claude-mem-lite 4.0.3 → 5.0.0

package/.claude-plugin/marketplace.json

@@ -10,7 +10,7 @@
   "plugins": [
     {
       "name": "claude-mem-lite",
-      "version": "4.0.3",
+      "version": "5.0.0",
       "source": "./",
       "description": "Persistent long-term memory for Claude Code via MCP — captures coding decisions, bugfixes, and context across sessions. Hybrid FTS5 + TF-IDF search with episode batching. Single SQLite DB, no external services. A lighter, lower-cost alternative to claude-mem (episode batching + a smaller model; cost savings are an internal estimate, not a measured benchmark)."
     }

package/.claude-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "claude-mem-lite",
-  "version": "4.0.3",
+  "version": "5.0.0",
   "description": "Persistent long-term memory for Claude Code via MCP — captures coding decisions, bugfixes, and context across sessions. Hybrid FTS5 + TF-IDF search with episode batching. Single SQLite DB, no external services. A lighter, lower-cost alternative to claude-mem (episode batching + a smaller model; cost savings are an internal estimate, not a measured benchmark).",
   "author": {
     "name": "sdsrss"

package/README.md

@@ -105,7 +105,6 @@ How claude-mem-lite differs from the major neighbors in the LLM-memory space (ve
 - **Atomic writes** -- All file writes (episodes, CLAUDE.md) use write-to-tmp + rename to prevent corruption on crash
 - **Robust locking** -- PID-aware lock files with automatic stale/orphan cleanup (>30s timeout or dead PID)
 - **Stale session cleanup** -- Sessions active for >24h are automatically marked as abandoned on next start
-- **Resource registry** -- Indexes installed skills and agents with FTS5 search, composite scoring, and invocation tracking; searchable via `mem_registry` MCP tool
 - **Unified resource discovery** -- Shared filesystem traversal layer (`resource-discovery.mjs`) used by both runtime scanner and offline indexer, supporting flat directories, plugin nesting, and loose `.md` files
 - **Domain synonym expansion** -- Registry search queries expand to domain synonyms (e.g., "fix" → debug, bugfix, troubleshoot, diagnose, repair)
 - **Multi-provider LLM mode** -- Provider priority `ANTHROPIC_API_KEY` (direct Anthropic API) → `OPENROUTER_API_KEY` (OpenRouter, OpenAI-compatible — point it at any model via `OPENROUTER_MODEL`) → `claude -p` CLI fallback when no key is set
@@ -121,7 +120,6 @@ How claude-mem-lite differs from the major neighbors in the LLM-memory space (ve
 - **Configurable LLM model** -- Switch between Haiku (fast/cheap) and Sonnet (deeper analysis) via `CLAUDE_MEM_MODEL` env var
 - **DB auto-recovery** -- Detects and cleans corrupted WAL/SHM files on startup; periodic WAL checkpoints prevent unbounded growth
 - **Schema auto-migration** -- Idempotent `ALTER TABLE` migrations run on every startup, safely adding new columns and indexes without data loss
-- **Exploration bonus** -- New resources in the registry get a fair chance in composite ranking; zombie resources (high recommend, zero adopt) are penalized in scoring
 - **LLM concurrency control** -- File-based semaphore limits background workers to 2 concurrent LLM calls, preventing resource contention
 - **stdin overflow protection** -- Hook input truncated at 256KB with regex-based action salvage for oversized tool outputs
 - **Cross-session handoff** -- Captures session state (request, completed work, next steps, key files) on `/clear` or `/exit`, then injects context when the next session detects continuation intent via explicit keywords or FTS5 term overlap
@@ -224,7 +222,6 @@ rm -rf ~/claude-mem-lite/   # pre-v0.5 unhidden (if not auto-moved)
 ```
 ~/.claude-mem-lite/
   claude-mem-lite.db       # SQLite database — memory (WAL mode)
-  resource-registry.db     # SQLite database — skill/agent registry
   runtime/
     session-<project>    # Active session state
     ep-<project>.json    # Episode buffer
@@ -274,8 +271,6 @@ surface — reach them through the CLI column in the second table.
 | `mem_export` | `claude-mem-lite export` | JSON / JSONL dump, filters by project, type, date. |
 | `mem_fts_check` | `claude-mem-lite fts-check <check\|rebuild>` | FTS5 integrity + rebuild. |
 | `mem_browse` | `claude-mem-lite browse` | Tier-grouped dashboard (working / active / archive). |
-| `mem_registry` | `claude-mem-lite registry <action>` | List / search / import / remove skills + agents. |
-| `mem_use` | _MCP only_ | Load a skill / agent from the registry by name. |
 
 ### Skill Commands (in Claude Code chat)
 
@@ -472,29 +467,6 @@ Stop
   -> Spawn LLM summary worker (poll-based wait)
 ```
 
-### Resource Registry
-
-The resource registry (`registry.mjs`, `registry-retriever.mjs`) indexes installed skills and agents into a searchable FTS5 database. Unlike the previous proactive dispatch system, the registry is now on-demand — it's reachable via the `claude-mem-lite registry` CLI (primary path for Claude Code since v2.34.0 hides the `mem_registry` MCP tool from `tools/list`) or by direct `tools/call mem_registry` for MCP clients that know the name.
-
-```
-Registry pipeline:
-  -> registry-scanner.mjs discovers skills/agents on filesystem
-  -> resource-discovery.mjs handles flat dirs, plugin nesting, loose .md files
-  -> registry-indexer.mjs indexes content into FTS5 with metadata
-  -> registry-retriever.mjs provides BM25-ranked search with synonym expansion
-  -> mem_registry MCP tool exposes search/list/stats/import/remove/reindex actions
-
-Smart invocation (three layers):
-  L1 auto-load: UserPromptSubmit matches managed skill name in prompt
-     -> Loads content with path="~/.claude-mem-lite/managed/.../SKILL.md"
-     -> Guides: Read("path") or mem_use(name="..."), never Skill()
-  L2 bridge: PreToolUse hook intercepts Skill("name") for managed resources
-     -> Outputs content, prevents native handler failure
-  L3 explicit: mem_use(name="...") loads full content with reload path
-  Search: managed resources → Read(path), native plugins → Skill("full:name")
-```
-
-Composite scoring for search results: BM25 relevance (40%) + repo stars (15%) + success rate (15%) + adoption rate (10%) + freshness (10%) + exploration bonus (10%). Domain filtering ensures platform-specific resources (iOS, Go, Rust) only surface for matching projects.
 
 ### Episode Encoding
 
@@ -656,12 +628,6 @@ claude-mem-lite/
   format-utils.mjs     # String formatting: truncate, typeIcon, date/time/week formatting
   hash-utils.mjs       # MinHash signatures, Jaccard similarity for dedup
   bash-utils.mjs       # Bash output significance detection: errors, tests, builds, deploys
-  # Resource registry
-  registry.mjs         # Resource registry DB: schema, CRUD, FTS5, invocation tracking
-  registry-retriever.mjs # FTS5 retrieval with synonym expansion and composite scoring
-  registry-indexer.mjs # Resource indexing pipeline
-  registry-scanner.mjs # Filesystem scanner: reads content + hashes, delegates discovery
-  resource-discovery.mjs # Shared discovery layer: flat dirs, plugin nesting, loose .md files
   haiku-client.mjs     # Unified Haiku LLM wrapper: direct API or CLI fallback
   # Install & config
   install.mjs          # CLI installer: setup, uninstall, status, doctor (npx/git clone mode)
@@ -809,18 +775,6 @@ claude-mem-lite.
 | `CLAUDE_MEM_NO_TEMPLATE_REFRESH` | `1` stops SessionStart from refreshing the adopted `CLAUDE.md` managed block when the shipped template changes. | _(refreshes)_ |
 | `MEM_QUIET_HOOKS` | See Core above — the broadest injection-volume switch. | _(disabled)_ |
 
-### Registry import bounds
-
-`registry import-url` pulls from a third-party repository, so it is bounded. Entries past a
-bound are refused, not truncated, and the refusal is printed with the import result. Set any
-of these to `0` for the pre-v3.98 unlimited behavior; an unparseable or negative value keeps
-the default rather than removing the bound.
-
-| Variable | Description | Default |
-|----------|-------------|---------|
-| `CLAUDE_MEM_IMPORT_MAX_ITEMS` | Max skills/agents imported from one repository. | `200` |
-| `CLAUDE_MEM_IMPORT_MAX_FILE_BYTES` | Max size of a single `SKILL.md`/`AGENT.md`. Oversized entries are skipped; the rest still import. | `2097152` (2 MB) |
-| `CLAUDE_MEM_IMPORT_MAX_TOTAL_BYTES` | Byte budget for one import run. Exhausting it stops the walk and books the remainder as refused. | `52428800` (50 MB) |
 
 ### Retrieval tuning
 
@@ -840,7 +794,6 @@ benchmark and A/B harness are calibrated against — changing them invalidates t
 | `CLAUDE_MEM_ERROR_RECALL_BM25_MIN` | Relevance floor for the error-recall surface (memories injected after a failed Bash command). **Off by default.** Setting it to `10.5` (the calibrated value) makes the surface stay silent when its best-matching memory is not actually about the failure — the whole set is dropped, never trimmed row-by-row. **It is a real trade, not a free win:** measured on a live database at that threshold, injections fall ~37% and ~39% of firings go silent, concentrated in projects with few memories. Off by default because nothing shows the dropped rows were noise. Explore with `node benchmark/error-recall-suite.mjs --sweep`. | `0` (off) |
 | `CLAUDE_MEM_ERROR_RECALL_RERANK` | `off` restores the flat keyword ordering of the error-recall surface. **On by default**, and unlike the floor above it removes nothing: memories that share only the failed command's vocabulary are demoted below memories that mention the failure itself, and when a project has none of the latter the result is unchanged. Measured on a live database over 52 real failing commands × 15 projects: the lead memory matched no error term in 42.3% of firings before, 21.5% after, with the injected row count identical. | _(on)_ |
 | `CLAUDE_MEM_ERROR_RECALL_ON_FAILURE` | `off` stops the plugin from recalling memories when a Bash command **fails at the host level**. On by default. Claude Code delivers failed tool calls to a separate `PostToolUseFailure` hook event, so before this the surface only ever saw commands that exited `0` while printing error-ish text — a genuinely failing build recalled nothing. Denials from your own guardrails (sandbox, policy hooks, declined permission prompts) and commands you interrupted are never recalled for. | _(on)_ |
-| `CLAUDE_MEM_REGISTRY_CONFINE` | `off` lets registry enrichment read a resource whose `local_path` lies outside the managed data directory. **On by default.** Enrichment reads the file at `resources.local_path` and sends it to an LLM; the confinement check used to guard one of the four code paths that do this (`mem_registry(action="enrich")`) and not the other three, so `enrich <name>`, `enrich --all` and `import --enrich` read any path the row happened to hold. All four are gated now. Turn it off only if you deliberately registered resources outside `CLAUDE_MEM_DIR` and want them enriched; only `off` disables it — case-insensitively and ignoring surrounding whitespace — so a typo leaves the guard on. | _(on)_ |
 | `CLAUDE_MEM_UPS_IDENTIFIER_BYPASS` | `0` disables the bypass that lets an exact identifier match skip the score floors. | _(on)_ |
 | `CLAUDE_MEM_UPS_PROMPT_FALLBACK_LIMIT` | How many past-prompt rows the fallback arm may return. | `1` |
 | `MEM_COVERAGE_THRESHOLD` | Fraction of query terms a memory must cover to qualify (∈ [0,1]). | `0.4` |
@@ -885,7 +838,6 @@ what is already stored — only whether new work runs.
 | `CLAUDE_MEM_SKIP_MARKER_GC` | Skip the runtime-marker sweep. | _(runs)_ |
 | `CLAUDE_MEM_SKIP_UPDATE` | Skip the 24h auto-update check against GitHub Releases. | _(runs)_ |
 | `CLAUDE_MEM_SKIP_SIG_VERIFY` | Skip Ed25519 signature verification of a downloaded update. **Escape hatch — leaves updates unauthenticated.** | _(verifies)_ |
-| `CLAUDE_MEM_SKIP_REPOS` | Skip skill/agent registry seeding during install. | _(seeds)_ |
 | `CLAUDE_MEM_NO_LESSON_RETRY` | `1` disables the one-shot retry that re-asks for a missing `lesson_learned`. | _(retries)_ |
 | `CLAUDE_MEM_FLUSH_TIMEOUT` | Seconds the Stop hook waits for pending episode flushes. | `15` |
 | `CLAUDE_MEM_BACKUP_BUDGET_MB` | Disk budget for backup snapshots; the next maintain/save evicts oldest snapshots past the 7-day undo grace. | `256` |
@@ -897,7 +849,6 @@ and names can change between releases.
 
 | Variable | Description | Default |
 |----------|-------------|---------|
-| `CLAUDE_MEM_RECOMMEND_MODE` | Skill-recommendation engine: `shadow` (log would-be recommendations, inject nothing) or `off`. **`live` is parsed but not implemented** — live injection is Phase 2. Setting it runs shadow and prints one warning to stderr per process; `claude-mem-lite doctor` also reports it as an inert flag. | `shadow` |
 | `CLAUDE_MEM_TASK_IMPERATIVE` | `on`/`1` injects the single most relevant lesson at prompt position under an imperative template. | _(off)_ |
 | `CLAUDE_MEM_SUBAGENT_INJECT` | Dispatch-time memory injection for subagents. | _(off)_ |
 | `CLAUDE_MEM_SALIENCE` | Selects a comprehension-bridge arm (`bridge`, `bind`); unset = current default behavior. | _(unset)_ |

package/README.zh-CN.md

@@ -81,8 +81,6 @@
 - **原子写入** -- 所有文件写入（episode、CLAUDE.md）使用 write-to-tmp + rename 防止崩溃时损坏
 - **健壮锁机制** -- PID 感知的锁文件，自动清理过期（>30s）或孤儿（PID 已死）锁
 - **过期会话清理** -- 活跃超过 24 小时的会话在下次启动时自动标记为 abandoned
-- **智能调用** -- 三层调用系统：L1 自动加载（UserPromptSubmit 匹配 skill 名注入内容 + `Read()` 路径），L2 Bridge（PreToolUse 拦截 `Skill()` 误调），L3 显式调用（`mem_use` MCP 工具）。managed 资源用 `Read("~/.claude-mem-lite/managed/.../SKILL.md")`，原生插件用 `Skill("full:name")`
-- **资源注册表** -- 对已安装的 skill 和 agent 建立 FTS5 索引，支持复合评分和调用追踪。搜索结果区分 managed（Read 路径）vs native（Skill 全名）调用方式
 - **统一资源发现** -- 共享文件系统遍历层（`resource-discovery.mjs`），运行时扫描器和离线索引器共用，支持扁平目录、插件嵌套和松散 `.md` 文件
 - **领域同义词扩展** -- 注册表搜索查询自动扩展领域同义词（如 "修复" → fix, debug, bugfix, repair, error）
 - **持久化冷却机制** -- 5 分钟跨会话冷却 + 同会话去重，避免重复推荐 skill 自动加载
@@ -188,7 +186,6 @@ rm -rf ~/claude-mem-lite/   # v0.5 前的非隐藏目录（如未自动迁移）
 ```
 ~/.claude-mem-lite/
   claude-mem-lite.db       # SQLite 数据库 — 记忆（WAL 模式）
-  resource-registry.db     # SQLite 数据库 — skill/agent 注册表
   runtime/
     session-<project>    # 活跃会话状态
     ep-<project>.json    # Episode 缓冲区
@@ -233,8 +230,6 @@ v2.34.0 起服务端注册 17 个工具，但 `tools/list` 只暴露 6 个 **核
 | `mem_export` | `claude-mem-lite export` | JSON / JSONL 导出，支持项目/类型/日期过滤。 |
 | `mem_fts_check` | `claude-mem-lite fts-check <check\|rebuild>` | FTS5 完整性检查与重建。 |
 | `mem_browse` | `claude-mem-lite browse` | 分层仪表盘（working / active / archive）。 |
-| `mem_registry` | `claude-mem-lite registry <action>` | 列 / 搜索 / 导入 / 移除 skill / agent。 |
-| `mem_use` | _MCP only_ | 从 registry 按名载入 skill / agent。 |
 
 ### 技能命令（在 Claude Code 聊天中使用）
 
@@ -370,11 +365,6 @@ PostToolUse（每次工具执行）
   -> 为有意义的 episode 启动 LLM episode worker
   -> 错误触发回忆：搜索记忆中相关的历史修复
 
-PreToolUse（工具执行前）
-  -> L2 Skill Bridge：拦截对 managed 资源的 Skill() 调用
-     -> 匹配 managed 路径 → 输出内容 + mem_use() 提示
-     -> 未匹配 → 静默放行到原生 handler
-
 UserPromptSubmit（两个并行路径）
   -> [user-prompt-search.js] 通过 FTS5 + 活跃文件上下文自动搜索记忆
   -> [user-prompt-search.js] 注入相关历史观察（按时效和重要性加权）
@@ -393,36 +383,6 @@ Stop
   -> 启动 LLM 摘要 worker（轮询等待）
 ```
 
-### 智能调用系统
-
-三层调用系统确保 managed 资源（`~/.claude-mem-lite/managed/` 中的 skill 和 agent）能被正确调用：
-
-```
-L1 自动加载（UserPromptSubmit，<50ms）
-  -> 匹配 prompt 中的 managed skill/agent 名称
-  -> 加载 SKILL.md / {name}.md 内容
-  -> 输出：Read("~/.claude-mem-lite/managed/.../path.md") 调用指引
-  -> 截断时提供 mem_use(name="...") 备选
-
-L2 Bridge（PreToolUse Skill hook，<30ms）
-  -> 拦截 Skill("name") 调用，查询 managed 注册表
-  -> 匹配到 → 输出内容 + mem_use() 提示（防止原生 handler 报错）
-  -> 未匹配 → 放行到原生 Skill handler
-
-L3 显式调用（mem_use MCP 工具）
-  -> 按名称精确匹配 + FTS5 模糊回退
-  -> 返回完整内容 + 便携路径供 Read() 重载
-```
-
-**调用方式区分：**
-
-| 资源类型 | 位置 | 调用方式 |
-|---------|------|---------|
-| Managed skill | `~/.claude-mem-lite/managed/skills/` | `Read("~/.../SKILL.md")` 或 `mem_use(name="...")` |
-| Managed agent | `~/.claude-mem-lite/managed/agents/` | `Read("~/.../{name}.md")` 或 `mem_use(name="...", type="agent")` |
-| 原生插件 skill | `~/.claude/plugins/cache/` | `Skill("plugin:skill-name")` |
-| 用户自建 skill | `~/.claude/skills/` | `Skill("name")` |
-
 ### Episode 编码
 
 Episode 是一批相关操作（对同一组文件的编辑），由后台 LLM worker 处理：
@@ -548,13 +508,6 @@ claude-mem-lite/
   format-utils.mjs     # 字符串格式化：截断、类型图标、日期/时间格式化
   hash-utils.mjs       # MinHash 签名、Jaccard 相似度（去重用）
   bash-utils.mjs       # Bash 输出显著性检测：错误、测试、构建、部署
-  # 智能调度
-  dispatch.mjs         # 三级调度编排：快速过滤、上下文信号、FTS5、Haiku
-  dispatch-inject.mjs  # 注入模板渲染：skill/agent 推荐
-  registry.mjs         # 资源注册表 DB：schema、CRUD、FTS5、调用追踪
-  registry-retriever.mjs # FTS5 检索：同义词扩展与复合评分
-  registry-scanner.mjs # 文件系统扫描器：读取内容 + 哈希，委托发现层
-  resource-discovery.mjs # 共享发现层：扁平目录、插件嵌套、松散 .md 文件
   haiku-client.mjs     # 统一 Haiku LLM 封装：直连 API 或 CLI 回退
   # 安装与配置
   install.mjs          # CLI 安装器：设置、卸载、状态、诊断（npx/git clone 模式）
@@ -620,7 +573,6 @@ npm run benchmark:gate    # CI 门控：指标回退超过 5% 容差时失败
 | `MEM_QUIET_HOOKS` | 低噪声 hook。设为 `1` 时，SessionStart 注入去掉 `File Lessons` / `Key Context` 两节，`[mem] Related memories` 去掉 lesson 后缀，MCP server instructions 去掉 `WHEN TO USE` / `Decision rules` 两段。ID 与 `Recent` 表仍保留，`mem_get(ids=[…])` 可继续展开细节。适用于启用了 invited-memory adopt 流程或偏好最小化自动注入的用户。**v2.82.0 起此 env 不再阻挡 auto-adopt——如需关闭 auto-adopt 用 `MEM_NO_AUTO_ADOPT=1`。** | _(禁用)_ |
 | `MEM_NO_AUTO_ADOPT` | auto-adopt 全局关闭开关（v2.82.0+）。设为 `1` 阻止每次 SessionStart 在**所有**项目自动写入 `CLAUDE.md` 托管块。项目级关闭走 `claude-mem-lite adopt --disable`（写 `<memdir>/.mem-no-auto-adopt` 哨兵，存活于 marker 删除）。 | _(禁用)_ |
 | `MEM_NO_ADOPT_HINT` | 静音当前项目未 adopt 时 SessionStart 追加的那一行 "Invited-memory 未启用…" 提示。v2.82.1 起任何安装路径每次 SessionStart 都自动 adopt，所以该提示一般只在你显式 opt out（`MEM_NO_AUTO_ADOPT=1` 或 `claude-mem-lite adopt --disable`）的项目才会出现。 | _(禁用)_ |
-| `CLAUDE_MEM_REGISTRY_CONFINE` | 设为 `off` 允许 registry enrich 读取 `local_path` 位于托管数据目录之外的资源。**默认开启。** enrich 会读取 `resources.local_path` 指向的文件并发给 LLM；此前这道路径限制只接在四条读取路径中的一条（`mem_registry(action="enrich")`）上，`enrich <name>`、`enrich --all` 和 `import --enrich` 三条都是裸读。现在四条同门。仅当你确实把资源注册在 `CLAUDE_MEM_DIR` 之外并希望对其 enrich 时才关闭；只有精确值 `off` 会关闭它，拼错则守卫保持开启。 | _(开启)_ |
 
 ## 许可证
 

package/adopt-content.mjs

@@ -151,7 +151,6 @@ PreToolUse hook 在你 Read / Edit / Write 文件前已自动 \`mem_recall\` 该
 | 导出 JSON/JSONL | \`${CLI} export [--format jsonl]\` |
 | 统计总量 / 健康 | \`${CLI} stats [--days 30]\` |
 | 删除 / 更新某条 | \`${CLI} delete <id>[,<id>]\` · \`${CLI} update <id> [--title ...]\` |
-| skill-agent registry | \`${CLI} registry <list\\|search\\|import>\` |
 
 ## CLI 速查（常用检索）
 

package/cli/common.mjs

@@ -209,27 +209,21 @@ export const KNOWN_CLI_FLAGS = new Set([
   'age-days',
   'all',
   'anchor',
-  'batch',
   'before',
   'benchmark',
   'body',
   'branch',
-  'capability-summary',
-  'category',
   'closes-deferred',
   'concepts',
   'confirm',
   'days',
   'deep',
   'detail',
-  'domain-tags',
   'dry-run',
-  'enrich',
   'execute',
   'fields',
   'file',
   'files',
-  'floors',
   'force',
   'format',
   'from',
@@ -240,16 +234,11 @@ export const KNOWN_CLI_FLAGS = new Set([
   'importance',
   'include-compressed',
   'include-noise',
-  'intent-tags',
-  'invocation-name',
   'json',
   'key',
-  'keywords',
   'lesson',
   'lesson-learned',
   'limit',
-  'local-path',
-  'margins',
   'max',
   'memdir',
   'merge-ids',
@@ -265,9 +254,7 @@ export const KNOWN_CLI_FLAGS = new Set([
   'quality',
   'query',
   'reason',
-  'repo-url',
   'rerank',
-  'resource-type',
   'retain-days',
   'retry',
   'run',
@@ -279,16 +266,12 @@ export const KNOWN_CLI_FLAGS = new Set([
   'sort',
   'source',
   'status',
-  'sweep',
   'task',
-  'tech-stack',
   'text',
   'tier',
   'title',
   'to',
-  'trigger-patterns',
   'type',
-  'use-cases',
   'verbose',
   // Catalogued 2026-08-13 when suggestUnknownFlags started reporting EVERY unknown
   // flag: these are real, code-read flags that the old edit-distance gate happened to

package/cli.mjs

@@ -18,10 +18,7 @@ const CLI_COMMANDS = new Set([
   'maintain',
   'optimize',
   'fts-check',
-  'registry',
-  'import',
   'import-jsonl',
-  'enrich',
   'activity',
   'adopt',
   'unadopt',
@@ -29,6 +26,10 @@ const CLI_COMMANDS = new Set([
   'defer',
   'help',
 ]);
+// Removed with the skill/agent resource registry (docs/audits/20260906-145304.md).
+// Kept as a named set so a stale script or muscle-memory invocation gets the reason rather
+// than a bare "Unknown command" plus a misleading edit-distance suggestion.
+const REMOVED_COMMANDS = new Set(['registry', 'import', 'enrich']);
 const INSTALL_COMMANDS = new Set([
   'install',
   'uninstall',
@@ -95,6 +96,21 @@ if (cmd === '--version' || cmd === '-v' || cmd === '-V' || cmd === 'version') {
 } else if (INSTALL_COMMANDS.has(cmd)) {
   const { main } = await import('./install.mjs');
   await main(process.argv.slice(2));
+} else if (REMOVED_COMMANDS.has(cmd)) {
+  // Released-artifact discoverability signal for the skill-registry removal. Deliberately
+  // names NO version: this ships before the version is decided, and a hardcoded one is a
+  // guess in a third place (package.json and the CHANGELOG heading being the other two).
+  // The revert instruction is version-specific and correct, which is what a user needs.
+  // Deliberately NOT routed through the edit-distance suggester below: its nearest
+  // match for `import` is `import-jsonl`, a different feature that accepts a path
+  // argument, so a stale `import <github-url>` would be pointed at something that
+  // could plausibly run. Naming the removal is the only honest answer.
+  process.stderr.write(
+    `[mem] "${cmd}" was removed along with the skill/agent resource registry.\n` +
+      "[mem] Claude Code's own plugins/marketplace replace it. See CHANGELOG.md for the\n" +
+      '[mem] migration note; to revert, pin claude-mem-lite@4.0.4.\n',
+  );
+  process.exit(1);
 } else {
   process.stderr.write(`[mem] Unknown command: "${cmd}"\n`);
   // Suggest closest command by edit distance

package/commands/update.md

@@ -1,11 +1,11 @@
 ---
 name: update
-description: "Use when: search results seem noisy, after bulk imports, or for periodic memory/registry maintenance"
+description: "Use when: search results seem noisy or for periodic memory maintenance"
 ---
 
-# Memory & Registry Maintenance
+# Memory Maintenance
 
-Run intelligent maintenance on both the memory database and tool resource registry.
+Run intelligent maintenance on the memory database.
 
 ## Usage
 
@@ -24,15 +24,8 @@ When the user invokes `/mem:update`, perform the following maintenance cycle:
 5. Run `mem_compress(preview=false)` for old low-value observations
 6. **If pending purge items > 0**: Report the count to the user and ask for confirmation. If confirmed, call `mem_maintain(action="execute", operations=["purge_stale"])`. User may optionally specify `retain_days` (default 30) to control how many days of data to keep. Do NOT purge without explicit user confirmation.
 
-### Phase 2: Registry Maintenance
-
-1. Call `mem_registry(action="stats")` to get registry overview
-2. Call `mem_registry(action="reindex")` to rebuild FTS5 search index
-3. Report updated stats
-
-### Phase 3: Summary
+### Phase 2: Summary
 
 Summarize all maintenance actions taken in zh-CN:
 - Memory: observations cleaned, decayed, boosted, deduplicated, compressed
-- Registry: total resources, adoption rates, reindex status
 - Overall health assessment

package/format-utils.mjs

@@ -125,27 +125,6 @@ export function neutralizeSkillDelimiters(s) {
   return defangToFixpoint(s, SKILL_BLOCK_RE);
 }
 
-// <skill-bridge> is the wrapper scripts/pre-skill-bridge.js puts around a managed
-// skill body it injects as PreToolUse additionalContext. The body comes from a
-// third-party repo (tools/adopt import) — an untrusted boundary — so a literal
-// `</skill-bridge>` inside it would close the wrapper early and spill the rest of
-// the payload (e.g. a forged <system-reminder>) as undelimited context (audit
-// 2026-08-14 M-4). Not in CONTEXT_DELIMITER_RE for the same reason <skill-loaded>
-// isn't: the bridge's OWN wrapper must stay live, so the defang is applied per
-// call site to the untrusted body only.
-const SKILL_BRIDGE_RE = /<\/?skill-bridge(?:\s[^>]*)?>/gi;
-
-/**
- * Defang a literal `<skill-bridge>` opener/closer in untrusted text that is about
- * to be wrapped in a real skill-bridge block. Same fixpoint treatment as the
- * classes above. Never apply to the wrapper itself.
- * @param {string} s Input string (any type; coerced)
- * @returns {string} Text with skill-bridge delimiters defanged
- */
-export function neutralizeSkillBridgeDelimiters(s) {
-  return defangToFixpoint(s, SKILL_BRIDGE_RE);
-}
-
 /**
  * Render the PostToolUse error-recall hint block (hook.mjs::triggerErrorRecall).
  * The single most-relevant hit (rows[0]) that carries a lesson_learned gets its

package/hook-context.mjs

@@ -60,11 +60,16 @@ function mdCell(s) {
 
 export function computeAdaptiveWindows(db, project) {
   const sevenDaysAgo = Date.now() - 7 * DAY_MS;
+  // liveObsFilterSql, not a hand-written half of it (A20260906-R8-§11.3). This COUNT is a
+  // POOL-SIZING input: the windows returned below are handed to the recall queries at :201 /
+  // :468 / :502, all three of which filter live rows. Counting a strictly larger population
+  // than the window is applied to inverts the documented intent — 80 superseded rows read as
+  // >10 obs/day and earned the TIGHTEST window (12h) for finding the few live rows left.
   const row = db
     .prepare(
       `
     SELECT COUNT(*) as c FROM observations
-    WHERE project = ? AND created_at_epoch > ? AND COALESCE(compressed_into, 0) = 0
+    WHERE project = ? AND created_at_epoch > ? AND ${liveObsFilterSql('')}
   `,
     )
     .get(project, sevenDaysAgo);

package/hook-handoff.mjs

@@ -136,6 +136,15 @@ export function buildAndSaveHandoff(db, sessionId, project, type, episodeSnapsho
   const obsWindowParams = ccWindowStart !== null ? [ccWindowStart] : [];
 
   // 2. Completed — from observations (include narrative for richer handoff)
+  //
+  // `compressed_into` ONLY, deliberately — not a half-written liveObsFilterSql. Audit
+  // 2026-08-14 F4 ruled on exactly this line: `completed` is the session's own history, and a
+  // decision a later save overturned still happened, so erasing it here would misreport the
+  // session. Only key_decisions (:242) is re-presented to a LATER session as standing policy,
+  // and that one does filter superseded_at. The scope guard is the third case of "F4 — handoff
+  // key_decisions excludes a retracted decision" in tests/audit-silent-20260814.test.mjs.
+  // Audit R8 §11.3 proposed adding the filter here from a repo-wide regex sweep of the
+  // predicate shape; it was rejected on this reasoning, and the guard catches it.
   const completed = db
     .prepare(
       `

package/hook-update.mjs

@@ -7,7 +7,6 @@ import {
   readFileSync,
   writeFileSync,
   copyFileSync,
-  cpSync,
   readdirSync,
   existsSync,
   lstatSync,
@@ -404,11 +403,11 @@ export function getCurrentVersion() {
 }
 
 // SWITCHABLE_PATHS = everything in SOURCE_FILES plus the recursive dirs that
-// install.mjs copies as whole subtrees (scripts, registry, node_modules). It's
+// install.mjs copies as whole subtrees (scripts, node_modules). It's
 // built per-call from the *tarball's* manifest, not the locally-imported one —
 // see loadReleaseManifest comment for why.
 function buildSwitchablePaths(sourceFiles) {
-  return [...sourceFiles, 'scripts', 'registry', 'node_modules'];
+  return [...sourceFiles, 'scripts', 'node_modules'];
 }
 
 // Load the SOURCE_FILES / HOOK_SCRIPT_FILES manifest from the *extracted
@@ -989,6 +988,41 @@ export async function installExtractedRelease(sourceDir, targetDir = INSTALL_DIR
     rmSync(stagingDir, { recursive: true, force: true });
     rmSync(backupDir, { recursive: true, force: true });
 
+    // Post-update migration: reconcile settings.json against the release we just
+    // swapped in. `configureHooks()` already strips stale mem entries — but its only
+    // caller is `install()`, and this path never runs it, while `buildSwitchablePaths`
+    // replaces `scripts/` wholesale. So an upgrade that REMOVES a hook script deletes
+    // the file and leaves settings.json pointing at it; the launcher then reports a
+    // broken install on every fire, for a file that is not coming back (R9 review P1-1,
+    // first hit by the `PreToolUse:Skill` removal). Plugin-channel installs are
+    // unaffected — hooks/hooks.json is replaced wholesale — so this is npm-channel only.
+    //
+    // The reconciler lives in lib/hook-prune.mjs, not install.mjs: install.mjs already
+    // imports THIS file, so importing it back would close a cycle the repo's import-graph
+    // guard rejects. Failure is non-fatal — a stale entry is noisy, not broken, and must
+    // never roll back an otherwise-good update.
+    try {
+      const settingsPath = join(homedir(), '.claude', 'settings.json');
+      if (existsSync(settingsPath)) {
+        const { pruneDanglingMemHooks } = await import('./lib/hook-prune.mjs');
+        const before = JSON.parse(readFileSync(settingsPath, 'utf8'));
+        const { settings, removed } = pruneDanglingMemHooks(before, targetDir);
+        if (removed.length > 0) {
+          // Through the shared writer: settings.json is user-owned config and may be a
+          // symlink into a dotfiles repo, which a bare temp+rename would sever (P0-5).
+          const { atomicWriteFileSync } = await import('./lib/atomic-write.mjs');
+          atomicWriteFileSync(settingsPath, JSON.stringify(settings, null, 2) + '\n', { backup: true });
+          debugLog(
+            'INFO',
+            'update',
+            `pruned ${removed.length} dangling hook entr(ies): ${removed.join(', ')}`,
+          );
+        }
+      }
+    } catch (e) {
+      debugCatch(e, 'pruneDanglingMemHooks');
+    }
+
     // Post-update migration: clean stale global MCPs if plugin handles it.
     // Both "mem" (legacy, pre-v2.78) and "mem-lite" (current) are purged so a
     // user who manually ran `claude mcp add` in either era doesn't end up with
@@ -1194,15 +1228,6 @@ function copyReleaseIntoStaging(
     }
   }
 
-  // registry/ stays recursive — preinstalled.json is the only current entry
-  // but the directory is consumed wholesale by the registry indexer and may
-  // grow subtrees. Pre-v2.55 readdirSync+copyFileSync would EISDIR-throw on
-  // any subdir and silently roll back the entire update.
-  const sourceRegistry = join(sourceDir, 'registry');
-  if (existsSync(sourceRegistry)) {
-    cpSync(sourceRegistry, join(stagingDir, 'registry'), { recursive: true });
-  }
-
   const stagedScripts = join(stagingDir, 'scripts');
   if (existsSync(stagedScripts)) {
     for (const sf of readdirSync(stagedScripts).filter((n) => n.endsWith('.sh'))) {

package/hook.mjs

@@ -165,10 +165,10 @@ async function loadCacheGuard() {
 // Audit 2026-09-02 P1-8. `hook.mjs` is ONE entry point for seven events, so every static
 // import is paid by every event — PostToolUse, the highest-frequency one, was loading 85
 // modules (1.37 MB) to use a handful. The six modules below are loaded inside the handler
-// that needs them instead: registry-recommend, patha-exclude-meter, hook-update,
-// hook-optimize, adopt-cli, upgrade-banner. Measured marginal cost of the four largest,
-// same process: hook-update 9.4 ms, registry-recommend 5.1, hook-optimize 5.9, the rest
-// 0.5-2.4 each.
+// that needs them instead: patha-exclude-meter, hook-update, hook-optimize, adopt-cli,
+// upgrade-banner. Measured marginal cost of the largest, same process: hook-update
+// 9.4 ms, hook-optimize 5.9, the rest 0.5-2.4 each. (registry-recommend, 5.1 ms, was a
+// sixth entry until the skill-registry subsystem was removed — see docs/audits/20260906-145304.md.)
 //
 // A failing dynamic import lands in the dispatcher's tail catch -> recordHookError, the
 // same place a failing static import would have landed the whole process; each site keeps
@@ -642,23 +642,6 @@ async function handlePostToolUse() {
   if (SKIP_TOOLS.has(tool_name)) return;
   if (SKIP_PREFIXES.some((p) => tool_name.startsWith(p))) return;
 
-  // Shadow skill-adoption telemetry. mem_use is pre-filtered above, so the Skill tool is
-  // the only visible adoption signal (v1). Placed before the resp-length gate because a
-  // skill load's response shape varies. Never throws.
-  if (tool_name === 'Skill') {
-    const ti = typeof tool_input === 'string' ? tryParseJson(tool_input) : tool_input || {};
-    // hookData.session_id (CC UUID) pairs this adoption to the would-be reco from the
-    // UserPromptSubmit hook earlier in the same session (matched precision, B1).
-    // Lazy: only a `Skill` tool call needs this module, which is a small fraction of
-    // PostToolUse fires.
-    try {
-      const { recordSkillAdoption } = await import('./registry-recommend.mjs');
-      recordSkillAdoption('Skill', ti, inferProject(), hookData.session_id);
-    } catch {
-      /* telemetry only — never blocks capture */
-    }
-  }
-
   const resp = normalizeToolResponse(tool_response);
   if (!resp || resp.length < 10) return;
 
@@ -2254,14 +2237,7 @@ async function handleSessionStart() {
   } catch {
     /* best-effort */
   }
-  // Bound the shadow-recommendation log (daily JSONL shards, no GC at write time).
-  try {
-    const { gcOldShadowShards } = await import('./registry-recommend.mjs');
-    gcOldShadowShards();
-  } catch {
-    /* best-effort, never blocks SessionStart */
-  }
-  // Same for the opt-in metrics sink, which lives under DB_DIR. Runs even when
+  // Bound the opt-in metrics sink, which lives under DB_DIR. Runs even when
   // metrics are disabled, so shards left by a since-toggled-off run still get pruned.
   try {
     gcOldMetricShards(DB_DIR);

package/hooks/hooks.json

@@ -41,16 +41,6 @@
           }
         ]
       },
-      {
-        "matcher": "Skill",
-        "hooks": [
-          {
-            "type": "command",
-            "command": "node \"${CLAUDE_PLUGIN_ROOT}/scripts/hook-launcher.mjs\" scripts/pre-skill-bridge.js",
-            "timeout": 3
-          }
-        ]
-      },
       {
         "matcher": "Agent|Task",
         "hooks": [