@coreidentitylabs/open-graph-memory-mcp 1.0.1 → 1.0.3

package/README.md

@@ -16,6 +16,7 @@ You chat with your AI assistant
   ↓ Calls memory_add_entities / memory_add_relations
   ↓ Entities stored in knowledge graph (JSON or Neo4j)
   ↓ Before next task, agent calls memory_search or memory_get_context
+  ↓ Before complex tasks, agent calls memory_deep_analyze for rich multi-pass context
   ↓ Relevant historical context injected into prompt
   = AI remembers your project across sessions
 ```
@@ -118,6 +119,44 @@ Add to your MCP config (e.g. `mcp_config.json`):
 }
 ```
 
+## Real-World Applications & Use Cases
+
+Open-Memory is designed to solve specific challenges in complex, distributed, and fast-moving development environments:
+
+### 1. Cross-Repository Microservices
+
+In microservice architectures, knowledge is often fragmented across multiple repositories. Open-Memory acts as a decentralized knowledge bridge, allowing AI agents to:
+
+- **Track Cross-Service Dependencies**: Remember how service A interacts with service B via specific API contracts or message schemas.
+- **Maintain Architectural Consistency**: Store system-wide design patterns and security standards that apply to all repositories.
+- **Unified Onboarding**: Help new developers (and AI agents) understand the "big picture" by querying a persistent graph of how various services fit together.
+
+### 2. Guiding AI Agents with Full Context
+
+Traditional context windows are transient. By using a persistent memory, you can "train" your AI agents over time:
+
+- **Persistent Logic & Decisions**: Store the "why" behind past architectural choices so the AI doesn't suggest reverting them in future sessions.
+- **Project-Specific Knowledge**: Maintain a record of non-obvious business logic, domain-specific terminology, and custom internal tools.
+- **Contextual Recall**: When switching between tasks, the agent can call `memory_get_context` to instantly remember the relevant parts of the system it worked on last.
+
+### 3. Rapidly Changing Codebases
+
+When code is evolving fast, documentation often lags behind. Open-Memory helps keep pace by:
+
+- **Tracking In-Flight Changes**: Store ongoing refactorings and temporary architectural shifts that haven't been finalized yet.
+- **Delta Memory**: Use `memory_encode_text` to capture technical debt, "TODOs" discussed in chat, and emerging patterns before they are formally documented.
+- **Change History**: Query how a specific module's purpose or implementation has evolved over several iterations based on previous developer-AI interactions.
+
+### 4. Deep Analysis Before High-Stakes Decisions
+
+Before architectural changes, major refactors, or complex feature work, agents can call `memory_deep_analyze` to get a rich structured report on any topic:
+
+- **Key Entities by Centrality**: Surfaces the most connected concepts in the graph relevant to your topic
+- **Cluster Detection**: Reveals hidden communities of related decisions, patterns, and code structures
+- **Temporal Reasoning**: Shows how a technology, decision, or pattern evolved over calendar quarters
+- **Contradiction Detection**: Automatically flags conflicting facts stored across different sessions
+- **Suggested Next Steps**: Actionable pointers for deeper investigation before you commit to a direction
+
 ## Tools
 
 ### Write
@@ -139,6 +178,12 @@ Add to your MCP config (e.g. `mcp_config.json`):
 | `memory_get_relations` | Get relationships for an entity            |
 | `memory_get_context`   | Get formatted context for prompt injection |
 
+### Analysis
+
+| Tool                  | Description                                                                     | LLM Required |
+| --------------------- | ------------------------------------------------------------------------------- | :----------: |
+| `memory_deep_analyze` | Multi-pass deep analysis: centrality, clusters, temporal trends, contradictions |      ❌      |
+
 ### Management
 
 | Tool                   | Description                                      |
@@ -167,6 +212,8 @@ src/
 │   └── prompts.ts            # Extraction prompts
 ├── retrieval/
 │   └── search.ts             # Hybrid search engine
+├── analysis/
+│   └── deep-analyzer.ts      # Multi-pass deep analysis engine
 ├── evolution/
 │   └── consolidator.ts       # Memory consolidation
 ├── tools/
@@ -182,6 +229,33 @@ src/
 - **Encoding**: Optional server-side LLM pipeline (OpenAI, Ollama, Azure, etc.)
 - **Transport**: stdio (standard for IDE integrations)
 
+## References & Inspiration
+
+### Core Research and Surveys
+
+- **Yang, C., et al. (2026).** _Graph-based Agent Memory: Taxonomy, Techniques, and Applications._ This comprehensive survey provides the foundational taxonomy for structured topological models of experience, covering the memory lifecycle of extraction, storage, retrieval, and evolution.
+  - [Link: arXiv:2602.05665](https://arxiv.org/abs/2602.05665)
+  - [Repository: Awesome-GraphMemory](https://github.com/Graph-based-Agent-Memory)
+- **Yusuke, S. (2026).** _Graph-Based Agent Memory: A Complete Guide to Structure, Retrieval, and Evolution._ A detailed synthesis of design patterns for AI agents to solve context window limitations using graph structures.
+  - [Link: Medium Article](https://medium.com/@yusuke_s/graph-based-agent-memory-a-complete-guide-to-structure-retrieval-and-evolution-6824b22fb75e)
+
+### Architecture and Implementation Guides
+
+- **Lyon, W. (GraphGeeks).** _Building Intelligent Memory: Graph Databases for AI Agent Context and Retrieval._ A practical implementation guide focusing on Neo4j, Dgraph, and the Model Context Protocol (MCP) to perform context engineering and solve "AI amnesia".
+  - [Link: YouTube Video](https://www.youtube.com/watch?v=lyon_w_graphgeeks)
+  - [Project Demo: Go Fetch (Graph Oriented Fast Edge Traversal)](https://github.com/graph-geeks/go-fetch)
+- **MAGMA Architecture.** _Memory-Augmented Graph-based Multi-Agent Architecture._ This source details the four-layer graph model (Semantic, Temporal, Causal, and Entity) used for long-horizon task reasoning and dual-stream memory evolution.
+  - [Link: arXiv:2601.03236](https://arxiv.org/abs/2601.03236)
+
+### Key Protocols
+
+- **Model Context Protocol (MCP):** The standardized protocol used in this repository to expose graph memory search and storage tools to agents in VS Code and Google Antigravity.
+  - [Link: Model Context Protocol Official Site](https://modelcontextprotocol.io)
+
+---
+
+**Implementation Note:** This project, `open-graph-memory-mcp`, is an implementation of the Model Context Protocol (MCP) specifically designed to realize the Temporal and Knowledge Memory structures proposed in the research cited above.
+
 ## License
 
 MIT

package/dist/analysis/deep-analyzer.d.ts

@@ -0,0 +1,66 @@
+import type { StorageBackend } from "../types.js";
+export interface DeepAnalysisOptions {
+    topic: string;
+    /** Graph traversal depth from each seed node (default: 3) */
+    maxDepth?: number;
+    /** Number of seed nodes from initial hybrid search (default: 15) */
+    topK?: number;
+    /** Whether to include temporal trend analysis (default: true) */
+    includeTemporalAnalysis?: boolean;
+    /** Whether to detect topic clusters (default: true) */
+    includeClusters?: boolean;
+    /** Hard cap on nodes explored to bound performance (default: 100) */
+    maxNodes?: number;
+}
+export interface ClusterSummary {
+    /** Most connected node in this cluster */
+    centroid: string;
+    members: string[];
+    sharedRelations: string[];
+}
+export interface TemporalInsight {
+    /** Calendar quarter, e.g. "2025-Q1" */
+    period: string;
+    activeEntities: string[];
+    decisions: string[];
+    trend: "growing" | "stable" | "declining";
+}
+export interface DeepAnalysisReport {
+    topic: string;
+    analyzedAt: string;
+    seedNodes: {
+        name: string;
+        type: string;
+        score: number;
+    }[];
+    totalNodesExplored: number;
+    totalEdgesTraversed: number;
+    keyEntities: {
+        name: string;
+        type: string;
+        centrality: number;
+        description: string;
+    }[];
+    criticalPaths: {
+        path: string[];
+        relations: string[];
+        significance: string;
+    }[];
+    clusters: ClusterSummary[];
+    temporalInsights: TemporalInsight[];
+    contradictions: {
+        entity: string;
+        conflictDescription: string;
+    }[];
+    synthesizedSummary: string;
+    suggestedNextSteps: string[];
+    /** 0–1: based on seed coverage and graph density around the topic */
+    confidence: number;
+}
+/**
+ * Multi-pass deep analysis of a topic across the memory graph.
+ * Returns a structured DeepAnalysisReport with key entities, clusters,
+ * temporal trends, contradictions, and actionable next steps.
+ */
+export declare function deepAnalyze(store: StorageBackend, options: DeepAnalysisOptions): Promise<DeepAnalysisReport>;
+//# sourceMappingURL=deep-analyzer.d.ts.map

package/dist/analysis/deep-analyzer.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"deep-analyzer.d.ts","sourceRoot":"","sources":["../../src/analysis/deep-analyzer.ts"],"names":[],"mappings":"AAYA,OAAO,KAAK,EAAE,cAAc,EAA0B,MAAM,aAAa,CAAC;AAW1E,MAAM,WAAW,mBAAmB;IAClC,KAAK,EAAE,MAAM,CAAC;IACd,6DAA6D;IAC7D,QAAQ,CAAC,EAAE,MAAM,CAAC;IAClB,oEAAoE;IACpE,IAAI,CAAC,EAAE,MAAM,CAAC;IACd,iEAAiE;IACjE,uBAAuB,CAAC,EAAE,OAAO,CAAC;IAClC,uDAAuD;IACvD,eAAe,CAAC,EAAE,OAAO,CAAC;IAC1B,qEAAqE;IACrE,QAAQ,CAAC,EAAE,MAAM,CAAC;CACnB;AAED,MAAM,WAAW,cAAc;IAC7B,0CAA0C;IAC1C,QAAQ,EAAE,MAAM,CAAC;IACjB,OAAO,EAAE,MAAM,EAAE,CAAC;IAClB,eAAe,EAAE,MAAM,EAAE,CAAC;CAC3B;AAED,MAAM,WAAW,eAAe;IAC9B,uCAAuC;IACvC,MAAM,EAAE,MAAM,CAAC;IACf,cAAc,EAAE,MAAM,EAAE,CAAC;IACzB,SAAS,EAAE,MAAM,EAAE,CAAC;IACpB,KAAK,EAAE,SAAS,GAAG,QAAQ,GAAG,WAAW,CAAC;CAC3C;AAED,MAAM,WAAW,kBAAkB;IACjC,KAAK,EAAE,MAAM,CAAC;IACd,UAAU,EAAE,MAAM,CAAC;IACnB,SAAS,EAAE;QAAE,IAAI,EAAE,MAAM,CAAC;QAAC,IAAI,EAAE,MAAM,CAAC;QAAC,KAAK,EAAE,MAAM,CAAA;KAAE,EAAE,CAAC;IAC3D,kBAAkB,EAAE,MAAM,CAAC;IAC3B,mBAAmB,EAAE,MAAM,CAAC;IAC5B,WAAW,EAAE;QACX,IAAI,EAAE,MAAM,CAAC;QACb,IAAI,EAAE,MAAM,CAAC;QACb,UAAU,EAAE,MAAM,CAAC;QACnB,WAAW,EAAE,MAAM,CAAC;KACrB,EAAE,CAAC;IACJ,aAAa,EAAE;QACb,IAAI,EAAE,MAAM,EAAE,CAAC;QACf,SAAS,EAAE,MAAM,EAAE,CAAC;QACpB,YAAY,EAAE,MAAM,CAAC;KACtB,EAAE,CAAC;IACJ,QAAQ,EAAE,cAAc,EAAE,CAAC;IAC3B,gBAAgB,EAAE,eAAe,EAAE,CAAC;IACpC,cAAc,EAAE;QAAE,MAAM,EAAE,MAAM,CAAC;QAAC,mBAAmB,EAAE,MAAM,CAAA;KAAE,EAAE,CAAC;IAClE,kBAAkB,EAAE,MAAM,CAAC;IAC3B,kBAAkB,EAAE,MAAM,EAAE,CAAC;IAC7B,qEAAqE;IACrE,UAAU,EAAE,MAAM,CAAC;CACpB;AAMD;;;;GAIG;AACH,wBAAsB,WAAW,CAC/B,KAAK,EAAE,cAAc,EACrB,OAAO,EAAE,mBAAmB,GAC3B,OAAO,CAAC,kBAAkB,CAAC,CA8O7B"}

package/dist/analysis/deep-analyzer.js

@@ -0,0 +1,247 @@
+// =============================================================================
+// Deep Analysis Engine — Multi-Pass Graph Analysis for Topic Context
+// =============================================================================
+// Performs 5 analysis passes to produce a structured DeepAnalysisReport:
+//   Pass 1: Seed Search     — hybrid search to find top-K anchor nodes
+//   Pass 2: BFS Expansion   — full subgraph traversal to configurable depth
+//   Pass 3: Centrality      — degree centrality + cluster detection
+//   Pass 4: Temporal        — timeline of decisions/events by quarter
+//   Pass 5: Contradictions  — conflicting facts via cosine similarity
+//   Synthesis               — key entities, critical paths, summary, next steps
+// =============================================================================
+import { hybridSearch } from "../retrieval/search.js";
+import { generateLocalEmbedding, cosineSimilarity, } from "../encoding/embedder.js";
+// =============================================================================
+// Main Entry Point
+// =============================================================================
+/**
+ * Multi-pass deep analysis of a topic across the memory graph.
+ * Returns a structured DeepAnalysisReport with key entities, clusters,
+ * temporal trends, contradictions, and actionable next steps.
+ */
+export async function deepAnalyze(store, options) {
+    const { topic, maxDepth = 3, topK = 15, includeTemporalAnalysis = true, includeClusters = true, maxNodes = 100, } = options;
+    const analyzedAt = new Date().toISOString();
+    // ── Pass 1: Seed Search ───────────────────────────────────────────────────
+    const seedResults = await hybridSearch(store, { query: topic, topK });
+    const seedNodes = seedResults.nodes;
+    const visitedIds = new Set(seedNodes.map((n) => n.node.id));
+    const allNodes = new Map(seedNodes.map((n) => [n.node.id, n.node]));
+    const allEdges = new Map();
+    // ── Pass 2: BFS Subgraph Expansion ───────────────────────────────────────
+    let frontier = seedNodes.map((n) => n.node.id);
+    for (let depth = 0; depth < maxDepth && allNodes.size < maxNodes; depth++) {
+        const nextFrontier = [];
+        for (const nodeId of frontier) {
+            const subgraph = await store.getNeighborhood(nodeId, 1);
+            for (const edge of subgraph.edges) {
+                allEdges.set(edge.id, edge);
+            }
+            for (const neighbor of subgraph.nodes) {
+                if (!visitedIds.has(neighbor.id) && allNodes.size < maxNodes) {
+                    visitedIds.add(neighbor.id);
+                    allNodes.set(neighbor.id, neighbor);
+                    nextFrontier.push(neighbor.id);
+                }
+            }
+        }
+        frontier = nextFrontier;
+        if (frontier.length === 0)
+            break;
+    }
+    // ── Pass 3: Centrality & Cluster Analysis ─────────────────────────────────
+    // Degree centrality: count edge connections per node
+    const degreeCentrality = new Map();
+    for (const edge of allEdges.values()) {
+        degreeCentrality.set(edge.source, (degreeCentrality.get(edge.source) ?? 0) + 1);
+        degreeCentrality.set(edge.target, (degreeCentrality.get(edge.target) ?? 0) + 1);
+    }
+    const edgeCount = Math.max(allEdges.size, 1);
+    const keyEntities = [...allNodes.values()]
+        .map((n) => ({
+        name: n.name,
+        type: n.type,
+        centrality: (degreeCentrality.get(n.id) ?? 0) / edgeCount,
+        description: n.description.substring(0, 300),
+    }))
+        .sort((a, b) => b.centrality - a.centrality)
+        .slice(0, 10);
+    // Cluster detection: community members around hub nodes
+    const clusters = [];
+    if (includeClusters) {
+        for (const hub of keyEntities.slice(0, 3)) {
+            const hubNode = [...allNodes.values()].find((n) => n.name === hub.name);
+            if (!hubNode)
+                continue;
+            const hubEdges = [...allEdges.values()].filter((e) => e.source === hubNode.id || e.target === hubNode.id);
+            const memberIds = new Set(hubEdges.flatMap((e) => [e.source, e.target]));
+            memberIds.delete(hubNode.id);
+            const members = [...memberIds]
+                .map((id) => allNodes.get(id)?.name)
+                .filter(Boolean);
+            const sharedRelations = [...new Set(hubEdges.map((e) => e.relation))];
+            if (members.length > 0) {
+                clusters.push({
+                    centroid: hub.name,
+                    members: members.slice(0, 8),
+                    sharedRelations,
+                });
+            }
+        }
+    }
+    // ── Pass 4: Temporal Analysis ────────────────────────────────────────────
+    const temporalInsights = [];
+    if (includeTemporalAnalysis) {
+        const byQuarter = new Map();
+        for (const node of allNodes.values()) {
+            const d = new Date(node.createdAt);
+            const key = `${d.getFullYear()}-Q${Math.ceil((d.getMonth() + 1) / 3)}`;
+            if (!byQuarter.has(key))
+                byQuarter.set(key, []);
+            byQuarter.get(key).push(node);
+        }
+        const sortedPeriods = [...byQuarter.keys()].sort();
+        for (const period of sortedPeriods) {
+            const nodes = byQuarter.get(period);
+            const decisions = nodes
+                .filter((n) => n.type === "decision")
+                .map((n) => n.name);
+            const activeEntities = nodes
+                .filter((n) => n.type !== "conversation")
+                .map((n) => n.name)
+                .slice(0, 6);
+            const prevIdx = sortedPeriods.indexOf(period) - 1;
+            const prevCount = prevIdx >= 0 ? (byQuarter.get(sortedPeriods[prevIdx])?.length ?? 0) : 0;
+            const trend = nodes.length > prevCount
+                ? "growing"
+                : nodes.length < prevCount
+                    ? "declining"
+                    : "stable";
+            temporalInsights.push({ period, activeEntities, decisions, trend });
+        }
+    }
+    // ── Pass 5: Contradiction Detection ──────────────────────────────────────
+    // Heuristic: same-name nodes with semantically divergent descriptions
+    const contradictions = [];
+    const nameGroups = new Map();
+    for (const node of allNodes.values()) {
+        const key = node.name.toLowerCase();
+        if (!nameGroups.has(key))
+            nameGroups.set(key, []);
+        nameGroups.get(key).push(node);
+    }
+    for (const nodes of nameGroups.values()) {
+        if (nodes.length < 2)
+            continue;
+        const embedding0 = generateLocalEmbedding(nodes[0].description);
+        for (let i = 1; i < nodes.length; i++) {
+            const embeddingI = generateLocalEmbedding(nodes[i].description);
+            const sim = cosineSimilarity(embedding0, embeddingI);
+            if (sim < 0.5) {
+                contradictions.push({
+                    entity: nodes[0].name,
+                    conflictDescription: `Multiple descriptions with low similarity (${sim.toFixed(2)}): ` +
+                        `"${nodes[0].description.substring(0, 100)}" vs ` +
+                        `"${nodes[i].description.substring(0, 100)}"`,
+                });
+            }
+        }
+    }
+    // ── Critical Path Discovery ──────────────────────────────────────────────
+    const criticalPaths = discoverCriticalPaths(allEdges, allNodes, keyEntities.slice(0, 5));
+    // ── Synthesis ────────────────────────────────────────────────────────────
+    const topEntityNames = keyEntities
+        .slice(0, 5)
+        .map((e) => e.name)
+        .join(", ");
+    const synthesizedSummary = [
+        `Deep analysis of "${topic}" explored ${allNodes.size} nodes and ${allEdges.size} relationships.`,
+        keyEntities.length > 0
+            ? `Key entities by centrality: ${topEntityNames}.`
+            : "",
+        clusters.length > 0
+            ? `Identified ${clusters.length} cluster(s) around: ${clusters.map((c) => c.centroid).join(", ")}.`
+            : "",
+        contradictions.length > 0
+            ? `⚠️ ${contradictions.length} potential contradiction(s) detected.`
+            : "",
+        temporalInsights.length > 0
+            ? `Temporal span: ${temporalInsights[0].period} → ${temporalInsights.at(-1).period}.`
+            : "",
+    ]
+        .filter(Boolean)
+        .join(" ");
+    const suggestedNextSteps = buildSuggestedNextSteps(keyEntities, contradictions, clusters);
+    // Confidence = seed coverage ratio + edge density factor (0–1)
+    const edgeDensity = allEdges.size / Math.max(allNodes.size, 1);
+    const confidence = Math.min(1, (seedNodes.length / topK) * 0.5 + Math.min(edgeDensity / 5, 0.5));
+    // Update access counts for all explored nodes
+    for (const [id, node] of allNodes.entries()) {
+        await store.updateNode(id, {
+            accessCount: node.accessCount + 1,
+            lastAccessedAt: analyzedAt,
+        });
+    }
+    return {
+        topic,
+        analyzedAt,
+        seedNodes: seedNodes.map((n) => ({
+            name: n.node.name,
+            type: n.node.type,
+            score: n.score,
+        })),
+        totalNodesExplored: allNodes.size,
+        totalEdgesTraversed: allEdges.size,
+        keyEntities,
+        criticalPaths,
+        clusters,
+        temporalInsights,
+        contradictions,
+        synthesizedSummary,
+        suggestedNextSteps,
+        confidence,
+    };
+}
+// =============================================================================
+// Private Helpers
+// =============================================================================
+function discoverCriticalPaths(allEdges, allNodes, topEntities) {
+    const paths = [];
+    for (let i = 0; i < topEntities.length - 1; i++) {
+        const fromName = topEntities[i].name;
+        const toName = topEntities[i + 1].name;
+        const fromNode = [...allNodes.values()].find((n) => n.name === fromName);
+        const toNode = [...allNodes.values()].find((n) => n.name === toName);
+        if (!fromNode || !toNode)
+            continue;
+        const directEdge = [...allEdges.values()].find((e) => (e.source === fromNode.id && e.target === toNode.id) ||
+            (e.source === toNode.id && e.target === fromNode.id));
+        if (directEdge) {
+            paths.push({
+                path: [fromName, toName],
+                relations: [directEdge.relation],
+                significance: "Direct link between two high-centrality entities",
+            });
+        }
+    }
+    return paths.slice(0, 5);
+}
+function buildSuggestedNextSteps(keyEntities, contradictions, clusters) {
+    const steps = [];
+    if (contradictions.length > 0) {
+        steps.push(`Resolve contradictions for: ${contradictions.map((c) => c.entity).join(", ")}`);
+    }
+    const decisions = keyEntities.filter((e) => e.type === "decision");
+    if (decisions.length > 0) {
+        steps.push(`Review key decisions: ${decisions.map((e) => e.name).join(", ")}`);
+    }
+    if (clusters.length > 1) {
+        steps.push(`Investigate cluster relationships between "${clusters[0].centroid}" and "${clusters[1].centroid}"`);
+    }
+    steps.push(`Run memory_search with refined queries based on: ${keyEntities
+        .slice(0, 3)
+        .map((e) => e.name)
+        .join(", ")}`);
+    return steps.slice(0, 5);
+}
+//# sourceMappingURL=deep-analyzer.js.map

package/dist/analysis/deep-analyzer.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"deep-analyzer.js","sourceRoot":"","sources":["../../src/analysis/deep-analyzer.ts"],"names":[],"mappings":"AAAA,gFAAgF;AAChF,qEAAqE;AACrE,gFAAgF;AAChF,yEAAyE;AACzE,uEAAuE;AACvE,4EAA4E;AAC5E,oEAAoE;AACpE,sEAAsE;AACtE,sEAAsE;AACtE,gFAAgF;AAChF,gFAAgF;AAGhF,OAAO,EAAE,YAAY,EAAE,MAAM,wBAAwB,CAAC;AACtD,OAAO,EACL,sBAAsB,EACtB,gBAAgB,GACjB,MAAM,yBAAyB,CAAC;AA6DjC,gFAAgF;AAChF,mBAAmB;AACnB,gFAAgF;AAEhF;;;;GAIG;AACH,MAAM,CAAC,KAAK,UAAU,WAAW,CAC/B,KAAqB,EACrB,OAA4B;IAE5B,MAAM,EACJ,KAAK,EACL,QAAQ,GAAG,CAAC,EACZ,IAAI,GAAG,EAAE,EACT,uBAAuB,GAAG,IAAI,EAC9B,eAAe,GAAG,IAAI,EACtB,QAAQ,GAAG,GAAG,GACf,GAAG,OAAO,CAAC;IAEZ,MAAM,UAAU,GAAG,IAAI,IAAI,EAAE,CAAC,WAAW,EAAE,CAAC;IAE5C,6EAA6E;IAC7E,MAAM,WAAW,GAAG,MAAM,YAAY,CAAC,KAAK,EAAE,EAAE,KAAK,EAAE,KAAK,EAAE,IAAI,EAAE,CAAC,CAAC;IACtE,MAAM,SAAS,GAAG,WAAW,CAAC,KAAK,CAAC;IAEpC,MAAM,UAAU,GAAG,IAAI,GAAG,CAAS,SAAS,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,IAAI,CAAC,EAAE,CAAC,CAAC,CAAC;IACpE,MAAM,QAAQ,GAAG,IAAI,GAAG,CACtB,SAAS,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,EAAE,EAAE,CAAC,CAAC,IAAI,CAAC,CAAC,CAC1C,CAAC;IACF,MAAM,QAAQ,GAAG,IAAI,GAAG,EAAsB,CAAC;IAE/C,4EAA4E;IAC5E,IAAI,QAAQ,GAAG,SAAS,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,IAAI,CAAC,EAAE,CAAC,CAAC;IAE/C,KAAK,IAAI,KAAK,GAAG,CAAC,EAAE,KAAK,GAAG,QAAQ,IAAI,QAAQ,CAAC,IAAI,GAAG,QAAQ,EAAE,KAAK,EAAE,EAAE,CAAC;QAC1E,MAAM,YAAY,GAAa,EAAE,CAAC;QAElC,KAAK,MAAM,MAAM,IAAI,QAAQ,EAAE,CAAC;YAC9B,MAAM,QAAQ,GAAG,MAAM,KAAK,CAAC,eAAe,CAAC,MAAM,EAAE,CAAC,CAAC,CAAC;YAExD,KAAK,MAAM,IAAI,IAAI,QAAQ,CAAC,KAAK,EAAE,CAAC;gBAClC,QAAQ,CAAC,GAAG,CAAC,IAAI,CAAC,EAAE,EAAE,IAAI,CAAC,CAAC;YAC9B,CAAC;YAED,KAAK,MAAM,QAAQ,IAAI,QAAQ,CAAC,KAAK,EAAE,CAAC;gBACtC,IAAI,CAAC,UAAU,CAAC,GAAG,CAAC,QAAQ,CAAC,EAAE,CAAC,IAAI,QAAQ,CAAC,IAAI,GAAG,QAAQ,EAAE,CAAC;oBAC7D,UAAU,CAAC,GAAG,CAAC,QAAQ,CAAC,EAAE,CAAC,CAAC;oBAC5B,QAAQ,CAAC,GAAG,CAAC,QAAQ,CAAC,EAAE,EAAE,QAAQ,CAAC,CAAC;oBACpC,YAAY,CAAC,IAAI,CAAC,QAAQ,CAAC,EAAE,CAAC,CAAC;gBACjC,CAAC;YACH,CAAC;QACH,CAAC;QAED,QAAQ,GAAG,YAAY,CAAC;QACxB,IAAI,QAAQ,CAAC,MAAM,KAAK,CAAC;YAAE,MAAM;IACnC,CAAC;IAED,6EAA6E;IAC7E,qDAAqD;IACrD,MAAM,gBAAgB,GAAG,IAAI,GAAG,EAAkB,CAAC;IACnD,KAAK,MAAM,IAAI,IAAI,QAAQ,CAAC,MAAM,EAAE,EAAE,CAAC;QACrC,gBAAgB,CAAC,GAAG,CAClB,IAAI,CAAC,MAAM,EACX,CAAC,gBAAgB,CAAC,GAAG,CAAC,IAAI,CAAC,MAAM,CAAC,IAAI,CAAC,CAAC,GAAG,CAAC,CAC7C,CAAC;QACF,gBAAgB,CAAC,GAAG,CAClB,IAAI,CAAC,MAAM,EACX,CAAC,gBAAgB,CAAC,GAAG,CAAC,IAAI,CAAC,MAAM,CAAC,IAAI,CAAC,CAAC,GAAG,CAAC,CAC7C,CAAC;IACJ,CAAC;IAED,MAAM,SAAS,GAAG,IAAI,CAAC,GAAG,CAAC,QAAQ,CAAC,IAAI,EAAE,CAAC,CAAC,CAAC;IAC7C,MAAM,WAAW,GAAG,CAAC,GAAG,QAAQ,CAAC,MAAM,EAAE,CAAC;SACvC,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC;QACX,IAAI,EAAE,CAAC,CAAC,IAAI;QACZ,IAAI,EAAE,CAAC,CAAC,IAAI;QACZ,UAAU,EAAE,CAAC,gBAAgB,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,CAAC,IAAI,CAAC,CAAC,GAAG,SAAS;QACzD,WAAW,EAAE,CAAC,CAAC,WAAW,CAAC,SAAS,CAAC,CAAC,EAAE,GAAG,CAAC;KAC7C,CAAC,CAAC;SACF,IAAI,CAAC,CAAC,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,UAAU,GAAG,CAAC,CAAC,UAAU,CAAC;SAC3C,KAAK,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC;IAEhB,wDAAwD;IACxD,MAAM,QAAQ,GAAqB,EAAE,CAAC;IACtC,IAAI,eAAe,EAAE,CAAC;QACpB,KAAK,MAAM,GAAG,IAAI,WAAW,CAAC,KAAK,CAAC,CAAC,EAAE,CAAC,CAAC,EAAE,CAAC;YAC1C,MAAM,OAAO,GAAG,CAAC,GAAG,QAAQ,CAAC,MAAM,EAAE,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,IAAI,KAAK,GAAG,CAAC,IAAI,CAAC,CAAC;YACxE,IAAI,CAAC,OAAO;gBAAE,SAAS;YAEvB,MAAM,QAAQ,GAAG,CAAC,GAAG,QAAQ,CAAC,MAAM,EAAE,CAAC,CAAC,MAAM,CAC5C,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,MAAM,KAAK,OAAO,CAAC,EAAE,IAAI,CAAC,CAAC,MAAM,KAAK,OAAO,CAAC,EAAE,CAC1D,CAAC;YACF,MAAM,SAAS,GAAG,IAAI,GAAG,CAAC,QAAQ,CAAC,OAAO,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,CAAC,MAAM,EAAE,CAAC,CAAC,MAAM,CAAC,CAAC,CAAC,CAAC;YACzE,SAAS,CAAC,MAAM,CAAC,OAAO,CAAC,EAAE,CAAC,CAAC;YAE7B,MAAM,OAAO,GAAG,CAAC,GAAG,SAAS,CAAC;iBAC3B,GAAG,CAAC,CAAC,EAAE,EAAE,EAAE,CAAC,QAAQ,CAAC,GAAG,CAAC,EAAE,CAAC,EAAE,IAAI,CAAC;iBACnC,MAAM,CAAC,OAAO,CAAa,CAAC;YAE/B,MAAM,eAAe,GAAG,CAAC,GAAG,IAAI,GAAG,CAAC,QAAQ,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,QAAQ,CAAC,CAAC,CAAC,CAAC;YAEtE,IAAI,OAAO,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;gBACvB,QAAQ,CAAC,IAAI,CAAC;oBACZ,QAAQ,EAAE,GAAG,CAAC,IAAI;oBAClB,OAAO,EAAE,OAAO,CAAC,KAAK,CAAC,CAAC,EAAE,CAAC,CAAC;oBAC5B,eAAe;iBAChB,CAAC,CAAC;YACL,CAAC;QACH,CAAC;IACH,CAAC;IAED,4EAA4E;IAC5E,MAAM,gBAAgB,GAAsB,EAAE,CAAC;IAC/C,IAAI,uBAAuB,EAAE,CAAC;QAC5B,MAAM,SAAS,GAAG,IAAI,GAAG,EAAwB,CAAC;QAClD,KAAK,MAAM,IAAI,IAAI,QAAQ,CAAC,MAAM,EAAE,EAAE,CAAC;YACrC,MAAM,CAAC,GAAG,IAAI,IAAI,CAAC,IAAI,CAAC,SAAS,CAAC,CAAC;YACnC,MAAM,GAAG,GAAG,GAAG,CAAC,CAAC,WAAW,EAAE,KAAK,IAAI,CAAC,IAAI,CAAC,CAAC,CAAC,CAAC,QAAQ,EAAE,GAAG,CAAC,CAAC,GAAG,CAAC,CAAC,EAAE,CAAC;YACvE,IAAI,CAAC,SAAS,CAAC,GAAG,CAAC,GAAG,CAAC;gBAAE,SAAS,CAAC,GAAG,CAAC,GAAG,EAAE,EAAE,CAAC,CAAC;YAChD,SAAS,CAAC,GAAG,CAAC,GAAG,CAAE,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;QACjC,CAAC;QAED,MAAM,aAAa,GAAG,CAAC,GAAG,SAAS,CAAC,IAAI,EAAE,CAAC,CAAC,IAAI,EAAE,CAAC;QACnD,KAAK,MAAM,MAAM,IAAI,aAAa,EAAE,CAAC;YACnC,MAAM,KAAK,GAAG,SAAS,CAAC,GAAG,CAAC,MAAM,CAAE,CAAC;YACrC,MAAM,SAAS,GAAG,KAAK;iBACpB,MAAM,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,IAAI,KAAK,UAAU,CAAC;iBACpC,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC;YACtB,MAAM,cAAc,GAAG,KAAK;iBACzB,MAAM,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,IAAI,KAAK,cAAc,CAAC;iBACxC,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,IAAI,CAAC;iBAClB,KAAK,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC;YAEf,MAAM,OAAO,GAAG,aAAa,CAAC,OAAO,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC;YAClD,MAAM,SAAS,GACb,OAAO,IAAI,CAAC,CAAC,CAAC,CAAC,CAAC,SAAS,CAAC,GAAG,CAAC,aAAa,CAAC,OAAO,CAAC,CAAC,EAAE,MAAM,IAAI,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC;YAE1E,MAAM,KAAK,GACT,KAAK,CAAC,MAAM,GAAG,SAAS;gBACtB,CAAC,CAAC,SAAS;gBACX,CAAC,CAAC,KAAK,CAAC,MAAM,GAAG,SAAS;oBACxB,CAAC,CAAC,WAAW;oBACb,CAAC,CAAC,QAAQ,CAAC;YAEjB,gBAAgB,CAAC,IAAI,CAAC,EAAE,MAAM,EAAE,cAAc,EAAE,SAAS,EAAE,KAAK,EAAE,CAAC,CAAC;QACtE,CAAC;IACH,CAAC;IAED,4EAA4E;IAC5E,sEAAsE;IACtE,MAAM,cAAc,GAAsD,EAAE,CAAC;IAC7E,MAAM,UAAU,GAAG,IAAI,GAAG,EAAwB,CAAC;IACnD,KAAK,MAAM,IAAI,IAAI,QAAQ,CAAC,MAAM,EAAE,EAAE,CAAC;QACrC,MAAM,GAAG,GAAG,IAAI,CAAC,IAAI,CAAC,WAAW,EAAE,CAAC;QACpC,IAAI,CAAC,UAAU,CAAC,GAAG,CAAC,GAAG,CAAC;YAAE,UAAU,CAAC,GAAG,CAAC,GAAG,EAAE,EAAE,CAAC,CAAC;QAClD,UAAU,CAAC,GAAG,CAAC,GAAG,CAAE,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;IAClC,CAAC;IAED,KAAK,MAAM,KAAK,IAAI,UAAU,CAAC,MAAM,EAAE,EAAE,CAAC;QACxC,IAAI,KAAK,CAAC,MAAM,GAAG,CAAC;YAAE,SAAS;QAC/B,MAAM,UAAU,GAAG,sBAAsB,CAAC,KAAK,CAAC,CAAC,CAAC,CAAC,WAAW,CAAC,CAAC;QAChE,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,KAAK,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE,CAAC;YACtC,MAAM,UAAU,GAAG,sBAAsB,CAAC,KAAK,CAAC,CAAC,CAAC,CAAC,WAAW,CAAC,CAAC;YAChE,MAAM,GAAG,GAAG,gBAAgB,CAAC,UAAU,EAAE,UAAU,CAAC,CAAC;YACrD,IAAI,GAAG,GAAG,GAAG,EAAE,CAAC;gBACd,cAAc,CAAC,IAAI,CAAC;oBAClB,MAAM,EAAE,KAAK,CAAC,CAAC,CAAC,CAAC,IAAI;oBACrB,mBAAmB,EACjB,8CAA8C,GAAG,CAAC,OAAO,CAAC,CAAC,CAAC,KAAK;wBACjE,IAAI,KAAK,CAAC,CAAC,CAAC,CAAC,WAAW,CAAC,SAAS,CAAC,CAAC,EAAE,GAAG,CAAC,OAAO;wBACjD,IAAI,KAAK,CAAC,CAAC,CAAC,CAAC,WAAW,CAAC,SAAS,CAAC,CAAC,EAAE,GAAG,CAAC,GAAG;iBAChD,CAAC,CAAC;YACL,CAAC;QACH,CAAC;IACH,CAAC;IAED,4EAA4E;IAC5E,MAAM,aAAa,GAAG,qBAAqB,CACzC,QAAQ,EACR,QAAQ,EACR,WAAW,CAAC,KAAK,CAAC,CAAC,EAAE,CAAC,CAAC,CACxB,CAAC;IAEF,4EAA4E;IAC5E,MAAM,cAAc,GAAG,WAAW;SAC/B,KAAK,CAAC,CAAC,EAAE,CAAC,CAAC;SACX,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,IAAI,CAAC;SAClB,IAAI,CAAC,IAAI,CAAC,CAAC;IAEd,MAAM,kBAAkB,GAAG;QACzB,qBAAqB,KAAK,cAAc,QAAQ,CAAC,IAAI,cAAc,QAAQ,CAAC,IAAI,iBAAiB;QACjG,WAAW,CAAC,MAAM,GAAG,CAAC;YACpB,CAAC,CAAC,+BAA+B,cAAc,GAAG;YAClD,CAAC,CAAC,EAAE;QACN,QAAQ,CAAC,MAAM,GAAG,CAAC;YACjB,CAAC,CAAC,cAAc,QAAQ,CAAC,MAAM,uBAAuB,QAAQ,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,QAAQ,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC,GAAG;YACnG,CAAC,CAAC,EAAE;QACN,cAAc,CAAC,MAAM,GAAG,CAAC;YACvB,CAAC,CAAC,MAAM,cAAc,CAAC,MAAM,uCAAuC;YACpE,CAAC,CAAC,EAAE;QACN,gBAAgB,CAAC,MAAM,GAAG,CAAC;YACzB,CAAC,CAAC,kBAAkB,gBAAgB,CAAC,CAAC,CAAC,CAAC,MAAM,MAAM,gBAAgB,CAAC,EAAE,CAAC,CAAC,CAAC,CAAE,CAAC,MAAM,GAAG;YACtF,CAAC,CAAC,EAAE;KACP;SACE,MAAM,CAAC,OAAO,CAAC;SACf,IAAI,CAAC,GAAG,CAAC,CAAC;IAEb,MAAM,kBAAkB,GAAG,uBAAuB,CAChD,WAAW,EACX,cAAc,EACd,QAAQ,CACT,CAAC;IAEF,+DAA+D;IAC/D,MAAM,WAAW,GAAG,QAAQ,CAAC,IAAI,GAAG,IAAI,CAAC,GAAG,CAAC,QAAQ,CAAC,IAAI,EAAE,CAAC,CAAC,CAAC;IAC/D,MAAM,UAAU,GAAG,IAAI,CAAC,GAAG,CACzB,CAAC,EACD,CAAC,SAAS,CAAC,MAAM,GAAG,IAAI,CAAC,GAAG,GAAG,GAAG,IAAI,CAAC,GAAG,CAAC,WAAW,GAAG,CAAC,EAAE,GAAG,CAAC,CACjE,CAAC;IAEF,8CAA8C;IAC9C,KAAK,MAAM,CAAC,EAAE,EAAE,IAAI,CAAC,IAAI,QAAQ,CAAC,OAAO,EAAE,EAAE,CAAC;QAC5C,MAAM,KAAK,CAAC,UAAU,CAAC,EAAE,EAAE;YACzB,WAAW,EAAE,IAAI,CAAC,WAAW,GAAG,CAAC;YACjC,cAAc,EAAE,UAAU;SAC3B,CAAC,CAAC;IACL,CAAC;IAED,OAAO;QACL,KAAK;QACL,UAAU;QACV,SAAS,EAAE,SAAS,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC;YAC/B,IAAI,EAAE,CAAC,CAAC,IAAI,CAAC,IAAI;YACjB,IAAI,EAAE,CAAC,CAAC,IAAI,CAAC,IAAI;YACjB,KAAK,EAAE,CAAC,CAAC,KAAK;SACf,CAAC,CAAC;QACH,kBAAkB,EAAE,QAAQ,CAAC,IAAI;QACjC,mBAAmB,EAAE,QAAQ,CAAC,IAAI;QAClC,WAAW;QACX,aAAa;QACb,QAAQ;QACR,gBAAgB;QAChB,cAAc;QACd,kBAAkB;QAClB,kBAAkB;QAClB,UAAU;KACX,CAAC;AACJ,CAAC;AAED,gFAAgF;AAChF,kBAAkB;AAClB,gFAAgF;AAEhF,SAAS,qBAAqB,CAC5B,QAAiC,EACjC,QAAiC,EACjC,WAA+B;IAE/B,MAAM,KAAK,GACT,EAAE,CAAC;IAEL,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,WAAW,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC;QAChD,MAAM,QAAQ,GAAG,WAAW,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC;QACrC,MAAM,MAAM,GAAG,WAAW,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,IAAI,CAAC;QACvC,MAAM,QAAQ,GAAG,CAAC,GAAG,QAAQ,CAAC,MAAM,EAAE,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,IAAI,KAAK,QAAQ,CAAC,CAAC;QACzE,MAAM,MAAM,GAAG,CAAC,GAAG,QAAQ,CAAC,MAAM,EAAE,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,IAAI,KAAK,MAAM,CAAC,CAAC;QACrE,IAAI,CAAC,QAAQ,IAAI,CAAC,MAAM;YAAE,SAAS;QAEnC,MAAM,UAAU,GAAG,CAAC,GAAG,QAAQ,CAAC,MAAM,EAAE,CAAC,CAAC,IAAI,CAC5C,CAAC,CAAC,EAAE,EAAE,CACJ,CAAC,CAAC,CAAC,MAAM,KAAK,QAAQ,CAAC,EAAE,IAAI,CAAC,CAAC,MAAM,KAAK,MAAM,CAAC,EAAE,CAAC;YACpD,CAAC,CAAC,CAAC,MAAM,KAAK,MAAM,CAAC,EAAE,IAAI,CAAC,CAAC,MAAM,KAAK,QAAQ,CAAC,EAAE,CAAC,CACvD,CAAC;QAEF,IAAI,UAAU,EAAE,CAAC;YACf,KAAK,CAAC,IAAI,CAAC;gBACT,IAAI,EAAE,CAAC,QAAQ,EAAE,MAAM,CAAC;gBACxB,SAAS,EAAE,CAAC,UAAU,CAAC,QAAQ,CAAC;gBAChC,YAAY,EAAE,kDAAkD;aACjE,CAAC,CAAC;QACL,CAAC;IACH,CAAC;IAED,OAAO,KAAK,CAAC,KAAK,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC;AAC3B,CAAC;AAED,SAAS,uBAAuB,CAC9B,WAA6C,EAC7C,cAAoC,EACpC,QAA0B;IAE1B,MAAM,KAAK,GAAa,EAAE,CAAC;IAE3B,IAAI,cAAc,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;QAC9B,KAAK,CAAC,IAAI,CACR,+BAA+B,cAAc,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,MAAM,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC,EAAE,CAChF,CAAC;IACJ,CAAC;IAED,MAAM,SAAS,GAAG,WAAW,CAAC,MAAM,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,IAAI,KAAK,UAAU,CAAC,CAAC;IACnE,IAAI,SAAS,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;QACzB,KAAK,CAAC,IAAI,CACR,yBAAyB,SAAS,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC,EAAE,CACnE,CAAC;IACJ,CAAC;IAED,IAAI,QAAQ,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;QACxB,KAAK,CAAC,IAAI,CACR,8CAA8C,QAAQ,CAAC,CAAC,CAAC,CAAC,QAAQ,UAAU,QAAQ,CAAC,CAAC,CAAC,CAAC,QAAQ,GAAG,CACpG,CAAC;IACJ,CAAC;IAED,KAAK,CAAC,IAAI,CACR,oDAAoD,WAAW;SAC5D,KAAK,CAAC,CAAC,EAAE,CAAC,CAAC;SACX,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,IAAI,CAAC;SAClB,IAAI,CAAC,IAAI,CAAC,EAAE,CAChB,CAAC;IAEF,OAAO,KAAK,CAAC,KAAK,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC;AAC3B,CAAC"}

package/dist/index.js

@@ -31,7 +31,7 @@ async function main() {
     // 4. Register tools and resources
     registerMemoryTools(server, store, llm);
     registerMemoryResources(server, store);
-    const toolCount = llm ? 12 : 11;
+    const toolCount = llm ? 13 : 12;
     console.error(`[open-memory] Registered ${toolCount} tools and 2 resources`);
     // 5. Connect via stdio transport
     const transport = new StdioServerTransport();

package/dist/tools/memory-tools.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"memory-tools.d.ts","sourceRoot":"","sources":["../../src/tools/memory-tools.ts"],"names":[],"mappings":"AAIA,OAAO,EAAE,SAAS,EAAE,MAAM,yCAAyC,CAAC;AAGpE,OAAO,KAAK,EACV,cAAc,EAKd,WAAW,EACZ,MAAM,aAAa,CAAC;AAerB,wBAAgB,mBAAmB,CACjC,MAAM,EAAE,SAAS,EACjB,KAAK,EAAE,cAAc,EACrB,GAAG,CAAC,EAAE,WAAW,GAAG,IAAI,GACvB,IAAI,CA6BN"}
+{"version":3,"file":"memory-tools.d.ts","sourceRoot":"","sources":["../../src/tools/memory-tools.ts"],"names":[],"mappings":"AAIA,OAAO,EAAE,SAAS,EAAE,MAAM,yCAAyC,CAAC;AAGpE,OAAO,KAAK,EACV,cAAc,EAKd,WAAW,EACZ,MAAM,aAAa,CAAC;AAgBrB,wBAAgB,mBAAmB,CACjC,MAAM,EAAE,SAAS,EACjB,KAAK,EAAE,cAAc,EACrB,GAAG,CAAC,EAAE,WAAW,GAAG,IAAI,GACvB,IAAI,CAiCN"}

package/dist/tools/memory-tools.js

@@ -7,6 +7,7 @@ import { generateLocalEmbedding } from "../encoding/embedder.js";
 import { hybridSearch, getContextForTopic } from "../retrieval/search.js";
 import { consolidateMemory } from "../evolution/consolidator.js";
 import { encodeText } from "../encoding/pipeline.js";
+import { deepAnalyze } from "../analysis/deep-analyzer.js";
 import { DEFAULT_PAGE_SIZE, MAX_PAGE_SIZE, CHARACTER_LIMIT, } from "../constants.js";
 // =============================================================================
 // Tool Registration
@@ -26,6 +27,8 @@ export function registerMemoryTools(server, store, llm) {
     registerDeleteEntity(server, store);
     registerConsolidate(server, store);
     registerStatus(server, store);
+    // ---- ANALYSIS TOOLS ----
+    registerDeepAnalyze(server, store);
     // ---- OPTIONAL SERVER-SIDE ENCODING ----
     if (llm) {
         registerEncodeText(server, store, llm);
@@ -433,10 +436,12 @@ Provide either entity name or ID.`,
                 };
             }
             // Update access count
-            await store.updateNode(node.id, {
+            const updatedNode = await store.updateNode(node.id, {
                 accessCount: node.accessCount + 1,
                 lastAccessedAt: new Date().toISOString(),
             });
+            if (updatedNode)
+                node = updatedNode;
             // Get relationships
             const edges = await store.getEdgesForNode(node.id);
             // Resolve edge node names
@@ -856,6 +861,102 @@ Set autoStore=false to preview extraction results without saving.`,
     });
 }
 // =============================================================================
+// Analysis Tools
+// =============================================================================
+function registerDeepAnalyze(server, store) {
+    const InputSchema = z.object({
+        topic: z
+            .string()
+            .min(1)
+            .describe("The topic or question to perform deep analysis on"),
+        maxDepth: z
+            .number()
+            .int()
+            .min(1)
+            .max(5)
+            .default(3)
+            .describe("Graph traversal depth from each seed node (1=shallow, 5=deep)"),
+        topK: z
+            .number()
+            .int()
+            .min(5)
+            .max(30)
+            .default(15)
+            .describe("Number of seed nodes from initial hybrid search"),
+        includeTemporalAnalysis: z
+            .boolean()
+            .default(true)
+            .describe("Include temporal trend analysis grouped by calendar quarter"),
+        includeClusters: z
+            .boolean()
+            .default(true)
+            .describe("Detect topic clusters around hub entities"),
+        maxNodes: z
+            .number()
+            .int()
+            .min(20)
+            .max(200)
+            .default(100)
+            .describe("Hard cap on nodes explored — keeps performance bounded"),
+    });
+    server.registerTool("memory_deep_analyze", {
+        title: "Deep Analysis of a Topic",
+        description: `Perform multi-pass, graph-aware deep analysis of a topic across the entire memory graph.
+
+Goes far beyond memory_search by running 5 structured passes:
+1. Seed Search        — hybrid search to find top-K anchor nodes
+2. BFS Expansion      — full subgraph traversal to configurable depth
+3. Centrality/Cluster — surface hub entities and topic communities
+4. Temporal Analysis  — trend timeline grouped by quarter
+5. Contradiction Check — flags conflicting facts via semantic similarity
+
+Use before major decisions, refactors, or complex implementations.
+Returns: DeepAnalysisReport with keyEntities, clusters, temporalInsights, contradictions, synthesizedSummary, and suggestedNextSteps.`,
+        inputSchema: InputSchema,
+        annotations: {
+            readOnlyHint: true,
+            destructiveHint: false,
+            idempotentHint: false, // updates access counts
+            openWorldHint: false,
+        },
+    }, async (params) => {
+        try {
+            const report = await deepAnalyze(store, {
+                topic: params.topic,
+                maxDepth: params.maxDepth,
+                topK: params.topK,
+                includeTemporalAnalysis: params.includeTemporalAnalysis,
+                includeClusters: params.includeClusters,
+                maxNodes: params.maxNodes,
+            });
+            let text = JSON.stringify(report, null, 2);
+            if (text.length > CHARACTER_LIMIT) {
+                // Slim down to essentials when exceeding the token budget
+                const slim = {
+                    topic: report.topic,
+                    analyzedAt: report.analyzedAt,
+                    totalNodesExplored: report.totalNodesExplored,
+                    totalEdgesTraversed: report.totalEdgesTraversed,
+                    confidence: report.confidence,
+                    keyEntities: report.keyEntities.slice(0, 5),
+                    clusters: report.clusters.slice(0, 2),
+                    contradictions: report.contradictions.slice(0, 3),
+                    synthesizedSummary: report.synthesizedSummary,
+                    suggestedNextSteps: report.suggestedNextSteps,
+                    truncated: true,
+                };
+                text = JSON.stringify(slim, null, 2);
+            }
+            return {
+                content: [{ type: "text", text }],
+            };
+        }
+        catch (error) {
+            return errorResponse(error);
+        }
+    });
+}
+// =============================================================================
 // Helpers
 // =============================================================================
 function errorResponse(error) {