.7`, `0.1.0-rc.8`, and `0.1.1-rc.1`.
-> - DSH `0.1.0-rc.6` users must pin `dsh-tdai-memory@0.2.11`, the last release
->   carrying the legacy settings-allowlist compatibility patch.
-
-`node-llama-cpp` is an optional peer used only by the fully local embedding
-backend. It is intentionally not installed by default because its native build
-requires explicit pnpm build approval. Remote OpenAI-compatible embeddings do
-not need it. Users who select the local backend should install and approve
-`node-llama-cpp` in the target DSH profile separately.
-
-## Known trade-offs
-
-- **Extraction model**: `mimo-v2.5` extracts correctly but takes 20-30s per
-  call (background execution, does not block the conversation);
-  `deepseek-v4-flash` is fast but its JSON output is non-compliant (extracts 0)
-- **dedup**: LLM conflict-detection output parsing is unstable (once caused
-  stored=0); off by default; enable only with a more reliable model
-- **L1 memory vectors**: written with storage (8088 embedding is fast); L0
-  vectors run as a background task, drained by `destroy()` on headless exit
-- **Search tools** (`toolsEnabled`, on by default): the switch lives in the
-  Plugins page and applies in place — the row config is `volatile`, so the
-  plugin re-syncs its registration without a restart. If both tools are
-  missing, look for the plugin's
-  `[tdai-memory] tools registered: tdai_memory_search, tdai_conversation_search`
-  log line (v0.4.1+). Before v0.4.1 it read `config.toolsEnabled` off the
-  Volatile wrapper cordis hands to `apply` — a property that never exists — so
-  the failure was completely silent (issue #3)
-- **Upgrades**: after pulling new upstream code, rerun
-  `npx tsc -p dsh-tsconfig.json` in the tdai project dir (output in `dist-dsh/`)
-
-## License
-
-MIT
-
-
-Saves use the configuration form API. Rejected writes retain the draft and show an error. Changing a model preserves credentials, and disabling a default-on option explicitly stores `false`.
+- [历史详细文档](https://github.com/Scorp1o117/dsh-tdai-memory/blob/7e1dd9c9b1214f5fc531f560296e492af204c4e1/README.zh.md)保留架构、配置示例和旧版说明，其中安装命令与兼容性信息可能已过时。
+- 本项目移植 TencentDB Agent Memory；随仓库分发的第三方代码和许可证请参见对应目录。
+- [npm](https://www.npmjs.com/package/dsh-tdai-memory) · [问题反馈](https://github.com/Scorp1o117/dsh-tdai-memory/issues) · [MIT License](LICENSE)

package/README.zh.md

@@ -1,124 +1,5 @@
-# dsh-tdai-memory
+# DSH TDAI Memory（简体中文）
 
-## DSH 0.2.1-rc.3
+中文文档已迁移至仓库首页：[阅读中文版](README.md)。
 
-适配 DSH 0.2.1-rc.3，明确声明该预发布版本的 engines.dsh 与 peer 范围。宿主新增侧栏/会话「更多」导航菜单、修复新建会话丢失未发送草稿、DeepSeek API Key 路由改跟随官方 /models 模型列表；本插件使用的工具、system-prompt、settings 与 pi-ai 接口保持不变，也未使用任何被移除的宿主 API（`ctx.layout.entryVisibility`、`agentPresetNavigation`）。
-
-## DSH 0.2.1-rc.2
-
-适配 DSH 0.2.1-rc.2，明确声明该预发布版本的 engines.dsh 与 peer 范围。宿主新增 DeepSeek V4.1 Flash 1017 及 Search/Fetch API，本插件使用的接口保持不变。保留已有推理配置及自动补齐显式开启策略。
-
-## DSH 0.2.1-rc.1
-
-支持 DSH 0.2.1-rc.1，通过 `engines.dsh` 和一致的 peer 范围声明兼容，明确包含该预发布版本，并保留已有 0.2.0-rc.2 支持。
-
-插件跟随 DSH 的语言设置（DSH 0.2.0-rc.2 内置中文和 English），配置页面、状态提示和插件列表名称/简介同步切换。扩展语言使用宿主的回退链。切换语言保留未保存的设置，无需单独选择插件语言。
-
-## 配置入口（DSH 0.2.0-rc.2 起）
-
-在首页侧边栏打开 **插件 → 已安装 → dsh-tdai-memory**，直接在插件详情页配置并保存。配置页注册到官方的 `plugins.bundle.config` 接口；全局设置页不再重复显示配置入口。Web 与桌面版使用相同界面，本版要求 DSH 0.2.0-rc.2 或更新的 0.2.x 版本。现有配置无需迁移。
-
-选择 **自动记忆**，一次开启捕获、提取、召回和搜索；**仅搜索已有记忆** 关闭捕获、提取和自动召回，保留搜索工具；**暂停记忆功能** 关闭全部四项。主页面只展示两个模型的连接信息，独立开关、存储路径和调优参数保留在 **高级设置** 中。已有开关组合显示为 **自定义组合**，不会自动覆盖。选择模式后保存，并重启 DSH 使记忆流水线设置生效（搜索工具开关即时生效）。只写入编辑过的字段，密钥留空及未编辑的高级设置保持原值。
-
-**GitHub**: [Scorp1o117/dsh-tdai-memory](https://github.com/Scorp1o117/dsh-tdai-memory) · **npm**: [dsh-tdai-memory](https://www.npmjs.com/package/dsh-tdai-memory) · [English](README.md)
-
-[![Enhancement Suite](https://img.shields.io/badge/part%20of-Enhancement%20Suite-3964fe)](https://github.com/Scorp1o117/dsh-enhancement-suite) [![npm](https://img.shields.io/npm/v/dsh-enhancement-suite)](https://www.npmjs.com/package/dsh-enhancement-suite)
-
-属于 [DeepSeek Harness Enhancement Suite](https://github.com/Scorp1o117/dsh-enhancement-suite) —— Vision · Soul/Persona · 长期记忆 · 插件市场。
-
-把 **TencentDB Agent Memory**（腾讯云开源的四层记忆系统，原为 OpenClaw 插件）移植进 DeepSeek Harness。
-
-## 兼容性（v0.4.1）
-
-已在 DSH `0.1.7-rc.2`（Web）与 `0.2.0-rc.2`（桌面端运行时）的一次性 Profile 验证。桌面端使用独立的 `desktop` Profile；其他预发布版本暂未验证。
-
-## 桌面端安装
-
-在桌面端的“插件”页面安装，或使用桌面端“应用 → 管理 dsh 命令”注册的命令：
-
-```powershell
-dsh plugin --profile desktop add dsh-tdai-memory@0.3.6
-```
-
-重启桌面端以加载客户端插件。配置位于 `$DSH_HOME/profiles/desktop`。
-
-
-## 能力
-
-- **L0 对话捕获**：每轮对话（turn 结束、请求边界）自动写入原始对话（JSONL + SQLite + FTS + 向量）
-- **L1 结构化记忆**：后台管线用 LLM 从对话中提取事实/偏好/事件（persona / episodic / instruction），写入 `records/` + SQLite + FTS + 向量
-- **L2 场景 / L3 画像**：场景块与用户画像生成（管线自动调度）
-- **自动召回注入**：每次组装提示词时，按当前用户消息检索相关记忆 + 画像，作为动态上下文注入（模型"凭空想起"）
-- **工具**：`tdai_memory_search`（L1 结构化检索）、`tdai_conversation_search`（L0 原文检索）
-
-数据目录复用现有 `~/.memory-tencentdb/memory-tdai`，**旧记忆无缝继承**。
-
-`node-llama-cpp` 仅供完全本地的 embedding 后端使用，现在声明为可选 peer，默认安装不会再触发原生构建脚本。使用 OpenAI 兼容远程 embedding 时不需要它；选择本地后端的用户需在目标 DSH profile 中单独安装并批准 `node-llama-cpp` 的构建。
-
-从 `0.2.13` 起最低支持 DSH `0.1.0-rc.7`，已针对 `0.1.0-rc.7`、
-`0.1.0-rc.8` 和 `0.1.1-rc.1` 测试。仍使用 DSH `0.1.0-rc.6` 的用户请锁定
-`dsh-tdai-memory@0.2.11`；这是最后一个包含旧 settings 白名单兼容补丁的版本。
-
-## 架构（移植方式）
-
-| 层 | 内容 |
-|---|---|
-| 核心 | `tdai-memory-openclaw-plugin` 的 host-neutral 核心（`src/core`、`src/utils`），tsc 编译为 ESM（`dist-dsh/`），零改动 |
-| 宿主适配 | `StandaloneHostAdapter`（官方 standalone 模式，OpenAI 兼容直调） |
-| dsh 壳 | `index.js`：配置映射、`session/event`+`session/flush` 捕获、`agent.ctx` 上的 `system-prompt/assemble` waterfall 召回注入、工具注册、生命周期 |
-| 备用 | `recall-inject.js`：preset 行形态的召回注入（挂在 agent preset 内时用） |
-
-召回缓存默认 30 秒过期，每个注入实例最多保留 128 个会话。失败或超时的召回
-会在下次组装时重试。preset 行还支持 `timeoutMs`（默认 4000），到达期限后
-不注入记忆并继续组装提示词；提前完成时会清除超时定时器。期限到达后，底层
-召回任务可能仍会继续运行。
-
-关键接线（都是踩坑换来的）：
-
-- **捕获**：`session/flush` 监听器（await 语义，headless 退出前必完成）；`turn/start` 时间戳做 L0 cursor 下限；turn id 去重
-- **headless 一次性任务**：flush 内再等 `core.handleSessionEnd()`（L1 提取跑完才退出；否则 5s 关闭超时会杀掉提取）
-- **召回注入**：必须注册在 **`agent.ctx`** 上（组装在 agent 作用域进行，root 监听器收不到）；`session/created` 后延迟一 tick 从 `agents` 服务拿 agent
-
-## 配置（profile patch + settings）
-
-配置以 **settings 命名空间**驱动：profile patch 作为 base 层，`$DSH_HOME/settings.yaml` 的 `tdai-memory:` 节覆盖（LLM/embedding 密钥已迁到 settings.yaml）。首页侧边栏插件 → dsh-tdai-memory 可编辑全部字段（v0.2.0，含写-only 密钥）；TdaiCore 启动时构建，改动**重启后生效**。
-
-```yaml
-# $DSH_HOME/settings.yaml
-tdai-memory:
-  llm:
-    apiKey: 'sk-...'
-  embedding:
-    apiKey: 'local-no-key'
-```
-
-```yaml
-# profile patch（base 层）
-- id: tdai-memory
-  name: 'dsh-tdai-memory'
-  config:
-    extraction:
-      enabled: true
-      enableDedup: false      # dedup 的 LLM 输出解析不稳，先关
-    llm:                      # L1/L2/L3 提取模型（OpenAI 兼容）
-      baseUrl: 'https://opencode.ai/zen/go/v1'
-      model: 'mimo-v2.5'      # deepseek-v4-flash 对提取 JSON 输出不合格
-      sendSessionHeader: true # 给 LLM 请求带 x-opencode-session（OpenCode Go 等网关要求）
-      sessionId: ''           # 固定会话 ID；留空 = 数据目录下自动持久化 ID（重启不变）
-    embedding:                # 向量（OpenAI 兼容 /v1/embeddings）
-      baseUrl: 'http://127.0.0.1:8088/v1'
-      model: 'Qwen3-Embedding-0.6B'
-      dimensions: 1024
-      sendDimensions: false
-```
-
-## 已知取舍
-
-- **提取模型**：`mimo-v2.5` 提取正确但单次 20-30s（后台执行，不阻塞对话）；`deepseek-v4-flash` 快但 JSON 输出不合规（提取 0 条）
-- **dedup**：LLM 冲突检测输出解析不稳（曾导致 stored=0），默认关闭；开启需换更稳的模型
-- **L1 记忆向量**：随存储写入（8088 embedding 快）；L0 向量走后台任务，headless 退出时由 `destroy()` drain
-- **搜索工具**（`toolsEnabled`，默认开启）：开关在插件页，切换后**原地生效**（行配置是 `volatile`，无需重启即重新注册）。若两个工具完全缺失，可查插件日志 `[tdai-memory] tools registered: …`（v0.4.1 起）；v0.4.1 之前它读的是 cordis 交给 `apply` 的 Volatile 包装对象上的 `config.toolsEnabled` —— 那个属性永远不存在，因此整个失败过程完全静默（issue #3）
-- **升级**：上游拉新代码后，在 tdai 项目目录重跑 `npx tsc -p dsh-tsconfig.json`（产物在 `dist-dsh/`）
-
-
-保存使用配置表单 API；宿主拒绝写入时保留草稿并提示错误。只修改模型不会覆盖密钥，关闭默认开启的选项会显式保存 `false`。
+[English](README.en.md)

package/package.json

@@ -1,6 +1,6 @@
 {
     "name": "dsh-tdai-memory",
-    "version": "0.4.5",
+    "version": "0.4.6",
     "description": "DeepSeek Harness 的 TencentDB Agent Memory 移植：L0 对话捕获 → L1 结构化记忆提取 → L2 场景/L3 画像，自动召回注入 + 记忆/对话搜索工具；复用现有 ~/.memory-tencentdb/memory-tdai 数据；附 Web UI 设置栏。",
     "type": "module",
     "main": "index.js",
@@ -18,7 +18,7 @@
             "platform": "web"
         },
         "compatibility": {
-            "dsh": ">=0.2.0-rc.2 <0.3.0 || 0.2.1-rc.1 || 0.2.1-rc.2 || 0.2.1-rc.3",
+            "dsh": ">=0.2.0-rc.2 <0.3.0 || 0.2.1-rc.1 || 0.2.1-rc.2 || 0.2.1-rc.3 || 0.2.1-rc.4 || 0.2.1-rc.5",
             "dshReleases": {
                 "0.1.5-rc.2": "incompatible",
                 "0.1.5-rc.3": "incompatible",
@@ -31,7 +31,9 @@
                 "0.2.0-rc.2": "compatible",
                 "0.2.1-rc.1": "compatible",
                 "0.2.1-rc.2": "compatible",
-                "0.2.1-rc.3": "compatible"
+                "0.2.1-rc.3": "compatible",
+                "0.2.1-rc.4": "compatible",
+                "0.2.1-rc.5": "compatible"
             },
             "node": ">=22.18",
             "profiles": [
@@ -68,7 +70,7 @@
     "license": "MIT",
     "engines": {
         "node": ">=22.18",
-        "dsh": ">=0.2.0-rc.2 <0.3.0 || 0.2.1-rc.1 || 0.2.1-rc.2 || 0.2.1-rc.3"
+        "dsh": ">=0.2.0-rc.2 <0.3.0 || 0.2.1-rc.1 || 0.2.1-rc.2 || 0.2.1-rc.3 || 0.2.1-rc.4 || 0.2.1-rc.5"
     },
     "repository": {
         "type": "git",
@@ -87,10 +89,14 @@
         "sqlite-vec": "0.1.7-alpha.2",
         "undici": "^8.1.0"
     },
+    "devDependencies": {
+        "@deepseek-ai/dsh-settings": "0.2.1-rc.5",
+        "@deepseek-ai/dsh-tools": "0.2.1-rc.5"
+    },
     "peerDependencies": {
         "react": "^18.2.0",
-        "@deepseek-ai/dsh-tools": ">=0.2.0-rc.2 <0.3.0 || 0.2.1-rc.1 || 0.2.1-rc.2 || 0.2.1-rc.3",
-        "@deepseek-ai/dsh-settings": ">=0.2.0-rc.2 <0.3.0 || 0.2.1-rc.1 || 0.2.1-rc.2 || 0.2.1-rc.3",
+        "@deepseek-ai/dsh-tools": ">=0.2.0-rc.2 <0.3.0 || 0.2.1-rc.1 || 0.2.1-rc.2 || 0.2.1-rc.3 || 0.2.1-rc.4 || 0.2.1-rc.5",
+        "@deepseek-ai/dsh-settings": ">=0.2.0-rc.2 <0.3.0 || 0.2.1-rc.1 || 0.2.1-rc.2 || 0.2.1-rc.3 || 0.2.1-rc.4 || 0.2.1-rc.5",
         "node-llama-cpp": "^3.19.1"
     },
     "peerDependenciesMeta": {