node-red-contrib-knx-ultimate 4.3.5 → 4.3.7

package/nodes/locales/en/knxUltimateAI.json

@@ -3,9 +3,11 @@
     "title": "KNX AI (Traffic Analyzer)",
     "sections": {
       "capture": "Capture",
-      "analysis": "Analysis",
-      "anomalies": "Anomalies",
-      "llm": "LLM Assistant"
+      "storage": "Storage & Summary",
+      "detection": "Detection & Alerts",
+      "llmConnection": "AI Assistant Connection",
+      "llmContext": "AI Assistant Context",
+      "advanced": "Advanced Tuning"
     },
     "properties": {
       "server": "Gateway",
@@ -16,6 +18,8 @@
       "notifyreadrequest": "Capture GroupValue_Read",
       "analysisWindowSec": "Analysis window (seconds)",
       "historyWindowSec": "History window (seconds)",
+      "historyStoreToDisk": "Also archive captured telegrams to disk",
+      "historyStoreRetentionDays": "Disk archive retention (days)",
       "maxEvents": "Max stored events",
       "emitIntervalSec": "Auto emit summary (seconds, 0=off)",
       "topN": "Top list size",
@@ -33,17 +37,10 @@
       "llmApiKey": "API key",
       "llmModel": "Model",
       "llmSystemPrompt": "System prompt",
-      "llmTemperature": "Temperature",
-      "llmMaxTokens": "Max tokens",
-      "llmTimeoutMs": "Timeout (ms)",
-      "llmMaxEventsInPrompt": "Recent events included",
       "llmIncludeRaw": "Include raw payload hex",
-      "llmIncludeFlowContext": "Include Node-RED KNX node inventory",
-      "llmMaxFlowNodesInPrompt": "Max flow nodes included",
+      "llmIncludeFlowContext": "Include Node-RED project inventory",
       "llmIncludeDocsSnippets": "Include documentation snippets (help/README/examples)",
-      "llmDocsLanguage": "Docs language",
-      "llmDocsMaxSnippets": "Max docs snippets",
-      "llmDocsMaxChars": "Max docs chars"
+      "llmDocsLanguage": "Docs language"
     },
     "outputs": {
       "summary": "Summary/Stats",

package/nodes/locales/es/knxUltimateAI.html

@@ -31,6 +31,8 @@ Aquí tienes todos los campos tal como se muestran en el editor de KNX AI.
 ### Analysis
 - **Analysis window (seconds)**: ventana principal para resumen/rate.
 - **History window (seconds)**: ventana de retención del historial interno.
+- **Archivar tambien en disco los telegramas capturados**: guarda los telegramas tambien en `knxultimatestorage/knxai/history/<node-id>/YYYY-MM-DD.jsonl`, ademas de la RAM.
+- **Retencion del archivo en disco (dias)**: numero de dias que se conservan los archivos antes de borrarse automaticamente.
 - **Max stored events**: número máximo de telegramas en memoria.
 - **Auto emit summary (seconds, 0=off)**: intervalo periódico de resumen.
 - **Top list size**: cantidad de group addresses/fuentes en el top.
@@ -45,27 +47,32 @@ Aquí tienes todos los campos tal como se muestran en el editor de KNX AI.
 - **Flap window (seconds)**: ventana para detectar flapping/cambios rápidos.
 - **Max changes per GA in window (0=off)**: cambios máximos permitidos en ventana.
 
-### LLM Assistant
-- La pestaña **LLM Assistant** ahora aparece primero en el editor para facilitar la configuración.
+### Asistente IA
 - **Enable LLM assistant**: habilita funciones Ask/chat.
 - **Provider**: backend LLM (OpenAI-compatible u Ollama).
 - **Endpoint URL**: URL endpoint chat/completions.
 - **API key**: clave API (no requerida con Ollama local).
 - **Model**: ID/nombre de modelo.
-- **System prompt**: instrucción global para análisis KNX.
-- **Temperature**: temperatura de muestreo.
-- **Max tokens**: máximo de tokens de completado.
-- **Timeout (ms)**: timeout HTTP de peticiones LLM.
-- **Recent events included**: máximo de eventos recientes en el prompt.
+- **System prompt**: instrucción global para análisis KNX (Advanced).
+- Si el archivo en disco esta activo, **Ask** lo usa por defecto: respeta fechas/rangos explicitos y, si no los indicas, busca en las ultimas 24 horas mas los eventos actuales en RAM.
 - **Include raw payload hex**: incluye payload hex raw en el prompt.
-- **Include Node-RED KNX node inventory**: incluye inventario de flows en el prompt.
-- **Max flow nodes included**: límite de nodos flow incluidos.
+- **Incluir inventario del proyecto Node-RED**: incluye en el prompt el inventario de todo el proyecto Node-RED, con nodos KNX y otros nodos utiles como function/change/inject/template cuando contienen logica KNX o direcciones de grupo.
 - **Include documentation snippets (help/README/examples)**: incluye contexto de documentación.
 - **Docs language**: idioma preferido de snippets documentación.
-- **Max docs snippets**: máximo de snippets de documentación.
-- **Max docs chars**: máximo total de caracteres de documentación.
 - Botón **Refresh**: consulta el provider y carga modelos disponibles.
 
+### Advanced
+- **Analysis window (seconds)**: ventana principal para resumen/rate.
+- **Max stored events**: número máximo de telegramas en memoria.
+- **Top list size**: cantidad de group addresses/fuentes en el top.
+- **Pattern max lag (ms)**: diferencia temporal máxima para correlación.
+- **Pattern min occurrences**: ocurrencias mínimas antes de reportar patrón.
+- **Rate window (seconds)**: ventana deslizante para controles de rate.
+- **Max overall telegrams/sec (0=off)**: umbral en el bus global.
+- **Max telegrams/sec per GA (0=off)**: umbral por group address.
+- **Flap window (seconds)**: ventana para detectar flapping/cambios rápidos.
+- **Max changes per GA in window (0=off)**: cambios máximos permitidos en ventana.
+
 ### Configuración rápida de Ollama (local)
 - Selecciona **Provider = Ollama**.
 - Endpoint por defecto: `http://localhost:11434/api/chat`.

package/nodes/locales/es/knxUltimateAI.json

@@ -3,9 +3,11 @@
     "title": "KNX AI (Traffic Analyzer)",
     "sections": {
       "capture": "Capture",
-      "analysis": "Analysis",
-      "anomalies": "Anomalies",
-      "llm": "LLM Assistant"
+      "storage": "Historial y Resumen",
+      "detection": "Deteccion y Alertas",
+      "llmConnection": "Conexion del Asistente IA",
+      "llmContext": "Contexto del Asistente IA",
+      "advanced": "Ajustes Avanzados"
     },
     "properties": {
       "server": "Gateway",
@@ -16,6 +18,8 @@
       "notifyreadrequest": "Capture GroupValue_Read",
       "analysisWindowSec": "Analysis window (seconds)",
       "historyWindowSec": "History window (seconds)",
+      "historyStoreToDisk": "Archivar tambien en disco los telegramas capturados",
+      "historyStoreRetentionDays": "Retencion del archivo en disco (dias)",
       "maxEvents": "Max stored events",
       "emitIntervalSec": "Auto emit summary (seconds, 0=off)",
       "topN": "Top list size",
@@ -33,17 +37,10 @@
       "llmApiKey": "API key",
       "llmModel": "Model",
       "llmSystemPrompt": "System prompt",
-      "llmTemperature": "Temperature",
-      "llmMaxTokens": "Max tokens",
-      "llmTimeoutMs": "Timeout (ms)",
-      "llmMaxEventsInPrompt": "Recent events included",
       "llmIncludeRaw": "Include raw payload hex",
-      "llmIncludeFlowContext": "Include Node-RED KNX node inventory",
-      "llmMaxFlowNodesInPrompt": "Max flow nodes included",
+      "llmIncludeFlowContext": "Incluir inventario del proyecto Node-RED",
       "llmIncludeDocsSnippets": "Include documentation snippets (help/README/examples)",
-      "llmDocsLanguage": "Docs language",
-      "llmDocsMaxSnippets": "Max docs snippets",
-      "llmDocsMaxChars": "Max docs chars"
+      "llmDocsLanguage": "Docs language"
     },
     "outputs": {
       "summary": "Resumen/Estadísticas",

package/nodes/locales/fr/knxUltimateAI.html

@@ -31,6 +31,8 @@ Voici tous les champs tels qu'affichés dans l'éditeur KNX AI.
 ### Analysis
 - **Analysis window (seconds)** : fenêtre principale pour résumé/débits.
 - **History window (seconds)** : fenêtre de rétention de l'historique interne.
+- **Archiver aussi sur disque les telegrammes captures** : stocke aussi les télégrammes dans `knxultimatestorage/knxai/history/<node-id>/YYYY-MM-DD.jsonl`, en plus de la RAM.
+- **Retention de l'archive disque (jours)** : nombre de jours conservés sur disque avant suppression automatique des anciens fichiers.
 - **Max stored events** : nombre maximal de télégrammes en mémoire.
 - **Auto emit summary (seconds, 0=off)** : intervalle périodique d'émission du résumé.
 - **Top list size** : nombre de group addresses/sources dans le top.
@@ -45,27 +47,32 @@ Voici tous les champs tels qu'affichés dans l'éditeur KNX AI.
 - **Flap window (seconds)** : fenêtre de détection flapping/changements rapides.
 - **Max changes per GA in window (0=off)** : nombre max de changements autorisés.
 
-### LLM Assistant
-- L'onglet **LLM Assistant** est maintenant affiché en premier dans l'éditeur.
+### Assistant IA
 - **Enable LLM assistant** : active les fonctions Ask/chat.
 - **Provider** : backend LLM (OpenAI-compatible ou Ollama).
 - **Endpoint URL** : URL endpoint chat/completions.
 - **API key** : clé API (non requise avec Ollama local).
 - **Model** : ID/nom du modèle.
-- **System prompt** : instruction système globale pour l'analyse KNX.
-- **Temperature** : température d'échantillonnage.
-- **Max tokens** : nombre maximal de tokens de complétion.
-- **Timeout (ms)** : timeout HTTP des requêtes LLM.
-- **Recent events included** : nombre max d'événements récents dans le prompt.
+- **System prompt** : instruction système globale pour l'analyse KNX (Advanced).
+- Si l'archive disque est active, **Ask** l'utilise par défaut : les dates/plages explicites sont respectées, sinon l'assistant cherche sur les dernières 24 heures plus les événements RAM courants.
 - **Include raw payload hex** : inclut le payload hex brut dans le prompt.
-- **Include Node-RED KNX node inventory** : inclut l'inventaire des flows.
-- **Max flow nodes included** : limite de nœuds flow inclus.
+- **Inclure l'inventaire du projet Node-RED** : inclut dans le prompt l'inventaire de tout le projet Node-RED, avec les nœuds KNX et d'autres nœuds utiles comme function/change/inject/template lorsqu'ils contiennent de la logique KNX ou des adresses de groupe.
 - **Include documentation snippets (help/README/examples)** : inclut le contexte documentation.
 - **Docs language** : langue préférée des snippets documentation.
-- **Max docs snippets** : nombre max de snippets documentation.
-- **Max docs chars** : nombre total max de caractères documentation.
 - Bouton **Refresh** : interroge le provider et charge les modèles disponibles.
 
+### Advanced
+- **Analysis window (seconds)** : fenêtre principale pour résumé/débits.
+- **Max stored events** : nombre maximal de télégrammes en mémoire.
+- **Top list size** : nombre de group addresses/sources dans le top.
+- **Pattern max lag (ms)** : écart temporel max pour corrélation des patterns.
+- **Pattern min occurrences** : occurrences minimales avant signalement.
+- **Rate window (seconds)** : fenêtre glissante pour contrôles de débit.
+- **Max overall telegrams/sec (0=off)** : seuil sur le bus global.
+- **Max telegrams/sec per GA (0=off)** : seuil par group address.
+- **Flap window (seconds)** : fenêtre de détection flapping/changements rapides.
+- **Max changes per GA in window (0=off)** : nombre max de changements autorisés.
+
 ### Démarrage rapide Ollama (local)
 - Choisir **Provider = Ollama**.
 - Endpoint par défaut : `http://localhost:11434/api/chat`.

package/nodes/locales/fr/knxUltimateAI.json

@@ -3,9 +3,11 @@
     "title": "KNX AI (Traffic Analyzer)",
     "sections": {
       "capture": "Capture",
-      "analysis": "Analysis",
-      "anomalies": "Anomalies",
-      "llm": "LLM Assistant"
+      "storage": "Historique et Resume",
+      "detection": "Detection et Alertes",
+      "llmConnection": "Connexion Assistant IA",
+      "llmContext": "Contexte Assistant IA",
+      "advanced": "Reglages Avances"
     },
     "properties": {
       "server": "Gateway",
@@ -16,6 +18,8 @@
       "notifyreadrequest": "Capture GroupValue_Read",
       "analysisWindowSec": "Analysis window (seconds)",
       "historyWindowSec": "History window (seconds)",
+      "historyStoreToDisk": "Archiver aussi sur disque les telegrammes captures",
+      "historyStoreRetentionDays": "Retention de l'archive disque (jours)",
       "maxEvents": "Max stored events",
       "emitIntervalSec": "Auto emit summary (seconds, 0=off)",
       "topN": "Top list size",
@@ -33,17 +37,10 @@
       "llmApiKey": "API key",
       "llmModel": "Model",
       "llmSystemPrompt": "System prompt",
-      "llmTemperature": "Temperature",
-      "llmMaxTokens": "Max tokens",
-      "llmTimeoutMs": "Timeout (ms)",
-      "llmMaxEventsInPrompt": "Recent events included",
       "llmIncludeRaw": "Include raw payload hex",
-      "llmIncludeFlowContext": "Include Node-RED KNX node inventory",
-      "llmMaxFlowNodesInPrompt": "Max flow nodes included",
+      "llmIncludeFlowContext": "Inclure l'inventaire du projet Node-RED",
       "llmIncludeDocsSnippets": "Include documentation snippets (help/README/examples)",
-      "llmDocsLanguage": "Docs language",
-      "llmDocsMaxSnippets": "Max docs snippets",
-      "llmDocsMaxChars": "Max docs chars"
+      "llmDocsLanguage": "Docs language"
     },
     "outputs": {
       "summary": "Résumé/Stats",

package/nodes/locales/it/knxUltimateAI.html

@@ -31,6 +31,8 @@ Di seguito sono elencati tutti i campi presenti nell'editor del nodo KNX AI.
 ### Analisi
 - **Finestra analisi (secondi)**: finestra principale per summary/rate.
 - **Finestra storico (secondi)**: finestra di retention dello storico interno telegrammi.
+- **Archivia anche i telegrammi catturati su disco**: salva i telegrammi anche in `knxultimatestorage/knxai/history/<node-id>/YYYY-MM-DD.jsonl`, oltre che in RAM.
+- **Retention archivio su disco (giorni)**: numero di giorni mantenuti su disco prima della cancellazione automatica dei file piu' vecchi.
 - **Eventi massimi in memoria**: numero massimo di telegrammi mantenuti in RAM.
 - **Invia summary automatico (secondi, 0=off)**: intervallo di emissione summary periodica.
 - **Dimensione lista Top**: numero di group address/sorgenti nella classifica summary.
@@ -45,27 +47,32 @@ Di seguito sono elencati tutti i campi presenti nell'editor del nodo KNX AI.
 - **Finestra flap (secondi)**: finestra temporale per rilevare flapping/cambi rapidi.
 - **Max cambi per GA nella finestra (0=off)**: massimo numero di cambi consentiti.
 
-### Assistente LLM
-- Il tab **Assistente LLM** è mostrato per primo nell'editor, per semplificare la configurazione.
+### Assistente AI
 - **Abilita assistente LLM**: abilita funzioni Ask/chat.
 - **Provider**: backend LLM (OpenAI-compatible o Ollama).
 - **URL endpoint**: URL endpoint chat/completions.
 - **API key**: chiave API (non necessaria con Ollama locale).
 - **Modello**: ID/nome modello.
-- **Prompt di sistema**: istruzione globale del comportamento analisi KNX.
-- **Temperatura**: temperatura di campionamento.
-- **Token massimi**: massimo numero di token di completamento.
-- **Timeout (ms)**: timeout HTTP richieste LLM.
-- **Eventi recenti inclusi**: massimo eventi recenti passati al prompt.
+- **Prompt di sistema**: istruzione globale del comportamento analisi KNX (Advanced).
+- Se l'archivio su disco e' attivo, **Ask** lo usa di default: rispetta date/intervalli espliciti e, se non presenti, cerca nelle ultime 24 ore piu' gli eventi correnti in RAM.
 - **Includi payload raw in hex**: include payload raw esadecimale nel prompt.
-- **Includi inventario nodi KNX nei flow**: include inventario flow nel prompt.
-- **Max nodi del flow inclusi**: limite nodi inclusi dall'inventario flow.
+- **Includi inventario del progetto Node-RED**: include nel prompt l'inventario dell'intero progetto Node-RED, compresi nodi KNX e altri nodi utili come function/change/inject/template quando contengono logica KNX o group address.
 - **Includi estratti documentazione (help/README/esempi)**: include contesto docs.
 - **Lingua documentazione**: lingua preferita per gli estratti docs.
-- **Max estratti docs**: numero massimo di estratti docs.
-- **Max caratteri docs**: massimo totale caratteri docs.
 - Pulsante **Aggiorna**: interroga il provider e popola i modelli disponibili.
 
+### Advanced
+- **Finestra analisi (secondi)**: finestra principale per summary/rate.
+- **Eventi massimi in memoria**: numero massimo di telegrammi mantenuti in RAM.
+- **Dimensione lista Top**: numero di group address/sorgenti nella classifica summary.
+- **Ritardo massimo pattern (ms)**: differenza temporale massima per correlare pattern.
+- **Occorrenze minime pattern**: soglia minima prima di segnalare un pattern.
+- **Finestra rate (secondi)**: finestra scorrevole per i controlli di rate.
+- **Max telegrammi/sec totale (0=off)**: soglia telegrammi/s sull'intero BUS.
+- **Max telegrammi/sec per GA (0=off)**: soglia telegrammi/s per singolo group address.
+- **Finestra flap (secondi)**: finestra temporale per rilevare flapping/cambi rapidi.
+- **Max cambi per GA nella finestra (0=off)**: massimo numero di cambi consentiti.
+
 ### Setup rapido Ollama (locale)
 - Seleziona **Provider = Ollama**.
 - Endpoint predefinito: `http://localhost:11434/api/chat`.

package/nodes/locales/it/knxUltimateAI.json

@@ -3,9 +3,11 @@
     "title": "KNX AI (Analisi Traffico)",
     "sections": {
       "capture": "Cattura",
-      "analysis": "Analisi",
-      "anomalies": "Anomalie",
-      "llm": "Assistente LLM"
+      "storage": "Storico e Summary",
+      "detection": "Rilevamento e Avvisi",
+      "llmConnection": "Connessione Assistente AI",
+      "llmContext": "Contesto Assistente AI",
+      "advanced": "Tuning Avanzato"
     },
     "properties": {
       "server": "Gateway",
@@ -16,6 +18,8 @@
       "notifyreadrequest": "Cattura GroupValue_Read",
       "analysisWindowSec": "Finestra analisi (secondi)",
       "historyWindowSec": "Finestra storico (secondi)",
+      "historyStoreToDisk": "Archivia anche i telegrammi catturati su disco",
+      "historyStoreRetentionDays": "Retention archivio su disco (giorni)",
       "maxEvents": "Eventi massimi in memoria",
       "emitIntervalSec": "Invia summary automatico (secondi, 0=off)",
       "topN": "Dimensione lista Top",
@@ -33,17 +37,10 @@
       "llmApiKey": "API key",
       "llmModel": "Modello",
       "llmSystemPrompt": "Prompt di sistema",
-      "llmTemperature": "Temperatura",
-      "llmMaxTokens": "Token massimi",
-      "llmTimeoutMs": "Timeout (ms)",
-      "llmMaxEventsInPrompt": "Eventi recenti inclusi",
       "llmIncludeRaw": "Includi payload raw in hex",
-      "llmIncludeFlowContext": "Includi inventario nodi KNX nei flow",
-      "llmMaxFlowNodesInPrompt": "Max nodi del flow inclusi",
+      "llmIncludeFlowContext": "Includi inventario del progetto Node-RED",
       "llmIncludeDocsSnippets": "Includi estratti documentazione (help/README/esempi)",
-      "llmDocsLanguage": "Lingua documentazione",
-      "llmDocsMaxSnippets": "Max estratti docs",
-      "llmDocsMaxChars": "Max caratteri docs"
+      "llmDocsLanguage": "Lingua documentazione"
     },
     "outputs": {
       "summary": "Summary/Statistiche",

package/nodes/locales/zh-CN/knxUltimateAI.html

@@ -31,6 +31,8 @@
 ### Analysis
 - **Analysis window (seconds)**：摘要/速率统计主窗口。
 - **History window (seconds)**：内部历史保留窗口。
+- **同时将捕获的报文归档到磁盘**：除了 RAM，还会把报文保存到 `knxultimatestorage/knxai/history/<node-id>/YYYY-MM-DD.jsonl`。
+- **磁盘归档保留天数**：归档文件在磁盘上保留的天数，超期后旧文件会自动删除。
 - **Max stored events**：内存中保留的最大电报数量。
 - **Auto emit summary (seconds, 0=off)**：周期性输出摘要间隔。
 - **Top list size**：摘要中 top 组地址/来源数量。
@@ -45,27 +47,32 @@
 - **Flap window (seconds)**：抖动/快速变化检测窗口。
 - **Max changes per GA in window (0=off)**：窗口内允许的最大变化次数。
 
-### LLM Assistant
-- **LLM Assistant** 标签页现在固定在编辑器最前面，便于快速配置。
+### AI 助手
 - **Enable LLM assistant**：启用 Ask/chat 功能。
 - **Provider**：LLM 后端（OpenAI-compatible 或 Ollama）。
 - **Endpoint URL**：chat/completions 接口 URL。
 - **API key**：API Key（本地 Ollama 可不填）。
 - **Model**：模型 ID/名称。
-- **System prompt**：KNX 分析全局系统提示词。
-- **Temperature**：采样温度。
-- **Max tokens**：最大生成 token 数。
-- **Timeout (ms)**：LLM 请求 HTTP 超时。
-- **Recent events included**：提示词中包含的最近事件数量上限。
+- **System prompt**：KNX 分析全局系统提示词（Advanced）。
+- 如果启用了磁盘归档，**Ask** 默认会查询该归档：若问题里写了明确日期/时间范围就按其查询，否则默认查询最近 24 小时并补上当前 RAM 事件。
 - **Include raw payload hex**：在提示词中包含原始十六进制 payload。
-- **Include Node-RED KNX node inventory**：在提示词中包含 flow 清单。
-- **Max flow nodes included**：flow 清单中最多包含的节点数。
+- **包含 Node-RED 项目清单**：在提示词中加入整个 Node-RED 项目的节点清单，不仅包含 KNX 节点，也包含 function/change/inject/template 等在内且带有 KNX 逻辑或组地址的有用节点。
 - **Include documentation snippets (help/README/examples)**：在提示词中包含文档片段。
 - **Docs language**：文档片段优先语言。
-- **Max docs snippets**：文档片段最大条数。
-- **Max docs chars**：文档片段最大总字符数。
 - **Refresh** 按钮：请求 provider 并加载可用模型 ID。
 
+### Advanced
+- **Analysis window (seconds)**：摘要/速率统计主窗口。
+- **Max stored events**：内存中保留的最大电报数量。
+- **Top list size**：摘要中 top 组地址/来源数量。
+- **Pattern max lag (ms)**：模式关联允许的最大时间差。
+- **Pattern min occurrences**：报告模式前的最小出现次数。
+- **Rate window (seconds)**：异常速率检查滑动窗口。
+- **Max overall telegrams/sec (0=off)**：全总线 telegram/s 阈值。
+- **Max telegrams/sec per GA (0=off)**：单组地址 telegram/s 阈值。
+- **Flap window (seconds)**：抖动/快速变化检测窗口。
+- **Max changes per GA in window (0=off)**：窗口内允许的最大变化次数。
+
 ### Ollama 快速配置（本地）
 - 选择 **Provider = Ollama**。
 - 默认 endpoint：`http://localhost:11434/api/chat`。

package/nodes/locales/zh-CN/knxUltimateAI.json

@@ -3,9 +3,11 @@
     "title": "KNX AI (Traffic Analyzer)",
     "sections": {
       "capture": "Capture",
-      "analysis": "Analysis",
-      "anomalies": "Anomalies",
-      "llm": "LLM Assistant"
+      "storage": "历史与摘要",
+      "detection": "检测与告警",
+      "llmConnection": "AI 助手连接",
+      "llmContext": "AI 助手上下文",
+      "advanced": "高级调优"
     },
     "properties": {
       "server": "Gateway",
@@ -16,6 +18,8 @@
       "notifyreadrequest": "Capture GroupValue_Read",
       "analysisWindowSec": "Analysis window (seconds)",
       "historyWindowSec": "History window (seconds)",
+      "historyStoreToDisk": "同时将捕获的报文归档到磁盘",
+      "historyStoreRetentionDays": "磁盘归档保留天数",
       "maxEvents": "Max stored events",
       "emitIntervalSec": "Auto emit summary (seconds, 0=off)",
       "topN": "Top list size",
@@ -33,17 +37,10 @@
       "llmApiKey": "API key",
       "llmModel": "Model",
       "llmSystemPrompt": "System prompt",
-      "llmTemperature": "Temperature",
-      "llmMaxTokens": "Max tokens",
-      "llmTimeoutMs": "Timeout (ms)",
-      "llmMaxEventsInPrompt": "Recent events included",
       "llmIncludeRaw": "Include raw payload hex",
-      "llmIncludeFlowContext": "Include Node-RED KNX node inventory",
-      "llmMaxFlowNodesInPrompt": "Max flow nodes included",
+      "llmIncludeFlowContext": "包含 Node-RED 项目清单",
       "llmIncludeDocsSnippets": "Include documentation snippets (help/README/examples)",
-      "llmDocsLanguage": "Docs language",
-      "llmDocsMaxSnippets": "Max docs snippets",
-      "llmDocsMaxChars": "Max docs chars"
+      "llmDocsLanguage": "Docs language"
     },
     "outputs": {
       "summary": "摘要/统计",

package/package.json

@@ -3,7 +3,7 @@
   "engines": {
     "node": ">=20.18.1"
   },
-  "version": "4.3.5",
+  "version": "4.3.7",
   "description": "Control your KNX and KNX Secure intallation via Node-Red! A bunch of KNX nodes, with integrated Philips HUE control, ETS group address importer, and KNX routing between interfaces. Easy to use and highly configurable.",
   "files": [
     "nodes/",