@pcircle/memesh 4.2.11 → 4.4.0

package/.claude-plugin/marketplace.json

@@ -8,11 +8,11 @@
       "name": "memesh",
       "source": "./",
       "description": "MeMesh — Local memory for Claude Code and MCP coding agents. One SQLite file, zero cloud required.",
-      "version": "4.2.11",
+      "version": "4.4.0",
       "author": {
         "name": "PCIRCLE AI"
       },
-      "homepage": "https://pcircle.ai/memesh-llm-memory",
+      "homepage": "https://pcircle.com/memesh-llm-memory",
       "license": "MIT",
       "keywords": [
         "claude-code",

package/.claude-plugin/plugin.json

@@ -4,8 +4,8 @@
   "author": {
     "name": "PCIRCLE AI"
   },
-  "version": "4.2.11",
-  "homepage": "https://pcircle.ai/memesh-llm-memory",
+  "version": "4.4.0",
+  "homepage": "https://pcircle.com/memesh-llm-memory",
   "repository": "https://github.com/PCIRCLE-AI/memesh-llm-memory",
   "license": "MIT",
   "keywords": [

package/README.de.md

@@ -83,7 +83,7 @@ flowchart TB
 | `/memesh` skill im Claude-Code-Chat verwenden | Path A (Plugin) |
 | Auto-Capture in Claude Code (Session → Lessons → nächste Recall) | Path A (Plugin) |
 | `memesh remember` / `memesh recall` / `memesh doctor` im Terminal | Path B (npm-global) |
-| `memesh` direkt zum Öffnen des Dashboards (ohne `npx`-Startverzögerung) | Path B (npm-global) |
+| `memesh serve` direkt zum Öffnen des Dashboards (ohne `npx`-Startverzögerung) | Path B (npm-global) |
 | `memesh-mcp` an Cursor, Cline oder andere MCP-Clients anbinden | Path B (npm-global) |
 | Alles oben | **Beide installieren** — kein Konflikt |
 
@@ -149,7 +149,7 @@ memesh doctor
 Dashboard öffnen, um den Speicher zu erkunden:
 
 ```bash
-memesh
+memesh serve
 ```
 
 <p align="center">
@@ -173,7 +173,7 @@ memesh
 | **Claude Code verwenden** | Projektentscheidungen, dateispezifische Erkenntnisse und vergangene Fehler während der Arbeit automatisch abrufen |
 | **Power-User von Coding-Agenten** | Eine lokale Speicherschicht über MCP-kompatible Tools verteilen |
 | **ein Team mit KI-Coding-Workflows experimentiert** | Projektwissen ohne gehostete Infrastruktur aus- und importieren |
-| **Agent-Entwickler** | Lokalen Speicher via MCP, HTTP, CLI oder Python-SDK hinzufügen |
+| **Agent-Entwickler** | Lokalen Speicher via MCP, HTTP oder CLI hinzufügen |
 
 ---
 
@@ -298,7 +298,7 @@ Wenn npm eine installierte Version als veraltet kennzeichnet (typischerweise ein
 
 **⚠️ Konflikterkennung** — Wenn Sie zwei Memories haben, die sich widersprechen, warnt Sie MeMesh.
 
-**🕸️ Wissensgraph-Konnektivität** — `memesh kg backfill-relations --all-rules` verknüpft verwaiste Entitäten über Tag-Kookurrenz, Projekt-Clustering, Sitzungskontext und Namensähnlichkeit — ohne LLM. Reduziert die Waisenrate auf einer repräsentativen Wissensbasis von 89% auf unter 12%.
+**🕸️ Wissensgraph-Konnektivität** — `memesh kg backfill-relations --all-rules` verknüpft verwaiste Entitäten über Tag-Kookurrenz, Projekt-Clustering, Sitzungskontext und Namensähnlichkeit — ohne LLM.
 
 **📦 Team-Freigabe** — `memesh export > team-knowledge.json` → mit Team teilen → `memesh import team-knowledge.json`
 Importierte Bundles bleiben durchsuchbar, aber MeMesh injiziert importierte Memories nicht automatisch in Claude Hooks, bis Sie sie überprüfen oder lokal neu speichern.
@@ -330,7 +330,7 @@ memesh config set llm.api-key sk-ant-...
 Oder nutzen Sie den Dashboard-Settings-Reiter (visuelles Setup):
 
 ```bash
-memesh  # öffnet Dashboard → Settings-Reiter
+memesh serve  # öffnet Dashboard → Settings-Reiter
 ```
 
 ### Eigene Embeddings verwenden (optional)
@@ -425,7 +425,7 @@ Beim Session-Start erscheint ein einzeiliges Banner (pro Version alle 24h gedros
 ```bash
 git clone https://github.com/PCIRCLE-AI/memesh-llm-memory
 cd memesh-llm-memory && npm install && npm run build
-npm test             # 630 tests
+npm test
 npm run test:e2e-dashboard
 ```
 
@@ -434,5 +434,5 @@ Dashboard: `cd dashboard && npm install && npm run dev`
 ---
 
 <p align="center">
-  <strong>MIT</strong> — Erstellt von <a href="https://pcircle.ai">PCIRCLE AI</a>
+  <strong>MIT</strong> — Erstellt von <a href="https://pcircle.com">PCIRCLE AI</a>
 </p>

package/README.es.md

@@ -83,7 +83,7 @@ flowchart TB
 | Usar el skill `/memesh` dentro de una conversación de Claude Code | Path A (plugin) |
 | Auto-captura en Claude Code (sesión → lecciones → recall siguiente) | Path A (plugin) |
 | Ejecutar `memesh remember` / `memesh recall` / `memesh doctor` en cualquier terminal | Path B (npm-global) |
-| Abrir el dashboard con `memesh` (sin retraso de arranque de `npx`) | Path B (npm-global) |
+| Abrir el dashboard con `memesh serve` (sin retraso de arranque de `npx`) | Path B (npm-global) |
 | Conectar `memesh-mcp` a Cursor, Cline u otro cliente MCP | Path B (npm-global) |
 | Todo lo anterior | **Instala ambos** — no entran en conflicto |
 
@@ -176,7 +176,7 @@ memesh doctor
 Abre el dashboard para explorar tu memoria:
 
 ```bash
-memesh
+memesh serve
 ```
 
 <p align="center">
@@ -200,7 +200,7 @@ memesh
 | **Un desarrollador usando Claude Code** | Recuperar automáticamente decisiones del proyecto, lecciones específicas de archivos y fracasos anteriores mientras trabajas |
 | **Un usuario avanzado de agentes de codificación** | Compartir una capa de memoria local entre herramientas compatibles con MCP |
 | **Un equipo experimentando con flujos de trabajo de IA para codificación** | Exportar/importar conocimiento del proyecto sin introducir infraestructura alojada |
-| **Un desarrollador de agentes** | Añadir memoria local mediante MCP, HTTP, CLI o el SDK de Python |
+| **Un desarrollador de agentes** | Añadir memoria local mediante MCP, HTTP o la CLI |
 
 ---
 
@@ -326,7 +326,7 @@ Cuando npm marca una versión instalada como deprecada (típicamente un aviso de
 
 **⚠️ Detección de Conflictos** — Si tienes dos memorias que se contradicen, MeMesh te advierte.
 
-**🕸️ Conectividad del grafo de conocimiento** — `memesh kg backfill-relations --all-rules` vincula entidades huérfanas mediante co-ocurrencia de etiquetas, agrupación de proyectos, contexto de sesión y similitud de nombres — sin LLM. Reduce la tasa de huérfanos del 89% a menos del 12% en una base de conocimiento representativa.
+**🕸️ Conectividad del grafo de conocimiento** — `memesh kg backfill-relations --all-rules` vincula entidades huérfanas mediante co-ocurrencia de etiquetas, agrupación de proyectos, contexto de sesión y similitud de nombres — sin LLM.
 
 **📦 Compartir en Equipo** — `memesh export > team-knowledge.json` → comparte con tu equipo → `memesh import team-knowledge.json`
 Los bundles importados permanecen buscables, pero MeMesh no inyecta automáticamente memorias importadas en hooks de Claude hasta que las revises o las guardes localmente de nuevo.
@@ -358,7 +358,7 @@ memesh config set llm.api-key sk-ant-...
 O usa la pestaña Configuración del dashboard (configuración visual):
 
 ```bash
-memesh  # abre dashboard → pestaña Settings
+memesh serve  # abre dashboard → pestaña Settings
 ```
 
 ### Usa tus propios embeddings (opcional)
@@ -374,7 +374,7 @@ El embedder se configura **independientemente del LLM de chat** — cambiar `llm
 
 | | Nivel 0 (por defecto) | Nivel 1 (Modo Inteligente) |
 |---|---|---|
-| **Búsqueda** | FTS5 + sqlite-vec, 95.60% R@5 (~4ms por recall) | sin cambios — el recall es sin LLM en cada nivel |
+| **Búsqueda** | FTS5 + sqlite-vec, 95.60% R@5 | sin cambios — el recall es sin LLM en cada nivel |
 | **Auto-capture** | Patrones basados en reglas | + LLM extrae decisiones y lecciones |
 | **Auto-etiquetado** | Solo etiquetas manuales | + LLM genera etiquetas para nuevas memorias |
 | **Análisis de fallos** | No disponible | + LLM convierte errores de sesión en lecciones estructuradas |
@@ -453,7 +453,7 @@ Al inicio de sesión aparece un banner de una línea (limitado a una vez cada 24
 ```bash
 git clone https://github.com/PCIRCLE-AI/memesh-llm-memory
 cd memesh-llm-memory && npm install && npm run build
-npm test             # 630 tests
+npm test
 npm run test:e2e-dashboard
 ```
 
@@ -462,5 +462,5 @@ Dashboard: `cd dashboard && npm install && npm run dev`
 ---
 
 <p align="center">
-  <strong>MIT</strong> — Hecho por <a href="https://pcircle.ai">PCIRCLE AI</a>
+  <strong>MIT</strong> — Hecho por <a href="https://pcircle.com">PCIRCLE AI</a>
 </p>

package/README.fr.md

@@ -83,7 +83,7 @@ flowchart TB
 | Utiliser le skill `/memesh` dans une conversation Claude Code | Path A (plugin) |
 | Auto-capture dans Claude Code (session → leçons → recall suivant) | Path A (plugin) |
 | Exécuter `memesh remember` / `memesh recall` / `memesh doctor` dans n'importe quel terminal | Path B (npm-global) |
-| Ouvrir le dashboard via `memesh` (sans délai de démarrage `npx`) | Path B (npm-global) |
+| Ouvrir le dashboard via `memesh serve` (sans délai de démarrage `npx`) | Path B (npm-global) |
 | Brancher `memesh-mcp` à Cursor, Cline ou un autre client MCP | Path B (npm-global) |
 | Tout ce qui précède | **Installez les deux** — ils ne sont pas en conflit |
 
@@ -149,7 +149,7 @@ memesh doctor
 Ouvrez le tableau de bord pour explorer votre mémoire :
 
 ```bash
-memesh
+memesh serve
 ```
 
 <p align="center">
@@ -173,7 +173,7 @@ memesh
 | **Un développeur utilisant Claude Code** | Rappeler automatiquement les décisions du projet, les leçons spécifiques aux fichiers et les échecs passés au fur et à mesure du travail |
 | **Un utilisateur avancé d'agent de codage** | Partager une couche de mémoire locale unique sur les outils compatibles MCP |
 | **Une équipe expérimentant les workflows de codage IA** | Exporter/importer les connaissances du projet sans infrastructure hébergée |
-| **Un développeur d'agent** | Ajouter la mémoire locale via MCP, HTTP, CLI ou le SDK Python |
+| **Un développeur d'agent** | Ajouter la mémoire locale via MCP, HTTP ou la CLI |
 
 ---
 
@@ -299,7 +299,7 @@ Lorsque npm signale une version installée comme dépréciée (typiquement une a
 
 **⚠️ Détection De Conflits** — Si vous avez deux mémoires qui se contredisent, MeMesh vous avertit.
 
-**🕸️ Connectivité du graphe de connaissances** — `memesh kg backfill-relations --all-rules` relie les entités orphelines par cooccurrence de tags, clustering de projets, contexte de session et similarité de noms — sans LLM. Réduit le taux d'orphelins de 89% à moins de 12% sur une base de connaissances représentative.
+**🕸️ Connectivité du graphe de connaissances** — `memesh kg backfill-relations --all-rules` relie les entités orphelines par cooccurrence de tags, clustering de projets, contexte de session et similarité de noms — sans LLM.
 
 **📦 Partage D'Équipe** — `memesh export > team-knowledge.json` → partagez avec votre équipe → `memesh import team-knowledge.json`
 Les bundles importés restent consultables, mais MeMesh n'injecte pas automatiquement les mémoires importées dans les hooks Claude jusqu'à ce que vous les examiniez ou les rémémorisiez localement.
@@ -331,7 +331,7 @@ memesh config set llm.api-key sk-ant-...
 Ou utilisez l'onglet Settings du tableau de bord (configuration visuelle) :
 
 ```bash
-memesh  # ouvre le tableau de bord → onglet Settings
+memesh serve  # ouvre le tableau de bord → onglet Settings
 ```
 
 ### Utilisez vos propres embeddings (optionnel)
@@ -426,7 +426,7 @@ Au démarrage de session, une bannière sur une ligne s'affiche (limitée à une
 ```bash
 git clone https://github.com/PCIRCLE-AI/memesh-llm-memory
 cd memesh-llm-memory && npm install && npm run build
-npm test             # 630 tests
+npm test
 npm run test:e2e-dashboard
 ```
 
@@ -435,5 +435,5 @@ Tableau de bord : `cd dashboard && npm install && npm run dev`
 ---
 
 <p align="center">
-  <strong>MIT</strong> — Créé par <a href="https://pcircle.ai">PCIRCLE AI</a>
+  <strong>MIT</strong> — Créé par <a href="https://pcircle.com">PCIRCLE AI</a>
 </p>

package/README.ja.md

@@ -83,7 +83,7 @@ flowchart TB
 | Claude Code の会話で `/memesh` skill を使う | Path A（プラグイン）|
 | Claude Code で自動キャプチャ（session → 学習 → 次回リコール） | Path A（プラグイン）|
 | ターミナルで `memesh remember` / `memesh recall` / `memesh doctor` を実行 | Path B（npm-global）|
-| `memesh` でダッシュボードを直接起動（`npx` 起動遅延なし） | Path B（npm-global）|
+| `memesh serve` でダッシュボードを直接起動（`npx` 起動遅延なし） | Path B（npm-global）|
 | `memesh-mcp` を Cursor、Cline、その他の MCP クライアントに接続 | Path B（npm-global）|
 | すべて | **両方インストール** — 競合しません |
 
@@ -176,7 +176,7 @@ memesh doctor
 ダッシュボードを開いてメモリを探索します:
 
 ```bash
-memesh
+memesh serve
 ```
 
 <p align="center">
@@ -200,7 +200,7 @@ memesh
 | **Claude Code を使う開発者** | プロジェクト判断、ファイル固有の知見、過去の失敗が自動で呼び出される |
 | **コーディングエージェントのパワーユーザー** | 1 つのローカルメモリレイヤーを MCP 互換ツール全体で共有 |
 | **AI コーディングワークフローを実験中のチーム** | ホステッドインフラを導入せず、プロジェクト知識をエクスポート・インポート |
-| **エージェント開発者** | MCP、HTTP、CLI、Python SDK 経由でローカルメモリを追加 |
+| **エージェント開発者** | MCP、HTTP、CLI 経由でローカルメモリを追加 |
 
 ---
 
@@ -326,7 +326,7 @@ npm がインストール済みバージョンを非推奨としてフラグし
 
 **⚠️ 矛盾検出** — 互いに矛盾するメモリが 2 つある場合、MeMesh が警告します。
 
-**🕸️ ナレッジグラフ接続性** — `memesh kg backfill-relations --all-rules` はタグの共起・プロジェクトクラスタリング・セッションコンテキスト・名前類似度を使って孤立エンティティをリンク — LLM 不要。代表的なナレッジベースで孤立率を 89% から 12% 未満に削減。
+**🕸️ ナレッジグラフ接続性** — `memesh kg backfill-relations --all-rules` はタグの共起・プロジェクトクラスタリング・セッションコンテキスト・名前類似度を使って孤立エンティティをリンク — LLM 不要。
 
 **📦 チーム共有** — `memesh export > team-knowledge.json` → チームと共有 → `memesh import team-knowledge.json`。
 インポートされたバンドルは検索可能ですが、MeMesh はレビュー・ローカル再保存まで Claude フックへの自動注入はしません。
@@ -358,7 +358,7 @@ memesh config set llm.api-key sk-ant-...
 またはダッシュボード Settings タブで視覚的にセットアップ:
 
 ```bash
-memesh  # ダッシュボード → Settings タブを開く
+memesh serve  # ダッシュボード → Settings タブを開く
 ```
 
 ### 独自のエンベディングを使う(任意)
@@ -374,7 +374,7 @@ memesh config set embedder.model text-embedding-3-small
 
 | | レベル 0 (デフォルト) | レベル 1 (スマートモード) |
 |---|---|---|
-| **検索** | FTS5 + sqlite-vec、95.60% R@5(リコール 1 回あたり ~4ms) | 変更なし — リコールはどのレベルでも LLM フリー |
+| **検索** | FTS5 + sqlite-vec、95.60% R@5 | 変更なし — リコールはどのレベルでも LLM フリー |
 | **自動キャプチャ** | ルールベースパターン | + LLM が判断・教訓を抽出 |
 | **自動タグ付け** | 手動タグのみ | + LLM が新規メモリにタグを生成 |
 | **失敗分析** | 利用不可 | + LLM がセッションエラーを構造化教訓に変換 |
@@ -453,7 +453,7 @@ bash "$(npm prefix -g)/lib/node_modules/@pcircle/memesh/scripts/upgrade-plugin.s
 ```bash
 git clone https://github.com/PCIRCLE-AI/memesh-llm-memory
 cd memesh-llm-memory && npm install && npm run build
-npm test             # 630 tests
+npm test
 npm run test:e2e-dashboard
 ```
 
@@ -462,5 +462,5 @@ npm run test:e2e-dashboard
 ---
 
 <p align="center">
-  <strong>MIT</strong> — Made by <a href="https://pcircle.ai">PCIRCLE AI</a>
+  <strong>MIT</strong> — Made by <a href="https://pcircle.com">PCIRCLE AI</a>
 </p>

package/README.ko.md

@@ -83,7 +83,7 @@ flowchart TB
 | Claude Code 대화에서 `/memesh` skill 사용 | Path A(플러그인) |
 | Claude Code에서 자동 캡처(session → 교훈 → 다음 recall) | Path A(플러그인) |
 | 터미널에서 `memesh remember` / `memesh recall` / `memesh doctor` 실행 | Path B(npm-global) |
-| `memesh`로 대시보드 바로 열기(`npx` 시작 지연 없음) | Path B(npm-global) |
+| `memesh serve`로 대시보드 바로 열기(`npx` 시작 지연 없음) | Path B(npm-global) |
 | `memesh-mcp`를 Cursor, Cline 또는 기타 MCP 클라이언트에 연결 | Path B(npm-global) |
 | 위 전부 | **둘 다 설치** — 충돌 없음 |
 
@@ -176,7 +176,7 @@ memesh doctor
 대시보드를 열어서 메모리를 탐색합니다:
 
 ```bash
-memesh
+memesh serve
 ```
 
 <p align="center">
@@ -200,7 +200,7 @@ memesh
 | **Claude Code를 사용 중** | 프로젝트 결정, 파일별 교훈, 과거 실패를 작업 중에 자동으로 회상 |
 | **코딩 에이전트 파워 유저** | MCP 호환 도구 전체에서 하나의 로컬 메모리 레이어 공유 |
 | **팀이 AI 코딩 워크플로우 실험 중** | 호스팅 인프라 도입 없이 프로젝트 지식 내보내기/가져오기 |
-| **에이전트 개발자** | MCP, HTTP, CLI, Python SDK를 통해 로컬 메모리 추가 |
+| **에이전트 개발자** | MCP, HTTP, CLI를 통해 로컬 메모리 추가 |
 
 ---
 
@@ -326,7 +326,7 @@ npm이 설치된 버전을 deprecated로 플래그하면(일반적으로 보안
 
 **⚠️ 충돌 감지** — 서로 모순되는 메모리 두 개가 있으면 MeMesh가 경고합니다.
 
-**🕸️ 지식 그래프 연결성** — `memesh kg backfill-relations --all-rules`는 태그 공동 발생, 프로젝트 클러스터링, 세션 컨텍스트, 이름 유사성을 사용해 고아 엔티티를 연결 — LLM 불필요. 대표적인 지식 베이스에서 고아 비율을 89%에서 12% 미만으로 감소.
+**🕸️ 지식 그래프 연결성** — `memesh kg backfill-relations --all-rules`는 태그 공동 발생, 프로젝트 클러스터링, 세션 컨텍스트, 이름 유사성을 사용해 고아 엔티티를 연결 — LLM 불필요.
 
 **📦 팀 공유** — `memesh export > team-knowledge.json` → 팀과 공유 → `memesh import team-knowledge.json`
 임포트된 번들은 계속 검색 가능하지만, MeMesh는 검토하거나 로컬에 다시 저장할 때까지 Claude 훅에 임포트된 메모리를 자동 주입하지 않습니다.
@@ -358,7 +358,7 @@ memesh config set llm.api-key sk-ant-...
 또는 대시보드 Settings 탭 사용 (비주얼 설정):
 
 ```bash
-memesh  # 대시보드 열기 → Settings 탭
+memesh serve  # 대시보드 열기 → Settings 탭
 ```
 
 ### 자체 임베딩 사용 (선택)
@@ -374,7 +374,7 @@ memesh config set embedder.model text-embedding-3-small
 
 | | Level 0 (기본) | Level 1 (스마트 모드) |
 |---|---|---|
-| **검색** | FTS5 + sqlite-vec, R@5 95.60% (회상 1회당 ~4ms) | 변경 없음 — 회상은 모든 레벨에서 LLM-free |
+| **검색** | FTS5 + sqlite-vec, R@5 95.60% | 변경 없음 — 회상은 모든 레벨에서 LLM-free |
 | **자동 캡처** | 규칙 기반 패턴 | + LLM이 결정과 교훈 추출 |
 | **자동 태그 부여** | 수동 태그만 | + LLM이 새 메모리에 태그 생성 |
 | **실패 분석** | 사용 불가 | + LLM이 세션 에러를 구조화된 교훈으로 변환 |
@@ -453,7 +453,7 @@ bash "$(npm prefix -g)/lib/node_modules/@pcircle/memesh/scripts/upgrade-plugin.s
 ```bash
 git clone https://github.com/PCIRCLE-AI/memesh-llm-memory
 cd memesh-llm-memory && npm install && npm run build
-npm test             # 630 tests
+npm test
 npm run test:e2e-dashboard
 ```
 
@@ -462,5 +462,5 @@ npm run test:e2e-dashboard
 ---
 
 <p align="center">
-  <strong>MIT</strong> — Made by <a href="https://pcircle.ai">PCIRCLE AI</a>
+  <strong>MIT</strong> — Made by <a href="https://pcircle.com">PCIRCLE AI</a>
 </p>

package/README.md

@@ -83,7 +83,7 @@ flowchart TB
 | Use the `/memesh` skill inside a Claude Code conversation | Path A (plugin) |
 | Get auto-capture (sessions → lessons → recall) in Claude Code | Path A (plugin) |
 | Run `memesh remember` / `memesh recall` / `memesh doctor` in any terminal | Path B (npm-global) |
-| Open the local dashboard via `memesh` (no `npx` lookup delay) | Path B (npm-global) |
+| Open the local dashboard via `memesh serve` (no `npx` lookup delay) | Path B (npm-global) |
 | Plug `memesh-mcp` into Cursor, Cline, or another MCP client | Path B (npm-global) |
 | All of the above | **Install both** — they don't conflict |
 
@@ -178,7 +178,7 @@ memesh doctor
 Open the dashboard to explore your memory:
 
 ```bash
-memesh
+memesh serve
 ```
 
 <p align="center">
@@ -202,7 +202,7 @@ memesh
 | **A developer using Claude Code** | Auto-recall project decisions, file-specific lessons, and past failures as you work |
 | **A coding-agent power user** | Share one local memory layer across MCP-compatible tools |
 | **A team experimenting with AI coding workflows** | Export/import project knowledge without introducing hosted infrastructure |
-| **An agent developer** | Add local memory through MCP, HTTP, CLI, or the Python SDK |
+| **An agent developer** | Add local memory through MCP, HTTP, or the CLI |
 
 ---
 
@@ -328,7 +328,7 @@ When npm flags an installed version as deprecated (typically a security advisory
 
 **⚠️ Conflict Detection** — If you have two memories that contradict each other, MeMesh warns you.
 
-**🕸️ Knowledge Graph Connectivity** — `memesh kg backfill-relations --all-rules` links orphan entities using tag co-occurrence, project clustering, session context, and name similarity — no LLM required. Reduces orphan rate from 89% to under 12% on a representative knowledge base.
+**🕸️ Knowledge Graph Connectivity** — `memesh kg backfill-relations --all-rules` links orphan entities using tag co-occurrence, project clustering, session context, and name similarity — no LLM required.
 
 **📦 Team Sharing** — `memesh export > team-knowledge.json` → share with your team → `memesh import team-knowledge.json`
 Imported bundles stay searchable, but MeMesh does not auto-inject imported memories into Claude hooks until you review or re-store them locally.
@@ -360,7 +360,7 @@ memesh config set llm.api-key sk-ant-...
 Or use the dashboard Settings tab (visual setup):
 
 ```bash
-memesh  # opens dashboard → Settings tab
+memesh serve  # opens dashboard → Settings tab
 ```
 
 ### Bring-your-own embeddings (optional)
@@ -376,7 +376,7 @@ The embedder is configured **independently of the chat LLM** — changing `llm.p
 
 | | Level 0 (default) | Level 1 (Smart Mode) |
 |---|---|---|
-| **Search** | FTS5 + sqlite-vec, 95.60% R@5 (~4ms per recall) | unchanged — recall is LLM-free at every level |
+| **Search** | FTS5 + sqlite-vec, 95.60% R@5 | unchanged — recall is LLM-free at every level |
 | **Auto-capture** | Rule-based patterns | + LLM extracts decisions & lessons |
 | **Auto-tagging** | Manual tags only | + LLM generates tags for new memories |
 | **Failure analysis** | Not available | + LLM converts session errors into structured lessons |
@@ -455,7 +455,7 @@ Session start surfaces a one-line banner (throttled to once per 24h per version)
 ```bash
 git clone https://github.com/PCIRCLE-AI/memesh-llm-memory
 cd memesh-llm-memory && npm install && npm run build
-npm test             # 630 tests
+npm test
 npm run test:e2e-dashboard
 ```
 
@@ -464,5 +464,5 @@ Dashboard: `cd dashboard && npm install && npm run dev`
 ---
 
 <p align="center">
-  <strong>MIT</strong> — Made by <a href="https://pcircle.ai">PCIRCLE AI</a>
+  <strong>MIT</strong> — Made by <a href="https://pcircle.com">PCIRCLE AI</a>
 </p>

package/README.pt.md

@@ -83,7 +83,7 @@ flowchart TB
 | Usar o skill `/memesh` numa conversa do Claude Code | Path A (plugin) |
 | Auto-captura no Claude Code (sessão → lições → recall seguinte) | Path A (plugin) |
 | Rodar `memesh remember` / `memesh recall` / `memesh doctor` em qualquer terminal | Path B (npm-global) |
-| Abrir o dashboard via `memesh` (sem atraso de inicialização do `npx`) | Path B (npm-global) |
+| Abrir o dashboard via `memesh serve` (sem atraso de inicialização do `npx`) | Path B (npm-global) |
 | Conectar `memesh-mcp` ao Cursor, Cline ou outro cliente MCP | Path B (npm-global) |
 | Tudo acima | **Instale ambos** — não conflitam |
 
@@ -149,7 +149,7 @@ memesh doctor
 Abra o dashboard para explorar sua memória:
 
 ```bash
-memesh
+memesh serve
 ```
 
 <p align="center">
@@ -173,7 +173,7 @@ memesh
 | **Um dev usando Claude Code** | Recuperar automaticamente decisões de projeto, lições por arquivo e falhas passadas enquanto trabalha |
 | **Um power user de agentes de código** | Compartilhar uma camada de memória local entre ferramentas compatíveis com MCP |
 | **Uma equipe experimentando workflows de IA para código** | Exportar/importar conhecimento de projeto sem precisar de infraestrutura hospedada |
-| **Um desenvolvedor de agentes** | Adicionar memória local via MCP, HTTP, CLI ou o SDK Python |
+| **Um desenvolvedor de agentes** | Adicionar memória local via MCP, HTTP ou CLI |
 
 ---
 
@@ -299,7 +299,7 @@ Quando o npm sinaliza uma versão instalada como depreciada (tipicamente um advi
 
 **⚠️ Detecção de Conflitos** — Se você tem duas memórias que se contradizem, MeMesh te avisa.
 
-**🕸️ Conectividade do grafo de conhecimento** — `memesh kg backfill-relations --all-rules` liga entidades órfãs usando co-ocorrência de tags, agrupamento de projetos, contexto de sessão e similaridade de nomes — sem LLM. Reduz a taxa de órfãos de 89% para menos de 12% numa base de conhecimento representativa.
+**🕸️ Conectividade do grafo de conhecimento** — `memesh kg backfill-relations --all-rules` liga entidades órfãs usando co-ocorrência de tags, agrupamento de projetos, contexto de sessão e similaridade de nomes — sem LLM.
 
 **📦 Compartilhamento em Equipe** — `memesh export > team-knowledge.json` → compartilhe com sua equipe → `memesh import team-knowledge.json`
 Bundles importados permanecem pesquisáveis, mas MeMesh não injeta automaticamente memórias importadas nos hooks do Claude até você revisar ou re-armazená-las localmente.
@@ -331,7 +331,7 @@ memesh config set llm.api-key sk-ant-...
 Ou use a aba Settings do dashboard (setup visual):
 
 ```bash
-memesh  # abre dashboard → aba Settings
+memesh serve  # abre dashboard → aba Settings
 ```
 
 ### Use seus próprios embeddings (opcional)
@@ -347,7 +347,7 @@ O embedder é configurado **independentemente do LLM de chat** — mudar `llm.pr
 
 | | Level 0 (padrão) | Level 1 (Smart Mode) |
 |---|---|---|
-| **Busca** | FTS5 + sqlite-vec, 95,60% R@5 (~4ms por recall) | inalterado — recall é LLM-free em todos os níveis |
+| **Busca** | FTS5 + sqlite-vec, 95,60% R@5 | inalterado — recall é LLM-free em todos os níveis |
 | **Auto-capture** | Padrões baseados em regras | + LLM extrai decisões & lições |
 | **Auto-tagging** | Apenas tags manuais | + LLM gera tags para novas memórias |
 | **Análise de falhas** | Não disponível | + LLM converte erros de sessão em structured lessons |
@@ -426,7 +426,7 @@ No início da sessão aparece um banner de uma linha (limitado a uma vez por 24h
 ```bash
 git clone https://github.com/PCIRCLE-AI/memesh-llm-memory
 cd memesh-llm-memory && npm install && npm run build
-npm test             # 630 tests
+npm test
 npm run test:e2e-dashboard
 ```
 
@@ -435,5 +435,5 @@ Dashboard: `cd dashboard && npm install && npm run dev`
 ---
 
 <p align="center">
-  <strong>MIT</strong> — Feito por <a href="https://pcircle.ai">PCIRCLE AI</a>
+  <strong>MIT</strong> — Feito por <a href="https://pcircle.com">PCIRCLE AI</a>
 </p>

package/README.th.md

@@ -83,7 +83,7 @@ flowchart TB
 | ใช้ skill `/memesh` ในการสนทนา Claude Code | Path A (plugin) |
 | Auto-capture ใน Claude Code (session → บทเรียน → recall ครั้งถัดไป) | Path A (plugin) |
 | รัน `memesh remember` / `memesh recall` / `memesh doctor` ใน terminal | Path B (npm-global) |
-| เปิด dashboard ผ่าน `memesh` (ไม่มีดีเลย์ของ `npx`) | Path B (npm-global) |
+| เปิด dashboard ผ่าน `memesh serve` (ไม่มีดีเลย์ของ `npx`) | Path B (npm-global) |
 | เสียบ `memesh-mcp` เข้ากับ Cursor, Cline หรือ MCP client อื่น | Path B (npm-global) |
 | ทั้งหมดข้างต้น | **ติดตั้งทั้งสอง** — ไม่ขัดแย้งกัน |
 
@@ -164,7 +164,7 @@ memesh doctor
 เปิดแดชบอร์ดเพื่อสำรวจหน่วยความจำ:
 
 ```bash
-memesh
+memesh serve
 ```
 
 <p align="center">
@@ -188,7 +188,7 @@ memesh
 | **นักพัฒนาที่ใช้ Claude Code** | เรียกคืนการตัดสินใจโครงการ บทเรียนเฉพาะไฟล์ และความล้มเหลวที่ผ่านมาโดยอัตโนมัติขณะทำงาน |
 | **ผู้ใช้เอเจนต์คิดโค้ดขั้นสูง** | ใช้ชั้นหน่วยความจำภายในตัวเดียวร่วมกันบนเครื่องมือที่รองรับ MCP |
 | **ทีมทดลองเวิร์กโฟลว์คิดโค้ด AI** | ส่งออก/นำเข้าความรู้โครงการโดยไม่ต้องสถาปัตยกรรมบนคลาวด์ |
-| **นักพัฒนาเอเจนต์** | เพิ่มหน่วยความจำภายในผ่าน MCP HTTP CLI หรือ Python SDK |
+| **นักพัฒนาเอเจนต์** | เพิ่มหน่วยความจำภายในผ่าน MCP HTTP หรือ CLI |
 
 ---
 
@@ -314,7 +314,7 @@ memesh export-schema \
 
 **⚠️ การตรวจจับความขัดแย้ง** — ถ้าคุณมีหน่วยความจำสองอันที่ขัดแย้งกัน MeMesh เตือนคุณ
 
-**🕸️ การเชื่อมต่อกราฟความรู้** — `memesh kg backfill-relations --all-rules` เชื่อมโยงเอนทิตีกำพร้าโดยใช้การร่วมเกิดของแท็ก การจัดกลุ่มโครงการ บริบทเซสชัน และความคล้ายคลึงของชื่อ — ไม่ต้องใช้ LLM ลดอัตราเด็กกำพร้าจาก 89% เป็นต่ำกว่า 12% บนฐานความรู้ที่เป็นตัวแทน
+**🕸️ การเชื่อมต่อกราฟความรู้** — `memesh kg backfill-relations --all-rules` เชื่อมโยงเอนทิตีกำพร้าโดยใช้การร่วมเกิดของแท็ก การจัดกลุ่มโครงการ บริบทเซสชัน และความคล้ายคลึงของชื่อ — ไม่ต้องใช้ LLM
 
 **📦 การแบ่งปันทีม** — `memesh export > team-knowledge.json` → แบ่งปันกับทีม → `memesh import team-knowledge.json`
 บันเดิลนำเข้าสามารถค้นหาได้ แต่ MeMesh ไม่ได้แทรกหน่วยความจำนำเข้าโดยอัตโนมัติลงใน hook Claude จนกว่าคุณจะตรวจสอบหรือเก็บเป็นท้องถิ่นอีกครั้ง
@@ -346,7 +346,7 @@ memesh config set llm.api-key sk-ant-...
 หรือใช้แท็บ Settings แดชบอร์ด (การตั้งค่าสายตา):
 
 ```bash
-memesh  # opens dashboard → Settings tab
+memesh serve  # opens dashboard → Settings tab
 ```
 
 ### ใช้ embeddings ของคุณเอง (ไม่บังคับ)
@@ -362,7 +362,7 @@ embedder ถูกตั้งค่า**แยกจาก LLM แชท** —
 
 | | ระดับ 0 (ค่าเริ่มต้น) | ระดับ 1 (Smart Mode) |
 |---|---|---|
-| **Search** | FTS5 + sqlite-vec, 95.60% R@5 (~4ms ต่อการ recall) | คงเดิม — recall ไม่ใช้ LLM ในทุก level |
+| **Search** | FTS5 + sqlite-vec, 95.60% R@5 | คงเดิม — recall ไม่ใช้ LLM ในทุก level |
 | **Auto-capture** | รูปแบบตามกฎ | + LLM สกัดการตัดสินใจ & บทเรียน |
 | **Auto-tagging** | แท็กด้วยตนเองเท่านั้น | + LLM สร้างแท็กให้ entity ใหม่ |
 | **วิเคราะห์ความล้มเหลว** | ไม่พร้อมใช้ | + LLM แปลง session errors เป็น structured lessons |
@@ -440,7 +440,7 @@ bash "$(npm prefix -g)/lib/node_modules/@pcircle/memesh/scripts/upgrade-plugin.s
 ```bash
 git clone https://github.com/PCIRCLE-AI/memesh-llm-memory
 cd memesh-llm-memory && npm install && npm run build
-npm test             # 630 tests
+npm test
 npm run test:e2e-dashboard
 ```
 
@@ -449,5 +449,5 @@ Dashboard: `cd dashboard && npm install && npm run dev`
 ---
 
 <p align="center">
-  <strong>MIT</strong> — สร้างโดย <a href="https://pcircle.ai">PCIRCLE AI</a>
+  <strong>MIT</strong> — สร้างโดย <a href="https://pcircle.com">PCIRCLE AI</a>
 </p>

package/README.vi.md

@@ -83,7 +83,7 @@ flowchart TB
 | Dùng skill `/memesh` trong cuộc trò chuyện Claude Code | Path A (plugin) |
 | Tự động capture trong Claude Code (session → bài học → recall lần sau) | Path A (plugin) |
 | Chạy `memesh remember` / `memesh recall` / `memesh doctor` ở bất kỳ terminal nào | Path B (npm-global) |
-| Mở dashboard qua `memesh` (không bị trễ khởi động `npx`) | Path B (npm-global) |
+| Mở dashboard qua `memesh serve` (không bị trễ khởi động `npx`) | Path B (npm-global) |
 | Cắm `memesh-mcp` vào Cursor, Cline hoặc client MCP khác | Path B (npm-global) |
 | Tất cả các mục trên | **Cài cả hai** — không xung đột |
 
@@ -149,7 +149,7 @@ memesh doctor
 Mở dashboard để khám phá bộ nhớ của bạn:
 
 ```bash
-memesh
+memesh serve
 ```
 
 <p align="center">
@@ -173,7 +173,7 @@ memesh
 | **Nhà phát triển sử dụng Claude Code** | Tự động gọi lại các quyết định dự án, bài học cụ thể theo file, và những lỗi trong quá khứ khi bạn làm việc |
 | **Power user của coding agent** | Chia sẻ một tầng bộ nhớ cục bộ qua các công cụ tương thích MCP |
 | **Nhóm thử nghiệm workflow AI coding** | Export/import kiến thức dự án mà không cần hạ tầng được quản lý |
-| **Nhà phát triển agent** | Thêm bộ nhớ cục bộ thông qua MCP, HTTP, CLI, hoặc Python SDK |
+| **Nhà phát triển agent** | Thêm bộ nhớ cục bộ thông qua MCP, HTTP, hoặc CLI |
 
 ---
 
@@ -299,7 +299,7 @@ Khi npm gắn cờ phiên bản đã cài là deprecated (thường là security
 
 **⚠️ Conflict Detection** — Nếu bạn có hai memories mâu thuẫn với nhau, MeMesh cảnh báo bạn.
 
-**🕸️ Kết nối đồ thị tri thức** — `memesh kg backfill-relations --all-rules` liên kết các thực thể cô lập bằng cách sử dụng đồng xuất hiện thẻ, phân cụm dự án, ngữ cảnh phiên và độ tương đồng tên — không cần LLM. Giảm tỷ lệ cô lập từ 89% xuống dưới 12% trên cơ sở tri thức đại diện.
+**🕸️ Kết nối đồ thị tri thức** — `memesh kg backfill-relations --all-rules` liên kết các thực thể cô lập bằng cách sử dụng đồng xuất hiện thẻ, phân cụm dự án, ngữ cảnh phiên và độ tương đồng tên — không cần LLM.
 
 **📦 Team Sharing** — `memesh export > team-knowledge.json` → chia sẻ với team → `memesh import team-knowledge.json`
 Các imported bundles vẫn có thể tìm kiếm được, nhưng MeMesh không auto-inject imported memories vào Claude hooks cho đến khi bạn review hoặc re-store chúng locally.
@@ -331,7 +331,7 @@ memesh config set llm.api-key sk-ant-...
 Hoặc dùng dashboard Settings tab (visual setup):
 
 ```bash
-memesh  # mở dashboard → Settings tab
+memesh serve  # mở dashboard → Settings tab
 ```
 
 ### Dùng embeddings của riêng bạn (tùy chọn)
@@ -347,7 +347,7 @@ Embedder được cấu hình **độc lập với LLM chat** — thay đổi `l
 
 | | Level 0 (default) | Level 1 (Smart Mode) |
 |---|---|---|
-| **Search** | FTS5 + sqlite-vec, 95.60% R@5 (~4ms mỗi recall) | giữ nguyên — recall luôn LLM-free ở mọi level |
+| **Search** | FTS5 + sqlite-vec, 95.60% R@5 | giữ nguyên — recall luôn LLM-free ở mọi level |
 | **Auto-capture** | Rule-based patterns | + LLM extracts decisions & lessons |
 | **Auto-tagging** | Chỉ thẻ thủ công | + LLM tự động gắn nhãn entity mới |
 | **Phân tích lỗi** | Không có sẵn | + LLM chuyển session errors thành structured lessons |
@@ -426,7 +426,7 @@ Khi bắt đầu session, banner một dòng (throttle mỗi 24h mỗi version)
 ```bash
 git clone https://github.com/PCIRCLE-AI/memesh-llm-memory
 cd memesh-llm-memory && npm install && npm run build
-npm test             # 630 tests
+npm test
 npm run test:e2e-dashboard
 ```
 
@@ -435,5 +435,5 @@ Dashboard: `cd dashboard && npm install && npm run dev`
 ---
 
 <p align="center">
-  <strong>MIT</strong> — Được tạo bởi <a href="https://pcircle.ai">PCIRCLE AI</a>
+  <strong>MIT</strong> — Được tạo bởi <a href="https://pcircle.com">PCIRCLE AI</a>
 </p>