memory-eternal 0.10.22 → 0.10.24

package/.claude-plugin/plugin.json

@@ -1,7 +1,7 @@
 {
   "$schema": "https://anthropic.com/claude-code/plugin.schema.json",
   "name": "memory-eternal",
-  "version": "0.10.22",
+  "version": "0.10.24",
   "description": "记忆核心（Memory Eternal）：对话结束后自动沉淀知识卡到本地记忆库（~/.dsh/memory-vault），支持检索 / 召回 / 审核中心。",
   "author": {
     "name": "EternalNight996",

package/.codex-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "memory-eternal",
-  "version": "0.10.22",
+  "version": "0.10.24",
   "description": "记忆核心（Memory Eternal）：对话结束后自动沉淀知识卡到本地记忆库（~/.dsh/memory-vault），支持检索 / 召回 / 审核中心。",
   "author": {
     "name": "EternalNight996",

package/.cursor-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "memory-eternal",
-  "version": "0.10.22",
+  "version": "0.10.24",
   "description": "记忆核心（Memory Eternal）：对话结束后自动沉淀知识卡到本地记忆库（~/.dsh/memory-vault），支持检索 / 召回 / 审核中心。",
   "author": {
     "name": "EternalNight996",

package/README.en.md

@@ -361,14 +361,16 @@ curl http://127.0.0.1:7999/memory-eternal/api/web-info    # is the standalone we
 
 | Version | Date | Highlights |
 |---|---|---|
-| **v0.10.22** | 2026-10-13 | **"Conflict scan": similar cards may be merged, conflicts must be judged by a human (first slice of issue #36)**. Until now `/optimize` merged **blindly** on similarity (≥0.55) — yet two semantically close cards with mutually exclusive conclusions score just as high (0.93 measured), so a blind merge quietly "reconciles" the conflict, and `mergeCards` **permanently deletes** the sources. New `lib/refine.js` plus a "organize → 🧭 Conflict scan" panel section: ① **wide recall** (≥0.30, upper bound lifted to 1.0, deliberately including the near-duplicate band — the most dangerous one for blind merging); ② the model only classifies each pair as "duplicate / complement / conflict / unrelated" and states the conflict in one line plus who to keep; ③ **the human decides**: keep A / keep B / keep both / merge; the **scan never writes**, and `keepA|keepB` only moves the yielding card to the **recycle bin** (recoverable); ④ rulings are recorded in `refine-decisions.json` so the same pair is not raised again; ⑤ **safety interlock**: pairs ruled conflicting are **excluded from one-click optimize** entirely. Also fixes one source of redundancy: skill-catalog injections (`- \`skill-name\`: …`) are no longer captured as junk cards (real examples already existed in the vault). API: `/refine`, `/refine-scan`, `/refine-apply`; CLI: `dsh-memory refine` (lists candidates without calling a model; `--apply` records a ruling). `npm test` now runs **329 checks** |
-| **v0.10.21** | 2026-10-13 | **"DB unavailable" goes from silent to visible (issue #33, scoped by review)**. ① **Health check at open**: the old code ran `PRAGMA journal_mode=WAL` and assumed success (that pragma does not throw on failure — it just returns the actual mode), so a read-only file or unwritable directory let the plugin run on silently; now the real journal mode is **read back** and a **write probe** runs (create then drop a table, in the main DB rather than temp), the verdict lands in `dbStatus()`, and an unwritable DB immediately logs the reason plus where to look. ② **`/overview` degrades instead of 500**: when the DB or directory is unavailable it no longer returns 500 (probes, the panel and scripts read 500 as "this port is broken / not our service", which historically started the restart storm and port drift); it now returns 200 with `degraded:true`, `dbError` and `db:{journalMode,writable,error}` while keeping the `vaultDir` / `version` identity fields — so "this is us, but the DB is broken" can finally be expressed separately. ③ **New `dsh-memory doctor`**: DB health self-check (real journal mode / write probe / backup list / recovery guidance) with `--json` for scripting; when the DB is unwritable or suspected corrupt it prints **guidance that never touches data** (stop writers → copy the scene aside → list backups → swap → re-verify). **Deliberately not done** (review conclusion, now guarded by tests so nobody adds it back casually): `busy_timeout` (`DatabaseSync` is a synchronous API, so waiting on a lock blocks the host event loop) and automatic WAL→DELETE downgrade (it **permanently** changes the DB mode, and a rollback journal has worse concurrency, which could worsen lock contention). `npm test` now runs **322 checks** |
+| **v0.10.24** | 2026-10-13 | **"Same goal, different approach" now clusters too (third and final slice of issue #36)**. The two previous versions covered "merge duplicates" and "judge conflicts", but the last piece — one purpose, several workflows/codebases — is **undetectable by body similarity** ("compress with 7z", "distribute as tar.gz", "slim the npm package" read very differently yet share a goal). Approach: **the model first writes one line per card stating what problem it solves, then we cluster by goal** — goals are cached in `refine-goals.json` (each card extracted once; a second scan makes zero calls) and clustered with union-find; hits are grouped as "one goal, N cards" with two actions: **keep all** (records pairwise rulings so it stops asking) or **merge this group** (via the existing /merge; the new card enters the audit queue). *Read-only throughout — the only write is the goal cache.* Entry points: the panel (organize → 🎯 same goal, different approach) and APIs `/refine-goals`, `/refine-goals-keep`; CLI `dsh-memory refine --goals --keep --path A --path B` (the clustering itself needs a model, run it in the panel). `npm test` now runs **332 checks** |
+| **v0.10.23** | 2026-10-13 | **Existing "injected junk cards" can be cleaned up in one click (second slice of issue #36)**. The previous version stopped such cards from being **created** (skill catalogs / `<system-reminder>` arriving as user messages were captured as knowledge after their tags were stripped), but the **backlog remained**: a scan of a real vault immediately found **32** cards unrelated to any conversation (catalog lines like `- \`code-review\`: Review the changes…`, fragments like `用户system-reminder-you-are-teammate-c…`). Adds `findInjectedCards()` (a **proposal only**, it changes nothing; criteria: injection tags present / catalog lines ≥30% of the body / under 200 characters left after stripping / a fragmentary title) and `purgeInjected()` (**goes to the recycle bin, recoverable**, reporting failures per card). Entry points: the panel (organize → 🧹 injected junk cards: scan, recycle one, recycle all) and the CLI `dsh-memory refine --junk [--purge --path P] [--json]`; API `/refine-junk`, `/refine-junk-purge`. `npm test` now runs **331 checks** |
 
 <details>
-<summary>Older release notes (v0.10.20 and earlier · 47 releases, click to expand)</summary>
+<summary>Older release notes (v0.10.22 and earlier · 49 releases, click to expand)</summary>
 
 | Version | Date | Highlights |
 |---|---|---|
+| **v0.10.22** | 2026-10-13 | **"Conflict scan": similar cards may be merged, conflicts must be judged by a human (first slice of issue #36)**. Until now `/optimize` merged **blindly** on similarity (≥0.55) — yet two semantically close cards with mutually exclusive conclusions score just as high (0.93 measured), so a blind merge quietly "reconciles" the conflict, and `mergeCards` **permanently deletes** the sources. New `lib/refine.js` plus a "organize → 🧭 Conflict scan" panel section: ① **wide recall** (≥0.30, upper bound lifted to 1.0, deliberately including the near-duplicate band — the most dangerous one for blind merging); ② the model only classifies each pair as "duplicate / complement / conflict / unrelated" and states the conflict in one line plus who to keep; ③ **the human decides**: keep A / keep B / keep both / merge; the **scan never writes**, and `keepA|keepB` only moves the yielding card to the **recycle bin** (recoverable); ④ rulings are recorded in `refine-decisions.json` so the same pair is not raised again; ⑤ **safety interlock**: pairs ruled conflicting are **excluded from one-click optimize** entirely. Also fixes one source of redundancy: skill-catalog injections (`- \`skill-name\`: …`) are no longer captured as junk cards (real examples already existed in the vault). API: `/refine`, `/refine-scan`, `/refine-apply`; CLI: `dsh-memory refine` (lists candidates without calling a model; `--apply` records a ruling). `npm test` now runs **329 checks** |
+| **v0.10.21** | 2026-10-13 | **"DB unavailable" goes from silent to visible (issue #33, scoped by review)**. ① **Health check at open**: the old code ran `PRAGMA journal_mode=WAL` and assumed success (that pragma does not throw on failure — it just returns the actual mode), so a read-only file or unwritable directory let the plugin run on silently; now the real journal mode is **read back** and a **write probe** runs (create then drop a table, in the main DB rather than temp), the verdict lands in `dbStatus()`, and an unwritable DB immediately logs the reason plus where to look. ② **`/overview` degrades instead of 500**: when the DB or directory is unavailable it no longer returns 500 (probes, the panel and scripts read 500 as "this port is broken / not our service", which historically started the restart storm and port drift); it now returns 200 with `degraded:true`, `dbError` and `db:{journalMode,writable,error}` while keeping the `vaultDir` / `version` identity fields — so "this is us, but the DB is broken" can finally be expressed separately. ③ **New `dsh-memory doctor`**: DB health self-check (real journal mode / write probe / backup list / recovery guidance) with `--json` for scripting; when the DB is unwritable or suspected corrupt it prints **guidance that never touches data** (stop writers → copy the scene aside → list backups → swap → re-verify). **Deliberately not done** (review conclusion, now guarded by tests so nobody adds it back casually): `busy_timeout` (`DatabaseSync` is a synchronous API, so waiting on a lock blocks the host event loop) and automatic WAL→DELETE downgrade (it **permanently** changes the DB mode, and a rollback journal has worse concurrency, which could worsen lock contention). `npm test` now runs **322 checks** |
 | **v0.10.20** | 2026-10-12 | **Fixing the three-layer "DB unavailable → the watchdog kills its own web → the port drifts to 8006" scene (issues #31 / #33, measured by AkiMei0721)**. ① **A non-200 `/overview` is not a dead port**: `/overview` touches the database, so a read-only / WAL-corrupted DB makes it 500, and the watchdog declared the port offline → `spawnWeb()` (which first SIGTERMs its own previous child) → **it killed and respawned its own healthy web, drifting the port from 7999 all the way to 8006 (six instances)**; a new pure `decideTickAction()` treats **a reachable TCP port as alive** (`api-degraded-but-listening`, resetting the restart counter), and the readiness check for a new instance likewise accepts "listening but `/overview` non-200". ② **The version probe now asks `/version-check`** (no database access; falls back to `/overview`) — the root symptom was "the port is clearly alive yet no version comes back", so verify reported "(no response)" and `status` even misjudged it as "a pre-v0.10.4 build". ③ **No more crash in restricted sandboxes**: the default `stdio:'pipe'` makes `spawn()` **throw EPERM synchronously** (the old code only guarded the `'error'` event, which never ran), so `dsh-memory watchdog --reap` crashed and process enumeration stayed permanently empty; `spawn` is now wrapped in `try/catch` and stdout goes through a **temp file**, touching no named pipes. ④ The stall check also gained a "minimum uptime" (30 minutes): a freshly restarted process necessarily has zero stops, so it must not report "turn-stop events are broken" (a false alarm was measured 1.5 minutes after a restart). Adds `tests/issue-fixes-8.test.mjs` (4 checks) plus a too-early case in `tests/stall.test.mjs`. `npm test` now runs **318 checks** |
 | **v0.10.19** | 2026-10-12 | **`UNSUPPORTED_REASONING_EFFORT` no longer kills the whole distillation (issue #34)**. The scene (zafir-zhong and others): many **gateways** claim reasoning-effort support in their capability list and only reject `reasoningEffort: 'off'` when it is actually sent (measured: workbuddy/hy3, FallbacksChain/Auto, new-api-resp/gpt-6-luna), so the entire candidate is marked failed and the pipeline moves to the next provider — auto-capture stops dead for those users (that FAIL shows up in the diagnostics of #37 / #38 / #40 too). The existing design only covered "the provider **says** it does not support it → do not send the field" (#15); the runtime half was missing. Fix: an in-process memo `markEffortUnsupported()\/isEffortUnsupported()` — on the first rejection the call **retries once without the field** (same candidate, same provider), remembers that `provider/model`, and from then on the first request for it carries no effort at all; the memo is **keyed by provider/model**, so other providers are untouched. Guards added in `tests/capture-route.test.mjs` (sent → rejected → retried without and got a card → later only one request → other providers unaffected). `npm test` now runs **313 checks** |
 | **v0.10.18** | 2026-10-12 | **"Save config" was being rejected unconditionally (issue #35)**. The scene (Dzifxazy): a fresh install, change nothing, hit save → **HTTP 400 "field must not be empty: secretHint"**, the whole submission rejected, so none of the numbers or toggles the user did change were stored; the wording also pointed troubleshooting at "some numeric box left blank". Root cause: the host-side pre-check treated **any empty string** as "numeric field not filled" and only exempted a hard-coded list (vaultDir / captureProvider / captureModel / recallEmbedding / activeVault) — while `secretHint`'s default **is** an empty string (`z.string().default('')`) and was not on that list; since the panel submits the whole form, every save hit it. Fix: new `emptyNumericFields(patch, Config.dict)` decides which keys were cleared by looking at the schema (`type === 'number'`), and the hard-coded list is gone (future nullable string fields need nobody to remember an exemption). Adds `tests/config-validate.test.mjs` (2 checks: empty strings only count for number fields — including a brand-new string field that appears on no list as a regression nail, plus a source guard against reintroducing the list). `npm test` now runs **312 checks** |

package/README.md

@@ -393,14 +393,16 @@ curl http://127.0.0.1:7999/memory-eternal/api/web-info    # 独立 web 是否活
 
 | 版本 | 日期 | 关键改动 |
 |---|---|---|
-| **v0.10.22** | 2026-10-13 | **「矛盾扫描」：相似可以合并，矛盾必须人判断（issue #36 第一刀）**。现状是 `/optimize` 只按相似度（≥0.55）**盲合并** —— 而两张语义相近、结论互斥的卡相似度同样很高（实测 0.93），盲合并会把冲突悄悄「和解」掉，且 `mergeCards` 会**永久删除**源卡。新增 `lib/refine.js` + 客户端「整理 → 🧭 矛盾扫描」：① **宽召回**（≥0.30，上限放到 1.0，特意把近重复那一段纳入 —— 那正是盲合并最危险的区间）；② 模型只判「重复 / 补充 / 矛盾 / 无关」并给一句冲突点 + 建议保留谁；③ **人拍板**：保留 A / 保留 B / 都保留 / 合并；**扫描绝不写库**，`keepA|keepB` 只把让位那张送**回收站**（可还原）；④ 决策记进 `refine-decisions.json`，同一对不再重复打扰；⑤ **安全联锁**：被判为矛盾的对**绝不参与一键优化**的盲合并。另修一处冗余来源：技能目录（`- \`skill-name\`: …`）这类注入文本不再被抓成垃圾卡（实测库里已有这种卡）。API：`/refine`、`/refine-scan`、`/refine-apply`；CLI：`dsh-memory refine`（列候选不调模型，`--apply` 落裁决）。`npm test` 共 **329 项** |
-| **v0.10.21** | 2026-10-13 | **「库不可用」从「静默运行」变成「看得清」（issue #33，按评审裁剪范围）**。① **打开即体检**：老实现执行完 `PRAGMA journal_mode=WAL` 就当成功（这条 PRAGMA 失败时**不抛错**，只把真实模式当一行返回），文件只读 / 目录不可写时插件会一路静默跑下去；现在打开后**读回真实 journal 模式 + 真写一次探针**（建表→删表，写主库而非 temp 库），结论记进 `dbStatus()`，不可写时立即把原因与排查方向打到日志。② **`/overview` 降级返回**：库/目录不可用时不再吐 500（500 会被探针、面板、脚本读成「这个端口坏了 / 不是我们的服务」，历史上正是重启风暴与端口漂移的入口），改成 200 + `degraded:true` + `dbError` + `db:{journalMode,writable,error}`，并保住 `vaultDir` / `version` 身份字段 —— 「这是我们、但库有问题」从此可以被分开表达。③ **新增 `dsh-memory doctor`**：库健康自检（真实 journal 模式 / 写探针 / 备份清单 / 恢复引导），带 `--json` 可脚本化；不可写或疑似损坏时给出**只引导、不动数据**的步骤（先停写者 → 复制现场留证 → 查备份 → 替换 → 复核）。**刻意不做**（评审结论，已写成测试守卫防止顺手加回）：`busy_timeout`（`DatabaseSync` 是同步 API，撞锁会阻塞宿主事件循环）与 WAL→DELETE 自动降级（会**永久**改变库模式，且 rollback journal 并发更差、可能反过来加剧锁竞争）。`npm test` 共 **322 项** |
+| **v0.10.24** | 2026-10-13 | **「同目标、不同做法」也能归到一起了（issue #36 第三块，闭环）**。前两版解决了「冗余合并」与「矛盾判别」，但最后一块 —— 同一目的、不同工作流/代码 —— 按正文相似度**根本抓不到**（「用 7z 压缩」「用 tar.gz 分发」「npm 包瘦身」字面差很远、目标却一样）。做法：**先让模型给每张卡抽一句「这张卡解决什么问题」，再按目标聚类** —— 目标抽取结果缓存在 `refine-goals.json`（同一张卡只抽一次，第二次扫描 0 次调用），聚类用并查集；命中后按「同一目标 N 张卡」成组给出，动作二选一：**都保留**（记两两决策，下次不再提示）或**合并这一组**（走现有 /merge，新卡进审核队列）。*全程只读，唯一写入是目标缓存*。入口：面板「整理 → 🎯 同目标不同做法」+ API `/refine-goals`、`/refine-goals-keep`；CLI `dsh-memory refine --goals --keep --path A --path B`（聚类本身需要模型，在面板跑）。`npm test` 共 **332 项** |
+| **v0.10.23** | 2026-10-13 | **存量「注入垃圾卡」可一键清理（issue #36 第二刀）**。上一版修掉了**新生成**的注入垃圾卡（技能目录 / `<system-reminder>` 被当 user 消息 → 剥标签后仍被当知识沉淀），但**存量还在**：在真实库里一跑就找出 **32 张**「和对话毫无关系」的卡（`- \`code-review\`: Review the changes…` 这种目录行、`用户system-reminder-you-are-teammate-c…` 这种碎片）。新增 `findInjectedCards()`（只**提议**、不动数据；判据：含注入标签 / 技能目录条目占比 ≥30% / 剥离注入后只剩 <200 字 / 标题是碎片）与 `purgeInjected()`（**走回收站、可还原**，逐个报告失败）。入口：面板「整理 → 🧹 注入垃圾卡」（扫描存量 / 单张回收 / 全部回收）+ CLI `dsh-memory refine --junk [--purge --path P] [--json]`；API `/refine-junk`、`/refine-junk-purge`。`npm test` 共 **331 项** |
 
 <details>
-<summary>更早的更新日志（v0.10.20 及以前 · 共 47 版，点开查看）</summary>
+<summary>更早的更新日志（v0.10.22 及以前 · 共 49 版，点开查看）</summary>
 
 | 版本 | 日期 | 关键改动 |
 |---|---|---|
+| **v0.10.22** | 2026-10-13 | **「矛盾扫描」：相似可以合并，矛盾必须人判断（issue #36 第一刀）**。现状是 `/optimize` 只按相似度（≥0.55）**盲合并** —— 而两张语义相近、结论互斥的卡相似度同样很高（实测 0.93），盲合并会把冲突悄悄「和解」掉，且 `mergeCards` 会**永久删除**源卡。新增 `lib/refine.js` + 客户端「整理 → 🧭 矛盾扫描」：① **宽召回**（≥0.30，上限放到 1.0，特意把近重复那一段纳入 —— 那正是盲合并最危险的区间）；② 模型只判「重复 / 补充 / 矛盾 / 无关」并给一句冲突点 + 建议保留谁；③ **人拍板**：保留 A / 保留 B / 都保留 / 合并；**扫描绝不写库**，`keepA|keepB` 只把让位那张送**回收站**（可还原）；④ 决策记进 `refine-decisions.json`，同一对不再重复打扰；⑤ **安全联锁**：被判为矛盾的对**绝不参与一键优化**的盲合并。另修一处冗余来源：技能目录（`- \`skill-name\`: …`）这类注入文本不再被抓成垃圾卡（实测库里已有这种卡）。API：`/refine`、`/refine-scan`、`/refine-apply`；CLI：`dsh-memory refine`（列候选不调模型，`--apply` 落裁决）。`npm test` 共 **329 项** |
+| **v0.10.21** | 2026-10-13 | **「库不可用」从「静默运行」变成「看得清」（issue #33，按评审裁剪范围）**。① **打开即体检**：老实现执行完 `PRAGMA journal_mode=WAL` 就当成功（这条 PRAGMA 失败时**不抛错**，只把真实模式当一行返回），文件只读 / 目录不可写时插件会一路静默跑下去；现在打开后**读回真实 journal 模式 + 真写一次探针**（建表→删表，写主库而非 temp 库），结论记进 `dbStatus()`，不可写时立即把原因与排查方向打到日志。② **`/overview` 降级返回**：库/目录不可用时不再吐 500（500 会被探针、面板、脚本读成「这个端口坏了 / 不是我们的服务」，历史上正是重启风暴与端口漂移的入口），改成 200 + `degraded:true` + `dbError` + `db:{journalMode,writable,error}`，并保住 `vaultDir` / `version` 身份字段 —— 「这是我们、但库有问题」从此可以被分开表达。③ **新增 `dsh-memory doctor`**：库健康自检（真实 journal 模式 / 写探针 / 备份清单 / 恢复引导），带 `--json` 可脚本化；不可写或疑似损坏时给出**只引导、不动数据**的步骤（先停写者 → 复制现场留证 → 查备份 → 替换 → 复核）。**刻意不做**（评审结论，已写成测试守卫防止顺手加回）：`busy_timeout`（`DatabaseSync` 是同步 API，撞锁会阻塞宿主事件循环）与 WAL→DELETE 自动降级（会**永久**改变库模式，且 rollback journal 并发更差、可能反过来加剧锁竞争）。`npm test` 共 **322 项** |
 | **v0.10.20** | 2026-10-12 | **「库不可用 → 看门狗自杀自拉 → 端口漂到 8006」三层现场一起修（issue #31 / #33，AkiMei0721 实测）**。① **`/overview` 非 200 ≠ 端口死了**：`/overview` 会碰数据库，库只读 / WAL 损坏时它返回 500，而看门狗据此判「离线」→ 走 `spawnWeb()`（开头先 SIGTERM 掉自己的上一个 child）→ **自杀自拉、端口从 7999 一路漂到 8006（6 个实例）**；新增 `decideTickAction()` 纯决策：只要 **TCP 还能连**就视为存活（`api-degraded-but-listening`，并重置重启计数），新实例的就绪判定同样接受「已监听但 `/overview` 非 200」。② **版本探测改走 `/version-check`**（不碰库；拿不到才回落 `/overview`）—— 病根是「端口上明明活着、版本却拿不到」，于是 verify 报「（无响应）」、`status` 甚至误判成「v0.10.4 之前的旧版」。③ **受限沙箱里不再崩**：默认 `stdio:'pipe'` 会在 `spawn()` 时**同步抛 EPERM**（老实现只在 `'error'` 事件里兜底，根本没机会执行）→ `dsh-memory watchdog --reap` 直接崩、进程枚举永远空名单；现在 `try/catch` 包住 spawn，并**用临时文件接 stdout**、完全不碰命名管道。④ 顺带给 stall 判据加「最短运行时长（30 分钟）」：刚落进程必然 0 收尾，不该据此报「收尾事件失效」（实测重启 1.5 分钟后就误报过）。新增 `tests/issue-fixes-8.test.mjs`（4 项）+ `tests/stall.test.mjs` 的 too-early 用例。`npm test` 共 **318 项** |
 | **v0.10.19** | 2026-10-12 | **`UNSUPPORTED_REASONING_EFFORT` 不再让整条蒸馏熔断（issue #34）**。现场（zafir-zhong 等）：很多**网关**在能力自述里说支持 reasoning effort，真把 `reasoningEffort: 'off'` 发出去才回 `UNSUPPORTED_REASONING_EFFORT`（实测：workbuddy/hy3、FallbacksChain/Auto、new-api-resp/gpt-6-luna），于是整个候选被判失败、换下一个 provider —— 用户的自动沉淀直接停摆（#37 / #38 / #40 的诊断信息里都能看到这条 FAIL）。既有设计只覆盖「provider **自述**不支持 → 不传该字段」（#15），缺了运行时这一半。修法：新增进程内记忆 `markEffortUnsupported()\/isEffortUnsupported()` —— 首次被拒时**不带该字段重试一次**（同一候选、同一 provider），并记住这个 `provider/model`，此后同一进程内它第一次请求就不再带 effort；记忆**按 provider/model 分开**，不误伤别的 provider。回归守卫加在 `tests/capture-route.test.mjs`（第一次带 → 被拒 → 第二次不带并拿到卡 → 后续只发一次 → 别的 provider 不受影响）。`npm test` 共 **313 项** |
 | **v0.10.18** | 2026-10-12 | **「保存配置」被无条件拒绝的修复（issue #35）**。现场（Dzifxazy）：全新安装、什么都不动、直接点保存 → **HTTP 400「字段不能为空：secretHint」**，整单被拒，用户改的数字与开关**一个都存不进去**；文案还会把排查方向误导到「某个数字格子没填」。根因：宿主侧的前置校验把**任何空字符串**都当成「数字没填」，只豁免一份硬编码名单（vaultDir / captureProvider / captureModel / recallEmbedding / activeVault）—— 而 `secretHint` 的默认值**就是**空串（`z.string().default('')`）又不在名单里；面板提交的是整张表单，于是每次保存必然命中。修法：新增 `emptyNumericFields(patch, Config.dict)`，**按 schema 的 `type === 'number'` 判定**「哪些键被清空」，彻底删掉硬编码名单（以后新增可空字符串字段不需要谁记得去补）。新增 `tests/config-validate.test.mjs`（2 项：空串只对 number 字段报「没填」——里面放了一个「不在任何名单里」的新字符串字段做回归钉子；源码守卫不许退回名单）。`npm test` 共 **312 项** |

package/bin/dsh-memory.mjs

@@ -12,6 +12,8 @@
 //   dsh-memory sweep <dir>                                挖掘 Claude Code 会话 JSONL
 //   dsh-memory doctor [--vault DIR] [--json]              库健康自检（journal 模式 / 写探针 / 备份清单 / 恢复引导）
 //   dsh-memory refine [--vault DIR] [--json]              矛盾扫描候选（不调模型）；--apply --a A --b B --action keepA|keepB|merge|keepBoth 落裁决
+//   dsh-memory refine --junk [--purge --path P] [--json]   注入垃圾卡（技能目录等注入文本沉淀成的卡）：列清单 / 送回收站
+//   dsh-memory refine --goals [--keep --path A --path B]   同目标不同做法：聚类需模型（面板里跑）；--keep 落"都保留"
 //
 // 环境变量：
 //   MEMORY_VAULT_DIR       vault 目录（默认 ~/.dsh/memory-vault）
@@ -170,6 +172,46 @@ async function main() {
       const { defaultVaultDir } = await import('../lib/capture-run.js')
       const { refineCandidates, refineApply, readDecisions } = await import('../lib/refine.js')
       const root = argOf('--vault') ? path.resolve(argOf('--vault')) : defaultVaultDir()
+      if (has('--goals')) {
+        // 「同目标、不同做法」聚类：需要模型（CLI 里通常拿不到 llm）→ 提示去面板；--keep 可落"都保留"
+        const { goalScan, markClusterKept } = await import('../lib/refine.js')
+        const paths = []
+        for (let i = 0; i < argv.length; i++) if (argv[i] === '--path' && argv[i + 1]) paths.push(argv[i + 1])
+        if (has('--keep')) {
+          const out = await markClusterKept(root, { paths })
+          if (has('--json')) { console.log(JSON.stringify(out, null, 2)); if (!out.ok) process.exitCode = 1; return }
+          if (out.ok) console.log('已记录「都保留」：' + out.recorded + ' 对')
+          else { console.error('失败：' + out.error); process.exitCode = 1 }
+          return
+        }
+        const out = await goalScan(root, { llm: null, limit: Number(argOf('--limit')) || 30 })
+        if (has('--json')) { console.log(JSON.stringify(out, null, 2)); if (!out.ok) process.exitCode = 1; return }
+        console.error('带模型的目标聚类请在 DSH 面板跑：记忆 → 整理 → 🎯 同目标不同做法')
+        console.error('（CLI 可用 --goals --keep --path A --path B 把一组标成"都保留"）')
+        process.exitCode = 1
+        return
+      }
+      if (has('--junk')) {
+        // 存量「注入垃圾卡」：--junk 列清单；--junk --purge --path <p>（可重复）送回收站
+        const { findInjectedCards, purgeInjected } = await import('../lib/refine.js')
+        if (has('--purge')) {
+          const paths = []
+          for (let i = 0; i < argv.length; i++) if (argv[i] === '--path' && argv[i + 1]) paths.push(argv[i + 1])
+          const out = await purgeInjected(root, { paths })
+          if (has('--json')) { console.log(JSON.stringify(out, null, 2)); if (!out.ok) process.exitCode = 1; return }
+          console.log('已送回收站 ' + out.removed.length + ' 张（可从回收中心还原）' + (out.failed.length ? '，失败 ' + out.failed.length + ' 张' : ''))
+          for (const x of out.failed) console.error('  ✗ ' + x.path + '：' + x.error)
+          if (!out.ok) process.exitCode = 1
+          return
+        }
+        const items = await findInjectedCards(root)
+        if (has('--json')) { console.log(JSON.stringify({ ok: true, count: items.length, items }, null, 2)); return }
+        console.log('注入垃圾卡（提议回收，未动数据）：' + items.length + ' 张')
+        for (const it of items) console.log('  ' + it.path + '  —— ' + it.reasons.join('；'))
+        console.log('')
+        console.log('回收（进回收站，可还原）：dsh-memory refine --junk --purge --path <路径> [--path <路径>…]')
+        return
+      }
       if (has('--apply')) {
         const out = await refineApply(root, {
           a: argOf('--a'), b: argOf('--b'), action: argOf('--action'),

package/lib/api.js

@@ -603,6 +603,51 @@ export function createApi(deps = {}) {
         json(res, out.ok ? 200 : 400, out)
         return
       }
+      case '/refine-goals': {
+        // 「同目标、不同做法」聚类（#36 第三块）：先让模型给每张卡抽一句「解决什么问题」（带缓存），
+        // 再按**目标**聚类 —— 正文相似度抓不到这种关系（7z 压缩 / tar 分发 / npm 瘦身 字面差很远）。
+        if (req.method !== 'POST') { json(res, 405, { ok: false, error: '需 POST' }); return }
+        const { goalScan } = await import('./refine.js')
+        let opts = {}
+        try { let raw = ''; for await (const chunk of req) raw += chunk; opts = JSON.parse(raw || '{}') } catch { opts = {} }
+        const out = await goalScan(vaultRoot, {
+          llm: getLlm(),
+          limit: Number(opts.limit) || 30,
+          threshold: Number(opts.threshold) || 0.55,
+          provider: opts.provider,
+          model: opts.model,
+        })
+        json(res, out.ok ? 200 : 400, out)
+        return
+      }
+      case '/refine-goals-keep': {
+        // 把一组「同目标」标成都保留（记两两决策，避免下次重复提示）
+        if (req.method !== 'POST') { json(res, 405, { ok: false, error: '需 POST' }); return }
+        const { markClusterKept } = await import('./refine.js')
+        let payload = {}
+        try { let raw = ''; for await (const chunk of req) raw += chunk; payload = JSON.parse(raw || '{}') } catch { return json(res, 400, { ok: false, error: 'JSON 解析失败' }) }
+        const out = await markClusterKept(vaultRoot, payload)
+        json(res, out.ok ? 200 : 400, out)
+        return
+      }
+      case '/refine-junk': {
+        // 注入垃圾卡的**提议清单**（只读）：正文其实是被注入的运行时文本（技能目录 / system-reminder…）。
+        // 新版本已不再生成这类卡，这里负责把**存量**找出来，由人决定是否回收。
+        const { findInjectedCards } = await import('./refine.js')
+        const items = await findInjectedCards(vaultRoot)
+        json(res, 200, { ok: true, count: items.length, items })
+        return
+      }
+      case '/refine-junk-purge': {
+        // POST {paths:[...]}：送**回收站**（可还原），不是永久删除
+        if (req.method !== 'POST') { json(res, 405, { ok: false, error: '需 POST' }); return }
+        const { purgeInjected } = await import('./refine.js')
+        let payload = {}
+        try { let raw = ''; for await (const chunk of req) raw += chunk; payload = JSON.parse(raw || '{}') } catch { return json(res, 400, { ok: false, error: 'JSON 解析失败' }) }
+        const out = await purgeInjected(vaultRoot, payload)
+        json(res, out.ok ? 200 : 400, out)
+        return
+      }
       case '/setup-run': {
         // 「补全 MCP」：真正执行 runSetup（写外部 agent 配置）。共享层（独立 web 也可用）
         if (req.method !== 'POST') return json(res, 405, { ok: false, error: '需 POST' })