dsh-memu-memory 1.1.3 → 1.2.1

package/AGENTS.md

@@ -14,7 +14,7 @@
 
 1. **不改 DSH 源码**，只用插件、公开 service、RPC、slot；只用公开扩展点。
 2. **不让主 agent 写全局记忆**：全局只由 curator 写，主 agent 只读。
-3. **动态记忆不进 system prompt**（静态 guide 可以）：走独立 plugin 消息，保护 KV cache 也让注入可折叠。`persona` 带 `complete: true` 的预设会丢弃本插件的 system prompt 段，这类会话的 guide 改由首步一条 `<memory-guide>` 消息投放（`withGuideMessage`，按 `binding.guideMessageAt` 只发一次，`lib/inject.js`）——仍然是 plugin 消息，不是把动态记忆写进 system prompt。
+3. **记忆规则与动态记忆都走 plugin 消息**：当前版本不再注册 system prompt 段；静态 guide 在回合首步以 `<memory-guide rules="…">` 投放（`rulesKey` 处理切域/改稿/压缩丢失），动态记忆走独立 plugin 消息，保护 KV cache 并让注入可折叠。
 4. **不每步注入**：普通回合首步注入；长工具回合只在出现尚未处理的真实人类消息批次时追加一次。工具日志、心跳不算人类消息。
 5. **不自动删除记忆**：永久记忆移出 = 边缘化；删库必须显式确认；删会话默认只解绑。
 6. **Python 不装进系统解释器 / 用户 site-packages**，只装 `engine/<version>/venv`，且 venv 里不留 pip（`--without-pip` 建，再用基础 Python 的 `pip --python` 指进去装）。

package/README.md

@@ -23,14 +23,14 @@
 
 ---
 
-`dsh-memu-memory` is built for long-form RP and companion chats. It gives a DSH conversation a memory that behaves like memory instead of a log: three retention layers that fade, a first-turn anchor plus at most one per-turn recall, provenance that cannot be quietly upgraded from "I inferred it" to "they said it", contradictions that wait for a decision instead of both being read as true, and an optional relationship graph with evidence. Speakers, characters, worlds and story branches form hard boundaries, and what a character is allowed to know stays separate from what actually happened.
+`dsh-memu-memory` is built for long-form RP and companion chats. It gives a DSH conversation a memory that behaves like memory instead of a log: three retention layers that fade, a first-turn anchor plus at most one per-turn recall, provenance that cannot be quietly upgraded from "I inferred it" to "they said it", contradictions that wait for a decision instead of both being read as true, near-identical facts collapse to one entry and one recall seat, and an optional relationship graph with evidence. Speakers, characters, worlds and story branches form hard boundaries, and what a character is allowed to know stays separate from what actually happened.
 
 It is **local-first by default**. The default `agent` backend stores full text in JSON indexes and recalls lexically — no Python, no embedding service, no API key. The `api` backend switches recall to the optional, pinned MemU engine (vector search). MemU is a plugin for people who want it, not a requirement for the plugin to start.
 
 > [!WARNING]
 > **Early-stage software.** This plugin is still early and may contain unforeseen errors, bugs or breaking changes. Back up your data, and please [report an issue](https://github.com/Tkingxiao/dsh-memu-memory/issues) or send a pull request — feedback helps the project improve.
 
-> **Status:** 1.1.2. The first public release was 1.0.0. The manual now lives in [`docs/`](docs/), with [`AGENTS.md`](AGENTS.md) as a short index. Back up `~/.dsh/dsh-memu-memory` before upgrades, and read [Compatibility & limits](#compatibility--limits) before storing data you cannot re-create.
+> **Status:** 1.2.1. The first public release was 1.0.0. The manual now lives in [`docs/`](docs/), with [`AGENTS.md`](AGENTS.md) as a short index. Back up `~/.dsh/dsh-memu-memory` before upgrades, and read [Compatibility & limits](#compatibility--limits) before storing data you cannot re-create.
 
 ## Screenshots
 
@@ -59,7 +59,7 @@ All screenshots are the real WebUI on isolated example data, not anyone's memori
 - **Three layers, one timeline.** `short_term` drafts expire after 72 hours by default; `long_term` is the body of ordinary memory; `must_remember` needs a reason and a source, decays over 365 days with a 0.35 strength floor, and can be *marginalized* but never auto-deleted to make room.
 - **Recall is a moment, not a dump.** The first turn after a session starts (or after compaction) gets an anchor block with budgets for pinned, core and global memory. After that, each user turn gets at most one recall. Deduplication and a token budget are enforced in code, not left to the prompt.
 - **A date is either an anniversary or one day.** `MM-DD` recurs every year (a birthday); `YYYY-MM-DD` fires the day before and the day itself, then is spent. A memory that says "tomorrow" is stored as the dated form, so an appointment does not come back every year.
-- **Memory fades unevenly.** Recency, access count, peak and end effects, primacy, habituation and flashbulb accuracy feed one strength score. Arousal and mood congruence are conditional rather than always-on: both are off by default and only engage when the mood is far enough from flat (|valence| >= 0.15, or |arousal - 0.35| >= 0.1), so in a calm conversation both multipliers are exactly 1. A vivid memory can be more confident than it is accurate, and the UI says so.
+- **Memory fades unevenly.** Recency, access count, peak and end effects, primacy, habituation and flashbulb accuracy feed one strength score. Mood is described by four components (joy / anger / sorrow / fear, summing to 1) plus arousal (intensity) and dominance (sense of control); a compound mood such as jealousy is a named mix of the four, not a new axis. Mood only reorders recall: it never enters the injected text and never rewrites a memory body. Gating is the OR of three independent conditions — the dominant share is >= 0.35, |arousal - 0.35| >= 0.1, or |dominance| >= 0.25; when none holds, both multipliers are exactly 1. Older entries carry no emotion and keep the old valence path; the four axes are never inferred back from valence. A vivid memory can be more confident than it is accurate, and the UI says so.
 - **Provenance is a fact, not a style.** Every entry carries `source` (`user_said` / `inferred` / `imagined` / `told`), `confidence`, `accuracy`, and disclosure rules. A vague recollection cannot impersonate a quote.
 - **Contradictions wait.** A write that rewrites or contradicts an existing entry lands as `pending_review`: it is not injected, not recalled, and appears in a review queue until someone accepts or rejects it.
 - **Secrets stay excluded.** `secret` visibility, group-chat channels, characters, worlds, branches and `knownBy` all constrain reads and edits. Automatic recall never injects a `secret` memory, and a `secret` is still readable by an explicit recall from the same folder — it hides a memory from the prompt, it is not a lock. What actually separates two conversations in one folder is the `context` scope (`channelId` / `characterId` / `worldId` / `branchId`); a blank scope on both sides means one shared scope, which is the point of a folder. The old `visibility: 'session'` level is gone — reading turns it into `shared` with its `context` untouched, and `node scripts/migrate-visibility.mjs` rewrites libraries that still store it (`--apply`, backs up first).
@@ -67,10 +67,40 @@ All screenshots are the real WebUI on isolated example data, not anyone's memori
 - **Images are pointers and captions, not pretend vision.** Attachments are deduplicated by SHA-256, quota-limited and tiered; an image without a description is remembered as "received an image", never as a scene the model did not see.
 - **Housekeeping happens in the background.** Consolidation runs in a seeded fork sub-session, never as an editable turn in your chat. Failed runs back off instead of reporting success.
 
+### Active memory, measured
+
+Idle consolidation is not "every quiet session gets a fork". Before a fork the plugin scores the evidence since the last cursor: human turns (`H`), content volume (`C`: human text, assistant text and deduped successful tool results, 2,000 characters per result) and effective tool work units (`U`), with `score = 0.35*min(H/3,2) + 0.45*min(C/4000,2) + 0.20*min(U/20,2)` (default threshold 0.8). A session idle for 15 minutes wakes below that score only when it is small; transcripts of 2,000+ events must earn their fork, because the host seeds the full prefix. A recent consolidation or a recent recall suppresses the idle fork.
+
+More is not better: thin turns on a large transcript are skipped by the guard (about 37% fewer prompt tokens on the same scenarios, at the cost of roughly 1 in 22 labels arriving a little later), but once the unconsolidated range passes 600 events the check runs unconditionally, so decisions made early in a long session do not stay out of memory forever.
+
+Four real long sessions from this machine were replayed read-only (no new chat) and their turns were counted:
+
+| Session | Events | Turns | Score-qualified | Below threshold | Fresh human turns after cursor |
+|---|---:|---:|---:|---:|---:|
+| `62056bff…` | 5,913 | 12 | 6 | 6 | 0 |
+| `adedb9ac…` | 5,766 | 23 | 16 | 7 | 0 |
+| `e38a0de0…` | 85 | 2 | 0 | 2 | 0 |
+| `edf7dc5c…` | 78 | 4 | 0 | 4 | 0 |
+
+Below-threshold turns on the two small sessions can still wake through starvation; on the 2,000+ event sessions they do not. None of the four had new human input after its real cursor, so none would fork again immediately.
+
+Provider-reported usage from real forks: one consolidation costs about **41k–217k input tokens** (including cache reads; it depends on memory-tool steps) and **0.8k–3.4k output**. At one 4-hour long session per day and 2–6 effective consolidations:
+
+| Consolidations/day | Monthly input | Monthly output |
+|---|---:|---:|
+| 2 | ~5.6M | ~0.11M |
+| 4 | ~11M | ~0.23M |
+| 6 | ~17M | ~0.34M |
+| 6 (worst measured fork each) | ~39M | ~0.7M |
+
+For scale, the whole 4-hour session tree measured here consumed 132M input and 1.28M output in one day, so active-memory forks are about **0.2%–1%** of the total. These are token counts, not money; cache reads are usually billed at a lower rate. The exact figure depends on turn count, memory-tool steps and how often the session actually goes idle.
+
 ## One WebUI
 
 The sidebar has exactly **one** entry: **MemU WebUI**. The memory wall and all settings live in the same navigation, with deep links such as `#/settings/memories`, `#/settings/folders` and `#/settings/stats`. The interface follows the browser locale (Chinese or English).
 
+**Private / global groups with multi-select.** The "Memories" page splits into **Private** and **Global** blocks (private first, default). The private block supports **multi-select**: bulk archive, restore and permanent delete (two-step confirm) — one failed row never eats the rest. Every row on the Global memory page has a "View" action that jumps straight to that entry on the wall.
+
 **In-place work.** Clicking a memory card opens a right-side drawer with the full text and the fields that decide how it behaves — title, content, tags, disclosure, confidence, open loop — plus save, archive, restore and "open in the browser" for heavier work. Archived memories can be deleted in the same drawer; the first click only arms it and the second confirms. Global memories are read-only in the drawer and link to their own page. Filters and scroll position stay where they were. Cards that carry images use the first usable attachment as a cover, and the drawer opens the full gallery; `secret` and `never_say` entries never receive a pixel URL, so they show a label instead.
 
 **Graph and list together.** The **Relations** tab renders the graph canvas and the adjacency list side by side (stacked on narrow screens), with the focus selector above them and link/entity/relationship tools below. There is no mode switch to forget.
@@ -79,13 +109,13 @@ The locally bundled D3 force layout preserves positions and the camera across ba
 
 | Page | What it is for |
 |---|---|
-| Memories | Cross-library wall, source/status/library/preview, in-place drawer, Relations tab |
+| Memories | Private/global grouped wall, source/status/library/preview, in-place drawer, multi-select bulk ops, Relations tab |
 | Engine & connection | Optional MemU engine install/switch, embedding endpoint, backups and restore |
 | Memory settings | Injection mode, token budgets, topK, relevance threshold, retention, automation |
 | Folders | Folder-level private/global ceiling, expandable conversations and their own switches |
 | Libraries | Create, rename, archive, restore, explicitly delete libraries |
 | Memory browser | Full-text editing, attachments, review queue, graph and bounded canvas, JSON/ZIP import/export |
-| Global memory | Domain-scoped global memories, evidence, review queue and the separate curator |
+| Global memory | Domain-scoped global memories, evidence, review queue and the separate curator; every row can jump to the wall |
 | Prompts | General / RP / Coding / Mixed variants, overrides and per-key reset; new conversations start on RP and can pick a domain in the first-run prompt |
 | Stats | Daily activity, hit distribution, health, never-recalled shortlist, 7-day write/read ratio |
 
@@ -296,7 +326,8 @@ the idle window starts from real message activity or the agent's return to idle.
 - **No cloud sync, no accounts, no cross-device sync.** This is a local plugin with no server-side identity.
 - **Images without vision:** without an optional caption/vision endpoint, an image memory is a pointer plus what a person wrote about it. The plugin does not claim to see pixels.
 - **Poisoned v4 sessions:** historical `format v4 message requires a producer-owned source kind` rows can be diagnosed but not rewritten; the plugin never edits private session files.
-- **Agent presets:** a conversation whose preset (`i-am-yuike`, `yuike-standard`, …) carries its own `cordis.yml` composes its model-facing plugins from that list alone, and its scope never delivers `agent/pre-step` to this plugin. At runtime `lib/preset-scope.js` attaches a second injection listener to that agent's own context, so **anchor and recall delivery works**; because a preset persona usually sets `complete: true`, system-prompt sections are dropped and the memory rules are delivered once as a `<memory-guide>` plugin message on the first step instead. The **final tool list** is a third, separate question: it follows the preset's own plugin list and the host scope, and it cannot be inferred from whether this plugin is listed — observations in this repo record both "no `memu_*` tools" and `memu_private_*` present in some presets' tool chains. **Per-preset tool availability still needs a real-host check (W3)**; until then do not read this as "presets have no tools", nor as "listing the plugin is enough". The plugin never edits anyone's preset files.
+- **Agent presets:** a conversation whose preset (`i-am-yuike`, `yuike-standard`, …) carries its own `cordis.yml` composes its model-facing plugins from that list alone, and its scope never delivers `agent/pre-step` to this plugin. At runtime `lib/preset-scope.js` attaches an injection listener to that agent's own context, so **anchor and recall delivery works**. All conversations now receive the memory rules as one `<memory-guide rules="…">` plugin message on the first step of a turn; the plugin no longer registers a system-prompt section, and the message is replaced after a domain/prompt change or a compaction (`rulesKey` / `guideLost`). The **final tool list** is a third, separate question: it follows the preset's own plugin list and the host scope, and it cannot be inferred from whether this plugin is listed — observations in this repo record both "no `memu_*` tools" and `memu_private_*` present in some presets' tool chains. **Per-preset tool availability still needs a real-host check (W3)**; until then do not read this as "presets have no tools", nor as "listing the plugin is enough". The plugin never edits anyone's preset files.
+- **Consolidation worker (C 包, 2026-10-30).** An idle or manual pass now works from a *fixed snapshot*: it takes the public `sessionQuery.observeSession()` lease (falling back to `readSession()` when the host offers no lease), cuts at the last complete `turn/end`, and drops unpaired tool calls and half steps — a live transcript whose last turn is still open is **deferred**, never consolidated. A job is settled only by a structured completion receipt that is cross-checked against the library index (an id the library does not contain is not a success, and a rewrite of an existing id counts as an update, not a new memory). Checked / pending / in-flight ranges and a lease are persisted separately in `consolidation.json`, so a crash, a timeout or a failed turn leaves the range pending instead of looking consolidated. **Limits:** the worker still runs inside the parent preset's own system prompt — it receives an explicit "you are a background consolidation worker, not this character" brief, which reduces but does not remove persona carry-over, because the host has no public API to replace that prompt; and when the conversation's tool surface carries no `memu_private_remember` / `memu_private_update` (or exposes memory only through PTC), consolidation reports `unavailable` (`worker-tools-missing` / `worker-tools-ptc-only`) rather than forking a worker that could not write. The parent's library, scope, domain and memory context are re-checked before dispatch, on every worker tool call, and again before any result is credited.
 - **"Tomorrow" is resolved at write time.** A relative day (今天 / 明天 / 后天 / today / tomorrow) together with a month-day settles into one dated memory; a memory that only says "sometime next month" keeps no date rather than inventing one. A date that was already written as `MM-DD` is not rewritten — the fix applies to new writes.
 - **Not in scope:** cloud MemU integration, source backfill for old entries, automatic repair of already-written v4 corruption.
 
@@ -320,6 +351,8 @@ The suite includes one real-engine integration case that is skipped unless `DSH_
 
 ## Changelog
 
+**1.2.1 (2026-10-31)**: fact-level dedup (write-side gate + recall-side cluster folding); consolidation now works in PTC (run_code) sessions; sidebar consolidation-failure alerts (cross-session, clearable, jump-to-library); the "currently in vector store" count follows real additions and deletions; private/global groups and multi-select bulk ops on the memory wall; one-click view from the global page; graph delete cascades with orphan-entity cleanup; remote consolidation counts only what actually landed.
+
 **1.1.3 (2026-10-30)**: managed local vectors are usable end to end (runtime/model download, switch-and-rebuild, progress with per-item failures); activity log with a Logs page; consolidation results move under the sidebar button; archive/restore and list refresh drop from seconds to milliseconds; a date written in the text is picked up when the field is empty; default topK 5/5; fixes the blank settings/WebUI panel (React #310).
 
 **1.1.2 (2026-10-29)**: anchor and recall delivery in agent-preset conversations; startup catch-up consolidation; panel TDZ crash fix.

package/README.zh.md

@@ -23,14 +23,14 @@
 
 ---
 
-`dsh-memu-memory` 面向长篇 RP 与陪伴对话，给 DSH 一份「像记忆」的记忆，而不是一本日志：三种留存层会淡；首轮有锚点，之后每个用户回合最多召回一次；来源不能被悄悄从「我推断」升级成「对方说过」；互相矛盾的记忆会挂起等待裁决，而不是两条都被当成真的；关系可以是一条带证据的边。说话人、角色、世界线与剧情分支构成硬边界，角色「知道什么」与实际发生过什么分开保存。
+`dsh-memu-memory` 面向长篇 RP 与陪伴对话，给 DSH 一份「像记忆」的记忆，而不是一本日志：三种留存层会淡；首轮有锚点，之后每个用户回合最多召回一次；来源不能被悄悄从「我推断」升级成「对方说过」；互相矛盾的记忆会挂起等待裁决，而不是两条都被当成真的；几乎相同的事实只入一条、召回时同簇只占一席；关系可以是一条带证据的边。说话人、角色、世界线与剧情分支构成硬边界，角色「知道什么」与实际发生过什么分开保存。
 
 默认**本地优先**。`agent` 后端把完整正文放在 JSON 索引里、用词法检索——不需要 Python、embedding 服务或 API key。`api` 后端才把检索交给可选的、锁定版本的 MemU 引擎（向量）。MemU 是「想要就用」的后端，不是插件能否启动的前提。
 
 > [!WARNING]
 > **早期版本提示。** 本插件仍处于早期阶段，可能存在无法预见的错误、bug 或不兼容变更。请先备份数据；欢迎[提交 Issue](https://github.com/Tkingxiao/dsh-memu-memory/issues) 或 PR 帮助改进。
 
-> **状态：** 1.1.0。首个公开发布版本为 1.0.0。细节文档已迁到 [`docs/`](docs/)，[`AGENTS.md`](AGENTS.md) 只做索引。升级前请备份 `~/.dsh/dsh-memu-memory`；如果要存无法重建的数据，请先读[兼容性与边界](#兼容性与边界)。
+> **状态：** 1.2.1。首个公开发布版本为 1.0.0。细节文档已迁到 [`docs/`](docs/)，[`AGENTS.md`](AGENTS.md) 只做索引。升级前请备份 `~/.dsh/dsh-memu-memory`；如果要存无法重建的数据，请先读[兼容性与边界](#兼容性与边界)。
 
 ## 界面
 
@@ -59,7 +59,7 @@
 - **三层，一条时间线。** `short_term` 草稿默认 72 小时到期；`long_term` 是记忆主体；`must_remember` 需要理由与来源，按 365 天半衰并带 0.35 强度地板，可以**边缘化**，但不会为腾名额被自动删除。
 - **召回是一个瞬间，不是一次倾倒。** 新会话或压缩后的第一个回合得到锚点块（pin / core / global 分段预算）；之后每个用户回合最多一次召回。去重与 token 预算由代码执行，不交给提示词自觉。
 - **日期要么是纪念日，要么是一天。** `MM-DD` 每年复发（生日）；`YYYY-MM-DD` 在前一天和当天各响一次，之后就算用过了。说「明天」的记忆存成带年份的日期，不会变成每年都冒出来的周年。
-- **记忆会不均匀地淡。** 新近、访问次数、峰终、首因、习惯化、闪光灯记忆的准确度共同参与强度；唤醒度与情绪同调是**条件项**而不是恒常生效——默认关闭，只有情绪离平缓够远时（|valence| >= 0.15，或 |arousal - 0.35| >= 0.1）才打开，平缓对话里两项乘数恒为 1。一条生动但可能记错的记忆会显示「细节未必准」，而不是假装确定。
+- **记忆会不均匀地淡。** 新近、访问次数、峰终、首因、习惯化、闪光灯记忆的准确度共同参与强度。情绪用四项构成（joy / anger / sorrow / fear，和为 1）加唤醒度 `arousal`（强度）与掌控感 `dominance` 描述；「嫉妒」这类复合情绪是四项的具名混合比例，不是新维度。情绪只影响召回排序：不进注入文本，也不改记忆正文。门控是三条独立条件的 OR——主导项占比 ≥ 0.35、`|arousal − 0.35| ≥ 0.1`、`|dominance| ≥ 0.25`；三条都不满足时两项乘数恒为 1。旧条目没有 emotion 时继续走原来的 valence 路径，不从旧 valence 反推四轴。一条生动但可能记错的记忆会显示「细节未必准」，而不是假装确定。
 - **来源是事实，不是文风。** 每条记忆带 `source`（`user_said` / `inferred` / `imagined` / `told`）、`confidence`、`accuracy` 与说出口径。模糊的推断不能冒充原话。
 - **矛盾先挂起。** 与已有记忆同题改写或说法相反的新条进入 `pending_review`：不注入、不召回，只出现在复核队列，等人或整理作业裁决。
 - **秘密不进自动召回。** `secret`、群聊频道、角色、世界线、剧情分支与 `knownBy` 共同约束读写；`secret` 永远不会被自动注入，但同文件夹里**主动检索仍然读得到**——它挡的是「自己冒出来」，不是保险箱。真正把同一文件夹里的两段对话隔开的是 `context` 作用域（`channelId` / `characterId` / `worldId` / `branchId`）；两边都没填就是同一个作用域，这正是「文件夹」的含义。旧的 `visibility: 'session'` 级别已删除：读到就按 `shared` 处理、`context` 原样保留，仍存着该级别的库用 `node scripts/migrate-visibility.mjs` 重写（`--apply` 才写，先备份）。
@@ -67,25 +67,55 @@
 - **图片是指针加描述，不是假装能看图。** 附件按 SHA-256 去重、有配额、分三级留存；没有描述时只记「收到一张图片」，绝不编造画面。
 - **整理在后台。** 整理跑在种子 fork 子会话里，不会在你的待发区多出一轮；失败退避重试，不伪记成功。
 
+### 主动记忆：实测触发与月开销
+
+空闲整理不是“每个安静会话都 fork”。进 fork 前会先给上次游标之后的证据打分：真人回合 `H`、内容量 `C`（真人文字 + 助手正文 + 去重后的成功工具结果，单结果最多 2000 字符）、有效工具工作单元 `U`，`score = 0.35*min(H/3,2) + 0.45*min(C/4000,2) + 0.20*min(U/20,2)`，默认门槛 0.8。空闲 15 分钟时，只有小会话能凭饥饿直接唤醒；2000 条事件以上的 transcript 必须分数达标，因为宿主会种完整前缀。近期整理过或注入过记忆的会话不 fork。
+
+整理不是越勤越好：大 transcript 的薄回合会被守卫跳过（同一批用例上约 −37% prompt token，代价是被跳过的那几个用例里约 1/22 条标签晚一点进记忆），但待整理区间超过 600 事件时无条件是检查一次，避免长会话早期的决定永远进不了记忆。
+
+用本机 4 个真实长会话做只读回放（不开新对话），统计合格回合：
+
+| 会话 | 事件数 | 回合数 | 分数触发 | 未达分数 | 游标后新真人回合 |
+|---|---:|---:|---:|---:|---:|
+| `62056bff…` | 5,913 | 12 | 6 | 6 | 0 |
+| `adedb9ac…` | 5,766 | 23 | 16 | 7 | 0 |
+| `e38a0de0…` | 85 | 2 | 0 | 2 | 0 |
+| `edf7dc5c…` | 78 | 4 | 0 | 4 | 0 |
+
+两条小会话的未达分数回合仍可能因饥饿唤醒；2000 条事件以上的大会话不会。四个会话在真实游标之后都没有新的真人输入，因此不会立即再 fork。
+
+从真实 fork 的 provider usage 汇总：一次整理约 **4.1 万~21.7 万 input tokens**（含缓存读，取决于记忆工具步数）+ **0.8k~3.4k output**。按每天一次 4 小时长会话、每天有效整理 2~6 次：
+
+| 每天整理次数 | 每月 input | 每月 output |
+|---|---:|---:|
+| 2 次 | ~5.6M | ~0.11M |
+| 4 次 | ~11M | ~0.23M |
+| 6 次 | ~17M | ~0.34M |
+| 6 次（都按实测最贵的单次） | ~39M | ~0.7M |
+
+作为参照，今天这棵 4 小时会话树实测 132M input + 1.28M output/日，主动记忆整理约占 **0.2%~1%**。以上是 token 量，不是金额；缓存读通常按更低单价计费。实际取决于回合数、记忆工具步数和会话真正空闲的次数。
+
 ## 一个 WebUI
 
 侧边栏只留**一个**入口：**MemU WebUI**。记忆墙与全部设置在同一套导航内，支持深链，如 `#/settings/memories`、`#/settings/folders`、`#/settings/stats`。
 
+**私有 / 全局分组与多选。** 「MemU 记忆」页按 **私有 / 全局** 分块（默认私有），私有块支持**多选**：批量归档、恢复、彻底删除（两段确认），一条失败不影响其余成果。「全局记忆」页每条记忆带「查看此记忆」，点击直达 MemU 记忆页对应条目。
+
 **原地操作。** 点记忆卡片会从右侧滑出详情抽屉：完整正文，以及决定它行为的字段——标题、正文、标签、说出口径、置信度、未了之事——外加保存、归档、恢复，以及需要重编辑时的「在记忆浏览中打开」。已归档的记忆可以直接在抽屉里删除：第一次点击只是准备，第二次才真正删除。全局记忆在抽屉里只读，并给出跳转。筛选条件与滚动位置保持不动。带图片的卡片会把第一张可用附件作为封面，抽屉里展开完整图片组；`secret` 与 `never_say` 条目不会拿到像素 URL，只显示标签占位。
 
 **关系图与邻接列表一起看。** 「关系」页签把关系画布与邻接列表并排显示（窄屏上下堆叠），焦点选择在上方，建边/实体/关系状态工具在下方；没有需要记得切换的模式。
 
-图谱使用本地打包的 D3 力导向布局，后台数据刷新保留节点位置与视角。支持拖拽节点、空白处平移、滚轮/双指缩放、适应全图与关联高亮；标签自动避让，设置中可调整斥力、连线距离，以及关闭标签或动态布局。图谱布局只表达已记录的关联，不会推断或写入新的关系。
+图谱使用本地打包的 D3 力导向布局，后台数据刷新保留节点位置与视角。支持拖拽节点、空白处平移、滚轮/双指缩放、适应全图与关联高亮；标签自动避让，设置中可调整斥力、连线距离，以及关闭标签或动态布局。图谱布局只表达已记录的关联，不会推断或写入新的关系。实体可以直接在图谱里删除（仍被引用时会明确拒绝）；永久删除一条记忆会连带清理指向它的边与孤立实体，孤立节点有启动扫描自动回收。
 
 | 页面 | 用途 |
 |---|---|
-| MemU 记忆 | 跨库记忆墙、来源/状态/库名/预览、原地详情、关系统一视图 |
+| MemU 记忆 | 私有/全局分组记忆墙、来源/状态/库名/预览、原地详情、多选批量操作、关系统一视图 |
 | 引擎与连接 | 可选 MemU 引擎的安装与切换、embedding 配置、备份恢复 |
 | 记忆设置 | 注入模式、token 预算、topK、相关度下限、留存与自动化 |
 | 文件夹 | 文件夹级私有/全局上限，展开对话修改各自选择 |
 | 记忆库 | 新建、改名、归档、恢复、显式删除 |
 | 记忆浏览 | 全文编辑、附件、复核池、有界图谱、JSON/ZIP 导入导出 |
-| 全局记忆 | 分域全局记忆、证据、复核队列与独立 curator |
+| 全局记忆 | 分域全局记忆、证据、复核队列与独立 curator，每条可跳转到记忆墙查看 |
 | 提示词 | 通用 / RP / Coding / Mixed 变体，覆盖与单项恢复；新会话默认 RP，并可在首次弹窗里直接选领域 |
 | 统计 | 每日活跃、命中分布、库健康度、最久没被想起的 5 条、7 天写读比 |
 
@@ -288,7 +318,7 @@ return { content: originalContent, meta: { ...originalMeta, memoryInbound: batch
 - **没有云同步、账号或跨设备同步。** 这是本地插件，没有服务端身份。
 - **没有视觉时：** 图片记忆只是「指针 + 人写的描述」。插件不会声称看过像素。
 - **v4 污染会话：** 历史 `format v4 message requires a producer-owned source kind` 行只能诊断，不能改写；插件不会编辑私有会话文件。
-- **人格预设（agent preset）：** 预设（`i-am-yuike`、`yuike-standard` 等）自带 `cordis.yml`，会话的模型可见插件集只由那份清单决定，它的 scope 也不会把 `agent/pre-step` 送到本插件。运行时 `lib/preset-scope.js` 会把第二个注入监听挂到该 agent 自己的 ctx 上，所以**锚点与召回的送达是能用的**；而 `persona` 通常带 `complete: true`、会把 system prompt 段整段丢弃，因此记忆规则改为首步一次性投放 `<memory-guide>` 插件消息。**最终工具面**是第三件事，由该预设自己的清单与宿主 scope 决定——不能从「有没有列上本插件」推断，仓库里的观察既有「没有 `memu_*` 工具」，也有某些预设的工具链里存在 `memu_private_*`。**逐预设的真实工具名单仍需真机核实（W3）**；在那之前，别读成「预设都没有工具」，也别读成「列上插件就一定可用」。插件不会改任何人的预设文件。
+- **人格预设（agent preset）：** 预设（`i-am-yuike`、`yuike-standard` 等）自带 `cordis.yml`，会话的模型可见插件集只由那份清单决定，它的 scope 也不会把 `agent/pre-step` 送到本插件。运行时 `lib/preset-scope.js` 会把注入监听挂到该 agent 自己的 ctx 上，所以**锚点与召回的送达是能用的**。所有会话现在都通过回合首步的一条 `<memory-guide rules="…">` 插件消息接收记忆规则；插件不再注册 system prompt 段，规则在切域、改稿或压缩后由 `rulesKey`/`guideLost` 替换。**最终工具面**是第三件事，由该预设自己的清单与宿主 scope 决定——不能从「有没有列上本插件」推断，仓库里的观察既有「没有 `memu_*` 工具」，也有某些预设的工具链里存在 `memu_private_*`。**逐预设的真实工具名单仍需真机核实（W3）**；在那之前，别读成「预设都没有工具」，也别读成「列上插件就一定可用」。插件不会改任何人的预设文件。
 - **相对日期在写入时结算：** 「今天 / 明天 / 后天 / today / tomorrow」加上一个月日落成带年份的日期；只说「下个月某天」的记忆不产生日期，也不替它编一个。已经按 `MM-DD` 写进去的旧条目不会被改写——这条只对新写入生效。
 - **不在范围：** 云端 MemU 联调、旧条目的 source 回填、已写坏 v4 消息的自动修复。
 
@@ -312,6 +342,8 @@ node --import ./tests/loader-register.mjs scripts/preview.mjs --port 7789
 
 ## 更新日志
 
+**1.2.1（2026-10-31）**：事实级去重（写入近似门 + 召回按簇折叠）；PTC（run_code）会话整理可用；侧边栏整理失败提醒（跨会话、可清除、可跳转）；「当前已入向量库」计数跟随真实增减；记忆墙私有/全局分组与多选批量操作；全局记忆一键查看；关系图删除级联与孤儿回收；整理写入计数说真话。
+
 **1.1.3（2026-10-30）**：受管本地向量可用（运行时/模型下载、切换重建、进度与失败明细）；活动日志与「日志」页；整理结果落到侧边栏按钮下方；归档/还原与列表刷新提速到毫秒级；日期漏填时从正文兜底；默认 topK 5/5；修复设置/WebUI 白屏。
 
 **1.1.2（2026-10-29）**：人格预设会话的锚点与召回送达；启动补扫空闲整理；面板 TDZ 崩溃修复。