n8n-nodes-tembory 1.0.9 → 1.0.12

package/README.md

@@ -2,10 +2,14 @@
 
 Node de memoria operacional da Tembory para agentes de IA no n8n.
 
-Versao atual: `1.0.1`.
+Versao atual: `1.0.12`.
 
 O Tembory entrega contexto rico para o AI Agent sem depender apenas do historico textual da conversa. Ele combina memoria semantica, working memory, decision state, fatos estaveis do lead, historico de tools, estado operacional, action ledger, timeline de entidades, compressao de memoria, grafo, mensagens recentes e diagnosticos.
 
+Nos presets de producao, o contexto e organizado como um pacote acionavel: secoes vazias sao removidas, evidencias de tools ficam canonicas em `tool_history`/`operational_state`, e blocos como `decision_state`, `working_memory` e `memory_compression` sao compactados sem perder IDs, proximas acoes, slots, reservas, confirmacoes e bloqueios de repeticao.
+
+Quando um modelo barato esta conectado ao Tembory, os presets de producao usam esse SLM para gerar um resumo ativo do contexto. Em modo `auto`, ele organiza tanto o estado atual (`working_memory`, `tool_history`, `operational_state`, `decision_state`) quanto memorias vetoriais recuperadas. Assim o primeiro turno ja pode receber highlights uteis mesmo antes de haver resultados do banco vetorial.
+
 ## Smoke tecnico
 
 Antes de publicar uma versao, rode:

package/dist/nodes/Mem0/Mem0Memory.node.js

@@ -27,7 +27,7 @@ const snapshotJson = (value) => {
     }
 };
 const truncate = (value, max = MAX_TEXT) => {
-    const text = typeof value === 'string' ? value : safeStringify(value);
+    const text = value === undefined || value === null ? '' : (typeof value === 'string' ? value : safeStringify(value));
     if (text.length <= max)
         return text;
     return `${text.slice(0, max)}\n... [truncated ${text.length - max} chars]`;
@@ -53,7 +53,7 @@ const saveClientVectorMemories = async (ctx, memories, ids = {}) => {
     const valid = memories.filter((memory) => String(memory === null || memory === void 0 ? void 0 : memory.text || '').trim());
     if (!valid.length)
         return null;
-    return GenericFunctions_1.mem0ApiRequest.call(ctx, 'POST', '/elefai/v1/vector-memories', {
+    return GenericFunctions_1.mem0ApiRequest.call(ctx, 'POST', '/tembory/v1/vector-memories', {
         memories: valid,
         user_id: ids.user_id,
         agent_id: ids.agent_id,
@@ -64,7 +64,7 @@ const searchClientVectorMemories = async (ctx, embedding, query, body = {}) => {
     const text = String(query || '').trim();
     if (!text)
         return { results: [] };
-    return GenericFunctions_1.mem0ApiRequest.call(ctx, 'POST', '/elefai/v1/vector-memories/search', {
+    return GenericFunctions_1.mem0ApiRequest.call(ctx, 'POST', '/tembory/v1/vector-memories/search', {
         ...body,
         query: text,
         embedding: await embedQueryCached(embedding, text),
@@ -167,11 +167,11 @@ const scoreOf = (m) => {
     var _a, _b, _c, _d;
     return (_d = (_c = (_b = (_a = m === null || m === void 0 ? void 0 : m.score) !== null && _a !== void 0 ? _a : m === null || m === void 0 ? void 0 : m.similarity) !== null && _b !== void 0 ? _b : m === null || m === void 0 ? void 0 : m.relevance) !== null && _c !== void 0 ? _c : m === null || m === void 0 ? void 0 : m.distance) !== null && _d !== void 0 ? _d : undefined;
 };
-const scoreMetaOf = (m) => (m && typeof m === 'object' && (m.elefaiScore || m._elefaiScore)) || {};
-const withElefaiScore = (m, elefaiScore) => {
+const scoreMetaOf = (m) => (m && typeof m === 'object' && (m.temboryScore || m._temboryScore)) || {};
+const withTemboryScore = (m, temboryScore) => {
     if (!m || typeof m !== 'object')
         return m;
-    return { ...m, elefaiScore };
+    return { ...m, temboryScore };
 };
 const normalizeFactValue = (value) => String(value || '').replace(/\*\*/g, '').replace(/\s+/g, ' ').replace(/[.,;:]+$/g, '').trim();
 const profileSourceRank = (source = '') => {
@@ -504,8 +504,8 @@ const deriveEntityTimeline = (profileFacts = {}, graph = [], recentMessages = []
     }
     return pruneByLimit(deduped.sort((a, b) => String(a.at || '').localeCompare(String(b.at || ''))), maxItems || 24);
 };
-const RECENT_MESSAGE_MARKER = '__elefai_recent_message_v1__';
-const TOOL_HISTORY_MARKER = '__elefai_tool_history_v1__';
+const RECENT_MESSAGE_MARKER = '__tembory_recent_message_v1__';
+const TOOL_HISTORY_MARKER = '__tembory_tool_history_v1__';
 const encodeRecentMessage = (recent, threadId) => `${RECENT_MESSAGE_MARKER}${safeStringify({
     role: recent.role || 'user',
     content: recent.content || '',
@@ -855,21 +855,21 @@ const extractToolCalls = (outputValues = {}) => {
 const getMemoryStore = (ctx) => {
     try {
         const data = ctx.getWorkflowStaticData('global');
-        data.elefaiBrain = data.elefaiBrain || {};
-        data.elefaiBrain.toolHistory = data.elefaiBrain.toolHistory || {};
-        data.elefaiBrain.recentMessages = data.elefaiBrain.recentMessages || {};
-        data.elefaiBrain.profileFacts = data.elefaiBrain.profileFacts || {};
-        data.elefaiBrain.workingMemory = data.elefaiBrain.workingMemory || {};
-        data.elefaiBrain.decisionState = data.elefaiBrain.decisionState || {};
-        data.elefaiBrain.memoryCompression = data.elefaiBrain.memoryCompression || {};
-        return data.elefaiBrain;
+        data.tembory = data.tembory || {};
+        data.tembory.toolHistory = data.tembory.toolHistory || {};
+        data.tembory.recentMessages = data.tembory.recentMessages || {};
+        data.tembory.profileFacts = data.tembory.profileFacts || {};
+        data.tembory.workingMemory = data.tembory.workingMemory || {};
+        data.tembory.decisionState = data.tembory.decisionState || {};
+        data.tembory.memoryCompression = data.tembory.memoryCompression || {};
+        return data.tembory;
     }
     catch {
-        global.__elefaiBrainMemory = global.__elefaiBrainMemory || { toolHistory: {}, recentMessages: {}, profileFacts: {}, workingMemory: {}, decisionState: {}, memoryCompression: {} };
-        global.__elefaiBrainMemory.workingMemory = global.__elefaiBrainMemory.workingMemory || {};
-        global.__elefaiBrainMemory.decisionState = global.__elefaiBrainMemory.decisionState || {};
-        global.__elefaiBrainMemory.memoryCompression = global.__elefaiBrainMemory.memoryCompression || {};
-        return global.__elefaiBrainMemory;
+        global.__temboryMemory = global.__temboryMemory || { toolHistory: {}, recentMessages: {}, profileFacts: {}, workingMemory: {}, decisionState: {}, memoryCompression: {} };
+        global.__temboryMemory.workingMemory = global.__temboryMemory.workingMemory || {};
+        global.__temboryMemory.decisionState = global.__temboryMemory.decisionState || {};
+        global.__temboryMemory.memoryCompression = global.__temboryMemory.memoryCompression || {};
+        return global.__temboryMemory;
     }
 };
 const namespacePart = (value) => String(value || '').trim().replace(/\s+/g, '_').replace(/[:|]+/g, '-');
@@ -882,6 +882,8 @@ const applyOperationalPreset = (advanced = {}) => {
     const preset = String(advanced.operationPreset || 'custom');
     const presets = {
         diagnostic: {
+            summarySource: 'auto',
+            includeConnectedModelSummary: true,
             includeContextHeader: true,
             includeSummary: true,
             includeScores: true,
@@ -906,6 +908,9 @@ const applyOperationalPreset = (advanced = {}) => {
             recentMessagesLastN: 8,
         },
         productionBalanced: {
+            summarySource: 'auto',
+            includeConnectedModelSummary: true,
+            compactStateSections: true,
             includeContextHeader: true,
             includeSummary: true,
             includeScores: false,
@@ -928,8 +933,15 @@ const applyOperationalPreset = (advanced = {}) => {
             lastN: 8,
             toolHistoryLastN: 10,
             recentMessagesLastN: 6,
+            vectorMemoryMaxChars: 360,
+            contextMaxChars: 10000,
+            connectedModelSummaryMaxChars: 900,
+            connectedModelSummaryInputMaxChars: 3200,
         },
         productionCheap: {
+            summarySource: 'activeContext',
+            includeConnectedModelSummary: true,
+            compactStateSections: true,
             includeContextHeader: true,
             includeSummary: false,
             includeScores: false,
@@ -952,8 +964,14 @@ const applyOperationalPreset = (advanced = {}) => {
             lastN: 4,
             toolHistoryLastN: 5,
             recentMessagesLastN: 4,
+            vectorMemoryMaxChars: 260,
+            contextMaxChars: 7000,
+            connectedModelSummaryMaxChars: 700,
+            connectedModelSummaryInputMaxChars: 2400,
         },
         productionNano: {
+            summarySource: 'auto',
+            includeConnectedModelSummary: true,
             compactForAgent: true,
             includeContextHeader: true,
             includeSummary: true,
@@ -980,8 +998,12 @@ const applyOperationalPreset = (advanced = {}) => {
             recentMessagesLastN: 2,
             vectorMemoryMaxChars: 220,
             contextMaxChars: 6000,
+            connectedModelSummaryMaxChars: 700,
+            connectedModelSummaryInputMaxChars: 2200,
         },
         audit: {
+            summarySource: 'auto',
+            includeConnectedModelSummary: true,
             includeContextHeader: true,
             includeSummary: true,
             includeScores: true,
@@ -1328,7 +1350,7 @@ const deriveContextHealth = ({ userId = '', project = '', vectorMemories = [], r
     const missing = Object.entries(checks).filter(([, ok]) => !ok).map(([key]) => key);
     const agenda = operationalState.agenda_state || {};
     return {
-        kind: 'elefai.brain.context_health.v1',
+        kind: 'tembory.context_health.v1',
         namespace: userId || null,
         project: project || null,
         quality_score: Math.max(0, Math.min(1000, score)),
@@ -1409,12 +1431,63 @@ const compactToolResult = (result, max = 360) => {
     return truncate(text, max);
 };
 const compactToolHistoryForAgent = (toolHistory = [], maxItems = 6, includeResults = true) => pruneByLimit(toolHistory || [], maxItems).map((tool) => ({
+    id: tool.call_id || tool.id,
     name: tool.name,
     status: tool.ok === false ? 'failed' : 'ok',
     at: tool.at,
     input: truncate(String(tool.input || ''), 180) || undefined,
     result: includeResults ? compactToolResult(tool.result, 360) : undefined,
 }));
+const cleanContextValue = (value) => {
+    if (Array.isArray(value)) {
+        return value.map(cleanContextValue).filter((item) => item !== undefined && item !== null && !(Array.isArray(item) && item.length === 0) && !(typeof item === 'object' && !Array.isArray(item) && Object.keys(item).length === 0));
+    }
+    if (!value || typeof value !== 'object')
+        return value === '' ? undefined : value;
+    const out = {};
+    for (const [key, item] of Object.entries(value)) {
+        const cleaned = cleanContextValue(item);
+        if (cleaned === undefined || cleaned === null)
+            continue;
+        if (Array.isArray(cleaned) && cleaned.length === 0)
+            continue;
+        if (typeof cleaned === 'object' && !Array.isArray(cleaned) && Object.keys(cleaned).length === 0)
+            continue;
+        out[key] = cleaned;
+    }
+    return out;
+};
+const compactWorkingMemoryForAgent = (memory = {}) => cleanContextValue({
+    current_goal: memory.current_goal,
+    current_task: memory.current_task,
+    last_user_intent: memory.last_user_intent,
+    last_user_message: truncate(memory.last_user_message, 220),
+    active_entities: memory.active_entities,
+    open_decisions: memory.open_decisions,
+    last_error: memory.last_error,
+    next_expected_action: memory.next_expected_action,
+    updated_at: memory.updated_at,
+});
+const compactDecisionStateForAgent = (state = {}) => cleanContextValue({
+    current_intent: state.current_intent,
+    active_decisions: (state.active_decisions || []).slice(-4).map((decision) => cleanContextValue({
+        id: decision.id,
+        status: decision.status,
+        decision: truncate(decision.decision || decision.summary || decision.reason, 260),
+        tool: decision.tool || decision.next_tool || decision.required_tool,
+        reservation_id: decision.reservation_id,
+        confirmation_id: decision.confirmation_id,
+        at: decision.at || decision.updated_at,
+    })),
+    do_not_repeat_tools: state.do_not_repeat_tools,
+    agenda_decision_state: state.agenda_decision_state,
+    latest_tool: state.latest_tool ? cleanContextValue({
+        name: state.latest_tool.name,
+        status: state.latest_tool.status || (state.latest_tool.ok === false ? 'failed' : 'ok'),
+        at: state.latest_tool.at,
+    }) : undefined,
+    conflict_policy: state.conflict_policy,
+});
 const compactOperationalStateForAgent = (state = {}) => {
     const agenda = state.agenda_state || {};
     return {
@@ -1444,6 +1517,21 @@ const compactMemoryCompressionForAgent = (compression = {}) => ({
     } : undefined,
     active_memory_count: ((compression.workflow_summary || {}).active_memory_count) || 0,
 });
+const compactActionLedgerForAgent = (ledger = [], maxItems = 6, includeResults = true) => pruneByLimit(ledger || [], maxItems).map((item) => cleanContextValue({
+    action_id: item.action_id,
+    kind: item.kind,
+    tool: item.tool || item.name,
+    status: item.status,
+    at: item.at,
+    result: includeResults ? compactToolResult(item.result, 260) : undefined,
+}));
+const compactEntityTimelineForAgent = (timeline = [], maxItems = 8) => pruneByLimit(timeline || [], maxItems).map((item) => cleanContextValue({
+    entity: item.entity || item.source || item.name,
+    event: truncate(item.event || item.fact || item.relation || item.kind || item.value, 220),
+    target: item.target,
+    at: item.at,
+    source: item.source_type || item.source,
+}));
 const compactVectorMemoriesForAgent = (vectorMemories = [], toolHistory = [], maxItems = 3) => {
     const hasStructuredTools = Array.isArray(toolHistory) && toolHistory.length > 0;
     return (vectorMemories || [])
@@ -1452,6 +1540,87 @@ const compactVectorMemoriesForAgent = (vectorMemories = [], toolHistory = [], ma
         .filter((text) => !hasStructuredTools || !/^\[tool_events_extracted\]/i.test(text))
         .slice(0, maxItems);
 };
+const modelResponseText = (response) => {
+    const content = response === null || response === void 0 ? void 0 : response.content;
+    if (typeof content === 'string')
+        return content;
+    if (Array.isArray(content)) {
+        return content.map((item) => {
+            if (typeof item === 'string')
+                return item;
+            if (item && typeof item === 'object' && typeof item.text === 'string')
+                return item.text;
+            if (item && typeof item === 'object' && typeof item.content === 'string')
+                return item.content;
+            return '';
+        }).filter(Boolean).join('\n');
+    }
+    if (typeof (response === null || response === void 0 ? void 0 : response.text) === 'string')
+        return response.text;
+    return response === undefined || response === null ? '' : String(response);
+};
+const cleanModelSummaryText = (value, max = 900) => {
+    let text = modelResponseText(value).trim();
+    if (!text)
+        return '';
+    try {
+        const parsed = JSON.parse(text);
+        text = modelResponseText(parsed) || (Array.isArray(parsed) ? parsed.map(modelResponseText).filter(Boolean).join('\n') : '');
+        if (!text && parsed && typeof parsed === 'object') {
+            const bullets = parsed.highlights || parsed.current_summary || parsed.summary || parsed.facts;
+            if (Array.isArray(bullets))
+                text = bullets.map((item) => `- ${typeof item === 'string' ? item : safeStringify(item)}`).join('\n');
+        }
+    }
+    catch { }
+    text = text
+        .replace(/^\s*\[\s*\{\s*"type"\s*:\s*"text"\s*,\s*"text"\s*:\s*"/, '')
+        .replace(/"\s*,\s*"annotations"\s*:\s*\[\s*\]\s*\}\s*\]\s*$/, '')
+        .replace(/\\n/g, '\n')
+        .trim();
+    return truncate(text, max);
+};
+const buildConnectedModelSummaryInput = ({ query, profileFacts, workingMemory, decisionState, memoryCompression, operationalState, toolHistory, recentMessages, vectorMemories, highlights, adv }) => {
+    const source = String(adv.summarySource || 'auto');
+    if (source === 'off' || source === 'disabled' || adv.includeConnectedModelSummary === false)
+        return '';
+    const includeVectors = source === 'auto' || source === 'vectorOnly' || source === 'vectors';
+    const includeActive = source === 'auto' || source === 'activeContext' || source === 'active';
+    const payload = cleanContextValue({
+        query: String(query || ''),
+        active_context: includeActive ? {
+            profile_facts: renderProfileFacts(profileFacts),
+            working_memory: compactWorkingMemoryForAgent(workingMemory || {}),
+            decision_state: compactDecisionStateForAgent(decisionState || {}),
+            operational_state: compactOperationalStateForAgent(operationalState || {}),
+            memory_compression: compactMemoryCompressionForAgent(memoryCompression || {}),
+            tool_history: compactToolHistoryForAgent(toolHistory || [], adv.summaryToolHistoryLastN || adv.toolHistoryLastN || 6, adv.includeToolResults !== false),
+            recent_messages: pruneByLimit(recentMessages || [], adv.summaryRecentMessagesLastN || 4).map((message) => cleanContextValue({
+                role: message.role,
+                content: truncate(message.content, 240),
+                at: message.at,
+            })),
+            highlights: (highlights || []).slice(0, adv.summaryHighlightsMaxItems || 4),
+        } : undefined,
+        vector_memories: includeVectors ? compactVectorMemoriesForAgent(vectorMemories || [], toolHistory || [], adv.summaryMaxFacts || 4) : undefined,
+    });
+    const hasActiveSignal = includeActive && Boolean((payload.active_context && Object.keys(payload.active_context).length) || payload.query);
+    const hasVectorSignal = includeVectors && Array.isArray(payload.vector_memories) && payload.vector_memories.length > 0;
+    if (!hasActiveSignal && !hasVectorSignal)
+        return '';
+    return truncate(safeStringify(payload), Number(adv.connectedModelSummaryInputMaxChars || 3000));
+};
+const invokeConnectedModelSummary = async (connectedLanguageModel, summaryInput, adv = {}) => {
+    if (!connectedLanguageModel || typeof connectedLanguageModel.invoke !== 'function' || !summaryInput)
+        return '';
+    const response = await connectedLanguageModel.invoke([
+        toBaseMessage({
+            role: 'user',
+            content: `Organize the Tembory memory context for the next agent turn. Return concise Portuguese bullets only. Preserve IDs, dates, tool names, next action, do-not-repeat instructions, and contradictions. Do not invent facts.\n\nContext:\n${summaryInput}`,
+        }),
+    ]);
+    return cleanModelSummaryText(response, Number(adv.connectedModelSummaryMaxChars || 900));
+};
 const contextSizeOfMessages = (messages = []) => {
     const perMessage = (messages || []).map((message, index) => {
         const content = String(message.content || '');
@@ -1460,7 +1629,7 @@ const contextSizeOfMessages = (messages = []) => {
     const chars = perMessage.reduce((sum, item) => sum + item.chars, 0);
     return { chars, approx_tokens: approxTokenCount((messages || []).map((m) => m.content || '').join('\n')), messages: perMessage };
 };
-const wrapElefaiMemory = (memory, ctx, memoryKey) => new Proxy(memory, {
+const wrapTemboryMemory = (memory, ctx, memoryKey) => new Proxy(memory, {
     get(target, prop) {
         if (prop === 'loadMemoryVariables') {
             return async (values = {}) => {
@@ -1469,7 +1638,7 @@ const wrapElefaiMemory = (memory, ctx, memoryKey) => new Proxy(memory, {
                 ]);
                 try {
                     const response = await target.loadMemoryVariables(values);
-                    const cacheHit = Boolean(response.elefaiBrainDiagnostics && response.elefaiBrainDiagnostics.cacheHit);
+                    const cacheHit = Boolean(response.temboryDiagnostics && response.temboryDiagnostics.cacheHit);
                     const chatHistory = cacheHit
                         ? [{ cached: true, messages: Array.isArray(response[memoryKey] || response.chatHistory) ? (response[memoryKey] || response.chatHistory).length : 0 }]
                         : snapshotJson(response[memoryKey] || response.chatHistory || []);
@@ -1479,23 +1648,23 @@ const wrapElefaiMemory = (memory, ctx, memoryKey) => new Proxy(memory, {
                                     action: 'loadMemoryVariables',
                                     cached: cacheHit || undefined,
                                     chatHistory,
-                                    context: response.elefaiBrainContext,
-                                    contextText: response.elefaiBrainContextText,
-                                    summary: response.elefaiBrainSummary,
-                                    connectedModelSummary: response.elefaiBrainConnectedModelSummary,
-                                    workingMemory: response.elefaiBrainWorkingMemory,
-                                    decisionState: response.elefaiBrainDecisionState,
-                                    memoryCompression: response.elefaiBrainMemoryCompression,
-                                    profileFacts: response.elefaiBrainProfileFacts,
-                                    operationalState: response.elefaiBrainOperationalState,
-                                    actionLedger: response.elefaiBrainActionLedger,
-                                    entityTimeline: response.elefaiBrainEntityTimeline,
-                                    vectorMemories: response.elefaiBrainVectorMemories,
-                                    graph: response.elefaiBrainGraph,
-                                    recentMessages: response.elefaiBrainRecentMessages,
-                                    recentHighlights: response.elefaiBrainRecentHighlights,
-                                    toolHistory: response.elefaiBrainToolHistory,
-                                    diagnostics: response.elefaiBrainDiagnostics,
+                                    context: response.temboryContext,
+                                    contextText: response.temboryContextText,
+                                    summary: response.temborySummary,
+                                    connectedModelSummary: response.temboryConnectedModelSummary,
+                                    workingMemory: response.temboryWorkingMemory,
+                                    decisionState: response.temboryDecisionState,
+                                    memoryCompression: response.temboryMemoryCompression,
+                                    profileFacts: response.temboryProfileFacts,
+                                    operationalState: response.temboryOperationalState,
+                                    actionLedger: response.temboryActionLedger,
+                                    entityTimeline: response.temboryEntityTimeline,
+                                    vectorMemories: response.temboryVectorMemories,
+                                    graph: response.temboryGraph,
+                                    recentMessages: response.temboryRecentMessages,
+                                    recentHighlights: response.temboryRecentHighlights,
+                                    toolHistory: response.temboryToolHistory,
+                                    diagnostics: response.temboryDiagnostics,
                                 },
                             }],
                     ]);
@@ -1529,17 +1698,18 @@ const wrapElefaiMemory = (memory, ctx, memoryKey) => new Proxy(memory, {
         return target[prop];
     },
 });
-const buildContextMessages = ({ payloadFormat, query, userId, profileFacts, workingMemory, decisionState, memoryCompression, operationalState, actionLedger, entityTimeline, vectorMemories, recentMessages, toolHistory, highlights, graph, diagnostics, adv }) => {
+const buildContextMessages = ({ payloadFormat, query, userId, profileFacts, workingMemory, decisionState, memoryCompression, operationalState, actionLedger, entityTimeline, vectorMemories, recentMessages, toolHistory, highlights, graph, diagnostics, connectedModelSummary, adv }) => {
     const includeHeader = adv.includeContextHeader !== false;
     const includeSummary = adv.includeSummary !== false;
     const includeScores = adv.includeScores !== false;
     const compactForAgent = Boolean(adv.compactForAgent);
+    const compactStateSections = Boolean(adv.compactStateSections);
     const sections = [];
     if (includeHeader) {
         sections.push({
             section: 'context_header',
             title: 'Tembory context',
-            value: compactForAgent
+            value: compactForAgent || compactStateSections
                 ? 'Read-only memory. Follow next_expected_action when present. Before calling downstream tools, verify required prior tool context in tool_history or operational_state. Do not repeat tools listed in do_not_repeat_tools.'
                 : 'Use this context as read-only memory. Prefer it over guessing. Do not mention internal section names to the user. Treat next_expected_action as an instruction, not as a suggestion. If it says to call a tool now, call that tool instead of asking the user the same question again. If the user asks to continue, chooses a slot, says ok/sim, reserve, confirm, update, cancel, or performs any downstream action that depends on a prior tool result, first verify the required prior result in tool_history, recent_messages, or vector memories. If the required prior result is absent, do not call the downstream tool; ask for the missing context or call the appropriate prerequisite tool.',
         });
@@ -1549,6 +1719,9 @@ const buildContextMessages = ({ payloadFormat, query, userId, profileFacts, work
             const summary = compactVectorMemoriesForAgent(vectorMemories, toolHistory, Number(adv.summaryMaxFacts || 3));
             sections.push({ section: 'summary', title: 'Summary', value: summary.length ? summary : null, why_null: summary.length ? undefined : 'no non-tool vector memories to summarize' });
         }
+        if (connectedModelSummary && adv.includeConnectedModelSummary !== false) {
+            sections.push({ section: 'connected_model_summary', title: 'SLM summary', value: connectedModelSummary });
+        }
         sections.push({
             section: 'working_memory',
             title: 'Working memory',
@@ -1588,7 +1761,7 @@ const buildContextMessages = ({ payloadFormat, query, userId, profileFacts, work
             why_null: adv.includeMemoryCompression === false ? 'memory compression disabled' : undefined,
         });
         const audit = {
-            kind: 'elefai.brain.context.v1',
+            kind: 'tembory.context.v1',
             meta: { user_id: userId, payloadFormat, query, compact_for_agent: true },
             sections,
             diagnostics,
@@ -1609,13 +1782,18 @@ const buildContextMessages = ({ payloadFormat, query, userId, profileFacts, work
         return [{ role: 'system', content: truncate(sections.map(renderCompactSection).join('\n\n'), Number(adv.contextMaxChars || 6000)) }];
     }
     if (includeSummary) {
-        const summary = vectorMemories.slice(0, Number(adv.summaryMaxFacts || 4)).map((memory) => contextMemoryText(memory, 360)).filter(Boolean);
+        const summary = compactStateSections
+            ? compactVectorMemoriesForAgent(vectorMemories, toolHistory, Number(adv.summaryMaxFacts || 4))
+            : vectorMemories.slice(0, Number(adv.summaryMaxFacts || 4)).map((memory) => contextMemoryText(memory, 360)).filter(Boolean);
         sections.push({ section: 'summary', title: 'Summary', value: summary.length ? summary : null, why_null: summary.length ? undefined : 'no vector memories to summarize' });
     }
+    if (connectedModelSummary && adv.includeConnectedModelSummary !== false) {
+        sections.push({ section: 'connected_model_summary', title: 'SLM summary', value: connectedModelSummary });
+    }
     sections.push({
         section: 'working_memory',
         title: 'Working memory',
-        value: adv.includeWorkingMemory === false ? null : (workingMemory || {}),
+        value: adv.includeWorkingMemory === false ? null : (compactStateSections ? compactWorkingMemoryForAgent(workingMemory || {}) : (workingMemory || {})),
         why_null: adv.includeWorkingMemory === false ? 'working memory disabled' : undefined,
     });
     sections.push({
@@ -1639,31 +1817,31 @@ const buildContextMessages = ({ payloadFormat, query, userId, profileFacts, work
     sections.push({
         section: 'decision_state',
         title: 'Decision state',
-        value: adv.includeDecisionState === false ? null : (decisionState || {}),
+        value: adv.includeDecisionState === false ? null : (compactStateSections ? compactDecisionStateForAgent(decisionState || {}) : (decisionState || {})),
         why_null: adv.includeDecisionState === false ? 'decision state disabled' : undefined,
     });
     sections.push({
         section: 'operational_state',
         title: 'Operational state',
-        value: adv.includeOperationalState === false ? null : operationalState,
+        value: adv.includeOperationalState === false ? null : (compactStateSections ? compactOperationalStateForAgent(operationalState || {}) : operationalState),
         why_null: adv.includeOperationalState === false ? 'operational state disabled' : undefined,
     });
     sections.push({
         section: 'action_ledger',
         title: 'Action ledger',
-        value: adv.includeActionLedger === false ? null : (actionLedger || []),
+        value: adv.includeActionLedger === false ? null : (compactStateSections ? compactActionLedgerForAgent(actionLedger || [], adv.actionLedgerMaxItems || 6, adv.includeToolResults !== false) : (actionLedger || [])),
         why_null: adv.includeActionLedger === false ? 'action ledger disabled' : undefined,
     });
     sections.push({
         section: 'entity_timeline',
         title: 'Entity timeline',
-        value: adv.includeEntityTimeline === false ? null : (entityTimeline || []),
+        value: adv.includeEntityTimeline === false ? null : (compactStateSections ? compactEntityTimelineForAgent(entityTimeline || [], adv.entityTimelineMaxItems || 8) : (entityTimeline || [])),
         why_null: adv.includeEntityTimeline === false ? 'entity timeline disabled' : undefined,
     });
     sections.push({
         section: 'vector',
         title: 'Vector memories',
-        value: vectorMemories.map((m) => {
+        value: compactStateSections ? compactVectorMemoriesForAgent(vectorMemories, toolHistory, Number(adv.maxReturn || adv.topK || 4)) : vectorMemories.map((m) => {
             const scoreMeta = scoreMetaOf(m);
             const rawScore = scoreOf(m);
             return {
@@ -1681,7 +1859,7 @@ const buildContextMessages = ({ payloadFormat, query, userId, profileFacts, work
     sections.push({
         section: 'memory_compression',
         title: 'Memory compression',
-        value: adv.includeMemoryCompression === false ? null : (memoryCompression || {}),
+        value: adv.includeMemoryCompression === false ? null : (compactStateSections ? compactMemoryCompressionForAgent(memoryCompression || {}) : (memoryCompression || {})),
         why_null: adv.includeMemoryCompression === false ? 'memory compression disabled' : undefined,
     });
     sections.push({
@@ -1693,7 +1871,11 @@ const buildContextMessages = ({ payloadFormat, query, userId, profileFacts, work
     sections.push({
         section: 'recent_messages',
         title: 'Recent messages',
-        value: recentMessages || [],
+        value: compactStateSections ? pruneByLimit(recentMessages || [], adv.recentMessagesLastN || 4).map((message) => cleanContextValue({
+            role: message.role,
+            content: truncate(message.content, 260),
+            at: message.at,
+        })) : recentMessages || [],
     });
     sections.push({
         section: 'recent_highlights',
@@ -1704,18 +1886,18 @@ const buildContextMessages = ({ payloadFormat, query, userId, profileFacts, work
         section: 'tool_history',
         title: 'Tool history',
         value: {
-            enabled: adv.includeToolHistory !== false,
-            items: adv.includeToolHistory === false ? [] : toolHistory.map((tool) => ({
-                id: tool.id,
-                turnId: tool.turnId,
-                sequence: tool.sequence,
+                enabled: adv.includeToolHistory !== false,
+                items: adv.includeToolHistory === false ? [] : (compactStateSections ? compactToolHistoryForAgent(toolHistory, adv.toolHistoryLastN || 8, adv.includeToolResults !== false) : toolHistory.map((tool) => ({
+                    id: tool.id,
+                    turnId: tool.turnId,
+                    sequence: tool.sequence,
                 name: tool.name,
                 input: tool.input,
                 ok: tool.ok,
                 at: tool.at,
-                source: tool.source,
-                result: adv.includeToolResults === false ? undefined : tool.result,
-            })),
+                    source: tool.source,
+                    result: adv.includeToolResults === false ? undefined : tool.result,
+            }))),
         },
     });
     if (adv.includeDiagnostics) {
@@ -1726,7 +1908,7 @@ const buildContextMessages = ({ payloadFormat, query, userId, profileFacts, work
         });
     }
     const audit = {
-        kind: 'elefai.brain.context.v1',
+        kind: 'tembory.context.v1',
         meta: { user_id: userId, payloadFormat, query },
         sections,
         diagnostics,
@@ -1740,17 +1922,32 @@ const buildContextMessages = ({ payloadFormat, query, userId, profileFacts, work
             return `## ${section.title}\n${section.value.length ? section.value.map((item) => typeof item === 'string' ? `- ${item}` : `- ${truncate(safeStringify(item), 900)}`).join('\n') : '(empty)'}`;
         return `## ${section.title}\n${truncate(safeStringify(section.value), 1600)}`;
     };
+    const sectionIsEmpty = (section) => {
+        if (!compactStateSections)
+            return false;
+        if (section.section === 'context_header' || section.section === 'tool_guard' || section.section === 'working_memory' || section.section === 'decision_state' || section.section === 'operational_state' || section.section === 'tool_history')
+            return false;
+        if (section.value === null || section.value === undefined)
+            return true;
+        if (Array.isArray(section.value) && section.value.length === 0)
+            return true;
+        if (typeof section.value === 'object' && !Array.isArray(section.value) && Object.keys(section.value).length === 0)
+            return true;
+        return false;
+    };
+    const renderableSections = compactStateSections ? sections.filter((section) => !sectionIsEmpty(section)) : sections;
     if (payloadFormat === 'auditJson') {
         return [{ role: 'system', content: JSON.stringify(audit, null, 2) }];
     }
     if (payloadFormat === 'auditBlocks') {
-        return sections.map((section) => ({ role: 'system', content: renderSection(section) }));
+        return renderableSections.map((section) => ({ role: 'system', content: renderSection(section) }));
     }
-    const text = sections.map(renderSection).join('\n\n');
+    const text = renderableSections.map(renderSection).join('\n\n');
+    const maxChars = Number(adv.contextMaxChars || 16000);
     if (payloadFormat === 'auditText' || payloadFormat === 'singleSystemMessage' || payloadFormat === 'structured') {
-        return [{ role: 'system', content: truncate(text, 16000) }];
+        return [{ role: 'system', content: truncate(text, maxChars) }];
     }
-    return [{ role: 'system', content: truncate(text, 16000) }];
+    return [{ role: 'system', content: truncate(text, maxChars) }];
 };
 class Mem0Memory {
     constructor() {
@@ -1898,8 +2095,25 @@ class Mem0Memory {
                         { displayName: 'MMR Lambda', name: 'mmrLambda', type: 'number', typeOptions: { minValue: 0, maxValue: 1, numberPrecision: 2 }, default: 0.5, description: 'Equilíbrio entre relevância e diversidade no MMR', displayOptions: { show: { '/retrievalMode': ['hybrid'] } } },
                         { displayName: 'Incluir Cabeçalho de Contexto', name: 'includeContextHeader', type: 'boolean', default: true },
                         { displayName: 'Compactar Contexto Para Agente', name: 'compactForAgent', type: 'boolean', default: false, description: 'Entrega ao agente apenas o estado acionável: working memory, decisões, estado operacional e histórico de tools compacto. Mantém detalhes completos no audit/debug.' },
-                        { displayName: 'Máximo de Caracteres do Contexto', name: 'contextMaxChars', type: 'number', default: 6000, description: 'Limite final aplicado quando Compactar Contexto Para Agente estiver ativo.' },
+                        { displayName: 'Organizar Seções de Produção', name: 'compactStateSections', type: 'boolean', default: false, description: 'Renderiza blocos de produção com campos canônicos, remove seções vazias e evita duplicar evidências já presentes em tool_history/operational_state.' },
+                        { displayName: 'Máximo de Caracteres do Contexto', name: 'contextMaxChars', type: 'number', default: 6000, description: 'Limite final aplicado ao contexto entregue ao agente em modos estruturados de produção.' },
                         { displayName: 'Incluir Resumo', name: 'includeSummary', type: 'boolean', default: true },
+                        { displayName: 'Incluir Resumo do SLM', name: 'includeConnectedModelSummary', type: 'boolean', default: true, description: 'Usa o modelo barato conectado ao Tembory para organizar contexto ativo e/ou memórias vetoriais em bullets curtos.' },
+                        {
+                            displayName: 'Fonte do Resumo do SLM',
+                            name: 'summarySource',
+                            type: 'options',
+                            options: [
+                                { name: 'Automático', value: 'auto' },
+                                { name: 'Contexto Ativo', value: 'activeContext' },
+                                { name: 'Somente Vetores', value: 'vectorOnly' },
+                                { name: 'Desligado', value: 'off' },
+                            ],
+                            default: 'auto',
+                            description: 'Define se o SLM resume working memory/tool state, memórias vetoriais, ambos ou nada.',
+                        },
+                        { displayName: 'Máximo de Caracteres de Entrada do SLM', name: 'connectedModelSummaryInputMaxChars', type: 'number', default: 3000 },
+                        { displayName: 'Máximo de Caracteres do Resumo do SLM', name: 'connectedModelSummaryMaxChars', type: 'number', default: 900 },
                         { displayName: 'Máximo de Fatos no Resumo', name: 'summaryMaxFacts', type: 'number', default: 4 },
                         { displayName: 'Incluir Scores', name: 'includeScores', type: 'boolean', default: true },
                         { displayName: 'Incluir Diagnóstico', name: 'includeDiagnostics', type: 'boolean', default: false },
@@ -1968,24 +2182,24 @@ class Mem0Memory {
                 if (loadCache.has(cacheKey)) {
                     result = snapshotJson(loadCache.get(cacheKey));
                     result.response = result.response || {};
-                    result.response.elefaiBrainDiagnostics = {
-                        ...(result.response.elefaiBrainDiagnostics || {}),
+                    result.response.temboryDiagnostics = {
+                        ...(result.response.temboryDiagnostics || {}),
                         cacheHit: true,
                         cacheScope: 'supplyData.loadMemoryVariables',
                     };
-                    if (result.response.elefaiBrainContext) {
-                        result.response.elefaiBrainContext = {
-                            kind: result.response.elefaiBrainContext.kind,
-                            userId: result.response.elefaiBrainContext.userId,
-                            project: result.response.elefaiBrainContext.project,
-                            query: result.response.elefaiBrainContext.query,
-                            retrievalMode: result.response.elefaiBrainContext.retrievalMode,
-                            payloadFormat: result.response.elefaiBrainContext.payloadFormat,
-                            contextQualityScore: result.response.elefaiBrainContext.contextQualityScore,
+                    if (result.response.temboryContext) {
+                        result.response.temboryContext = {
+                            kind: result.response.temboryContext.kind,
+                            userId: result.response.temboryContext.userId,
+                            project: result.response.temboryContext.project,
+                            query: result.response.temboryContext.query,
+                            retrievalMode: result.response.temboryContext.retrievalMode,
+                            payloadFormat: result.response.temboryContext.payloadFormat,
+                            contextQualityScore: result.response.temboryContext.contextQualityScore,
                             cacheHit: true,
                         };
                     }
-                    delete result.response.elefaiBrainContextText;
+                    delete result.response.temboryContextText;
                 }
                 else {
                     result = await Mem0Memory.prototype.loadMemoryVariablesForItem.call(this, itemIndex, inputValues);
@@ -2016,7 +2230,7 @@ class Mem0Memory {
                 await Mem0Memory.prototype.saveContextForItem.call(this, itemIndex, inputValues, outputValues);
             },
         };
-        return { response: wrapElefaiMemory(memory, this, memoryKey) };
+        return { response: wrapTemboryMemory(memory, this, memoryKey) };
     }
     async saveMessagesForItem(itemIndex, messages = []) {
         const inputValues = {};
@@ -2427,7 +2641,7 @@ class Mem0Memory {
                     }
                     ranked = selected;
                 }
-                const finalMemories = ranked.slice(0, maxReturn).map((r) => withElefaiScore(r.m, {
+                const finalMemories = ranked.slice(0, maxReturn).map((r) => withTemboryScore(r.m, {
                     semanticScore: r.semanticScore,
                     recencyScore: r.recencyScore,
                     hybridScore: r.hybrid,
@@ -2437,7 +2651,7 @@ class Mem0Memory {
                 payload = finalMemories.map((m) => { var _a, _b; return ({ role: 'system', content: (_b = (_a = m.memory) !== null && _a !== void 0 ? _a : m.text) !== null && _b !== void 0 ? _b : JSON.stringify(m) }); });
             }
             else {
-                vectorMemories = semMemories.map((m) => withElefaiScore(m, {
+                vectorMemories = semMemories.map((m) => withTemboryScore(m, {
                     semanticScore: scoreOf(m),
                     source: 'semantic',
                 }));
@@ -2528,34 +2742,6 @@ class Mem0Memory {
             }
             catch { }
         }
-        let connectedModelSummary = '';
-        if (connectedLanguageModel && typeof connectedLanguageModel.invoke === 'function' && vectorMemories.length && adv.includeSummary !== false) {
-            try {
-                const facts = vectorMemories.slice(0, Number(adv.summaryMaxFacts || 4)).map((memory) => contextMemoryText(memory, 500)).filter(Boolean).join('\n') || '(no vector memories found)';
-                if (facts) {
-                    const response = await connectedLanguageModel.invoke([
-                        toBaseMessage({
-                            role: 'user',
-                            content: `Summarize available Tembory context as concise factual bullets. If there are no memories, return exactly "(empty)".\n\nQuery: ${String(query || '')}\n\nMemories:\n${truncate(facts, 3000)}`,
-                        }),
-                    ]);
-                    connectedModelSummary = truncate(response?.content || response?.text || String(response || ''), 1200);
-                    connectedAi.languageModelSummary = true;
-                }
-            }
-            catch (error) {
-                connectedAi.errors.push(`languageModel.invoke: ${error.message || String(error)}`);
-            }
-        }
-        const diagnostics = {
-            vectorMemories: vectorMemories.length,
-            recentMessages: recentMessages.length,
-            toolHistory: toolHistory.length,
-            graph: graph.length,
-            project: project || null,
-            memoryNamespace: key,
-            connectedAi,
-        };
         const includeToolResults = adv.includeToolResults !== false;
         const operationalState = deriveOperationalState(toolHistory, renderProfileFacts(profileFacts), recentMessages, includeToolResults);
         const actionLedger = deriveActionLedger(toolHistory, adv.actionLedgerMaxItems || adv.toolHistoryLastN || 20, includeToolResults);
@@ -2582,6 +2768,38 @@ class Mem0Memory {
             vectorMemories,
             maxItems: adv.compressionMaxItems || 6,
         });
+        let connectedModelSummary = '';
+        if (connectedLanguageModel && typeof connectedLanguageModel.invoke === 'function' && adv.includeSummary !== false && adv.includeConnectedModelSummary !== false) {
+            try {
+                const summaryInput = buildConnectedModelSummaryInput({
+                    query,
+                    profileFacts,
+                    workingMemory,
+                    decisionState,
+                    memoryCompression,
+                    operationalState,
+                    toolHistory,
+                    recentMessages,
+                    vectorMemories,
+                    highlights,
+                    adv,
+                });
+                connectedModelSummary = await invokeConnectedModelSummary(connectedLanguageModel, summaryInput, adv);
+                connectedAi.languageModelSummary = Boolean(connectedModelSummary);
+            }
+            catch (error) {
+                connectedAi.errors.push(`languageModel.invoke: ${error.message || String(error)}`);
+            }
+        }
+        const diagnostics = {
+            vectorMemories: vectorMemories.length,
+            recentMessages: recentMessages.length,
+            toolHistory: toolHistory.length,
+            graph: graph.length,
+            project: project || null,
+            memoryNamespace: key,
+            connectedAi,
+        };
         const contextHealth = deriveContextHealth({
             userId: key,
             project,
@@ -2619,6 +2837,7 @@ class Mem0Memory {
             highlights,
             graph,
             diagnostics,
+            connectedModelSummary,
             adv,
         });
         diagnostics.contextSize = contextSizeOfMessages(payload);
@@ -2645,7 +2864,7 @@ class Mem0Memory {
         const summary = vectorMemories.slice(0, Number(adv.summaryMaxFacts || 4)).map((memory) => contextMemoryText(memory, 360)).filter(Boolean);
         const contextText = payload.map((message) => String(message.content || '')).join('\n\n');
         const audit = {
-            kind: 'elefai.brain.context.v1',
+            kind: 'tembory.context.v1',
             userId: key,
             project: project || undefined,
             query,
@@ -2716,25 +2935,25 @@ class Mem0Memory {
         return {
             response: {
                 [memoryKey]: payload,
-                elefaiBrainContext: audit,
-                elefaiBrainContextText: contextText,
-                elefaiBrainSummary: summary,
-                elefaiBrainConnectedModelSummary: connectedModelSummary,
-                elefaiBrainContextHealth: contextHealth,
-                elefaiBrainContextQualityScore: contextHealth.quality_score,
-                elefaiBrainWorkingMemory: workingMemory,
-                elefaiBrainDecisionState: decisionState,
-                elefaiBrainMemoryCompression: memoryCompression,
-                elefaiBrainProfileFacts: renderProfileFacts(profileFacts),
-                elefaiBrainOperationalState: operationalState,
-                elefaiBrainActionLedger: actionLedger,
-                elefaiBrainEntityTimeline: entityTimeline,
-                elefaiBrainVectorMemories: normalizedVectorMemories,
-                elefaiBrainGraph: graph,
-                elefaiBrainRecentMessages: recentMessages,
-                elefaiBrainRecentHighlights: highlights,
-                elefaiBrainToolHistory: audit.toolHistory,
-                elefaiBrainDiagnostics: diagnostics,
+                temboryContext: audit,
+                temboryContextText: contextText,
+                temborySummary: summary,
+                temboryConnectedModelSummary: connectedModelSummary,
+                temboryContextHealth: contextHealth,
+                temboryContextQualityScore: contextHealth.quality_score,
+                temboryWorkingMemory: workingMemory,
+                temboryDecisionState: decisionState,
+                temboryMemoryCompression: memoryCompression,
+                temboryProfileFacts: renderProfileFacts(profileFacts),
+                temboryOperationalState: operationalState,
+                temboryActionLedger: actionLedger,
+                temboryEntityTimeline: entityTimeline,
+                temboryVectorMemories: normalizedVectorMemories,
+                temboryGraph: graph,
+                temboryRecentMessages: recentMessages,
+                temboryRecentHighlights: highlights,
+                temboryToolHistory: audit.toolHistory,
+                temboryDiagnostics: diagnostics,
             },
         };
     }
@@ -2854,4 +3073,7 @@ exports.__private = {
     compactToolResult,
     compactToolHistoryForAgent,
     compactOperationalStateForAgent,
+    buildConnectedModelSummaryInput,
+    cleanModelSummaryText,
+    invokeConnectedModelSummary,
 };

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "n8n-nodes-tembory",
-  "version": "1.0.9",
+  "version": "1.0.12",
   "description": "Tembory node for n8n AI Agents with profile, tools, timeline, graph and semantic memory",
   "license": "MIT",
   "homepage": "https://tembory.com",

package/scripts/simulate-agent-context.js

@@ -42,20 +42,20 @@ const toolHistory = [
     ok: true,
     at: '2026-05-14T05:34:34.017Z',
     source: 'used_tools_text',
-    result: '[{"args":{"tool":"agenda_pre_reservar_horario","reservation_id":"RES-ELEFAI-LAB-001","selected_from_message":"dia 13","status":"pre_reserved","next_step":"confirmar_agendamento"}}]',
+    result: '[{"args":{"tool":"agenda_pre_reservar_horario","reservation_id":"RES-TEMBORY-LAB-001","selected_from_message":"dia 13","status":"pre_reserved","next_step":"confirmar_agendamento"}}]',
   },
 ];
 
 const vectorMemories = [
   {
-    memory: '[Used tools: Tool: agenda_pre_reservar_horario, Input: , Result: [{"args":{"tool":"agenda_pre_reservar_horario","reservation_id":"RES-ELEFAI-LAB-001","selected_from_message":"dia 13","status":"pre_reserved","next_step":"confirmar_agendamento"}}]] Calling agenda_pre_reservar_horario with input: {"id":"call_bfHlltcjI1vCagRkav0fj8Ha"}\nPré-reserva feita para o dia 13.\nSe quiser, posso confirmar agora.',
+    memory: '[Used tools: Tool: agenda_pre_reservar_horario, Input: , Result: [{"args":{"tool":"agenda_pre_reservar_horario","reservation_id":"RES-TEMBORY-LAB-001","selected_from_message":"dia 13","status":"pre_reserved","next_step":"confirmar_agendamento"}}]] Calling agenda_pre_reservar_horario with input: {"id":"call_bfHlltcjI1vCagRkav0fj8Ha"}\nPré-reserva feita para o dia 13.\nSe quiser, posso confirmar agora.',
     created_at: '2026-05-14T05:34:35.373020Z',
-    elefaiScore: { semanticScore: 0.71, recencyScore: 0.99, hybridScore: 0.81, source: 'semantic' },
+    temboryScore: { semanticScore: 0.71, recencyScore: 0.99, hybridScore: 0.81, source: 'semantic' },
   },
   {
     memory: '[Used tools: Tool: agenda_consultar_disponibilidade, Input: , Result: [{"args":{"tool":"agenda_consultar_disponibilidade","available_slots":"2026-05-12T09:00,2026-05-12T14:00,2026-05-13T10:30","timezone":"America/Sao_Paulo","next_step":"escolher_um_horario"}}]] Temos estes horários vagos:\n- 12/05 às 09:00\n- 12/05 às 14:00\n- 13/05 às 10:30',
     created_at: '2026-05-14T05:34:14.717988Z',
-    elefaiScore: { semanticScore: 0.7, recencyScore: 0.99, hybridScore: 0.8, source: 'semantic' },
+    temboryScore: { semanticScore: 0.7, recencyScore: 0.99, hybridScore: 0.8, source: 'semantic' },
   },
   { memory: 'Oi! Como posso ajudar?', created_at: '2026-05-14T05:33:47.194630Z' },
   { memory: 'oi', created_at: '2026-05-14T05:33:47.194355Z' },
@@ -87,7 +87,7 @@ const common = {
   toolHistory,
   highlights: [
     'tool agenda_consultar_disponibilidade: [{"args":{"available_slots":"2026-05-12T09:00,2026-05-12T14:00,2026-05-13T10:30"}}]',
-    'tool agenda_pre_reservar_horario: [{"args":{"reservation_id":"RES-ELEFAI-LAB-001","status":"pre_reserved"}}]',
+    'tool agenda_pre_reservar_horario: [{"args":{"reservation_id":"RES-TEMBORY-LAB-001","status":"pre_reserved"}}]',
   ],
   graph: [],
   diagnostics: {},
@@ -106,7 +106,7 @@ for (const [name, adv] of modes) {
     size: core.contextSizeOfMessages(messages),
     has_next_expected_action: /next_expected_action/.test(text),
     has_do_not_repeat_tools: /do_not_repeat_tools/.test(text),
-    has_reservation_id: /RES-ELEFAI-LAB-001/.test(text),
+    has_reservation_id: /RES-TEMBORY-LAB-001/.test(text),
     has_available_slots: /available_slots/.test(text),
     has_duplicate_calling_text: /Calling agenda_pre_reservar_horario/.test(text),
   }, null, 2));