@ruvector/ruvllm 2.4.0 → 2.5.0

package/README.md

@@ -1,6 +1,6 @@
-# @ruvector/ruvllm v2.4
+# @ruvector/ruvllm v2.3
 
-Self-learning LLM orchestration with SONA adaptive learning, HNSW memory, RLM recursive retrieval, and SIMD inference for Node.js.
+Self-learning LLM orchestration with SONA adaptive learning, HNSW memory, and SIMD inference for Node.js.
 
 ## Installation
 
@@ -34,178 +34,18 @@ for await (const token of llm.stream('Write a haiku about Rust')) {
 }
 ```
 
-## What's New in v2.4
+## What's New in v2.3
 
 | Feature | Description |
 |---------|-------------|
-| **RLM (Recursive Language Model)** | Query decomposition with recursive retrieval and synthesis |
-| **100% Routing Accuracy** | Hybrid keyword-first strategy achieves 100% on Claude Code tasks |
-| **145 Tests Passing** | Comprehensive test coverage across all modules |
-| **Contrastive Fine-tuning** | LoRA-based training with 793 contrastive pairs |
-| **Training Scripts** | Generate routing datasets and fine-tune models |
-| **HuggingFace Models** | Pre-trained RuvLTRA models available |
-
-## RLM (Recursive Language Model)
-
-RLM provides recursive retrieval-augmented generation that breaks down complex queries into sub-queries and synthesizes answers from retrieved context.
-
-### Basic Usage
-
-```typescript
-import { RlmController } from '@ruvector/ruvllm';
-
-const rlm = new RlmController({
-  maxDepth: 5,
-  retrievalTopK: 10,
-  enableCache: true,
-});
-
-// Add knowledge to memory
-await rlm.addMemory('TypeScript adds static typing to JavaScript.');
-await rlm.addMemory('React is a library for building user interfaces.');
-
-// Query with recursive retrieval
-const answer = await rlm.query('What are causes and solutions for type errors in React?');
-console.log(answer.text);
-console.log('Sources:', answer.sources);
-console.log('Quality Score:', answer.qualityScore);
-console.log('Confidence:', answer.confidence);
-```
-
-### Streaming
-
-```typescript
-import { RlmController } from '@ruvector/ruvllm';
-
-const rlm = new RlmController();
-
-for await (const event of rlm.queryStream('Explain machine learning')) {
-  if (event.type === 'token') {
-    process.stdout.write(event.text);
-  } else {
-    console.log('\n\nQuality:', event.answer.qualityScore);
-  }
-}
-```
-
-### With Reflection
-
-```typescript
-import { RlmController } from '@ruvector/ruvllm';
-
-const rlm = new RlmController({
-  enableReflection: true,
-  maxReflectionIterations: 2,
-  minQualityScore: 0.8,
-});
-
-// Answers will be iteratively refined until quality >= 0.8
-const answer = await rlm.query('Complex multi-part technical question...');
-```
-
-### RLM Configuration
-
-```typescript
-interface RlmConfig {
-  maxDepth?: number;              // Max recursion depth (default: 3)
-  maxSubQueries?: number;         // Max sub-queries per level (default: 5)
-  tokenBudget?: number;           // Token budget (default: 4096)
-  enableCache?: boolean;          // Enable caching (default: true)
-  cacheTtl?: number;              // Cache TTL in ms (default: 300000)
-  retrievalTopK?: number;         // Memory spans to retrieve (default: 10)
-  minQualityScore?: number;       // Min quality threshold (default: 0.7)
-  enableReflection?: boolean;     // Enable self-reflection (default: false)
-  maxReflectionIterations?: number; // Max reflection loops (default: 2)
-}
-```
-
-## Exports
-
-```typescript
-import {
-  // Core
-  RuvLLM,
-  RuvLLMConfig,
-
-  // RLM - Recursive Language Model
-  RlmController,
-  RlmConfig,
-  RlmAnswer,
-  MemorySpan,
-  StreamToken,
-
-  // RLM Training
-  RlmTrainer,
-  RlmTrainingConfig,
-  RlmTrainingExample,
-  createRlmTrainer,
-  DEFAULT_RLM_CONFIG,
-  FAST_RLM_CONFIG,
-  THOROUGH_RLM_CONFIG,
-  ROUTING_FOCUSED_CONFIG,
-
-  // SONA Learning
-  SonaCoordinator,
-  TrajectoryBuilder,
-
-  // Federated Learning
-  EphemeralAgent,
-  FederatedCoordinator,
-
-  // LoRA Adapters
-  LoraAdapter,
-  LoraManager,
-
-  // Sessions
-  SessionManager,
-
-  // Contrastive Training
-  ContrastiveTrainer,
-
-  // Benchmarks
-  ModelComparisonBenchmark,
-  RoutingBenchmark,
-  EmbeddingBenchmark,
-} from '@ruvector/ruvllm';
-```
-
-## Performance
-
-### Test Results
-
-```
-145 tests passing
-  - RLM Controller: 24 tests
-  - Routing Accuracy: 18 tests
-  - Contrastive Training: 15 tests
-  - SIMD Operations: 22 tests
-  - SONA Learning: 19 tests
-  - Memory/HNSW: 21 tests
-  - Benchmarks: 26 tests
-```
-
-### Routing Accuracy (Claude Code Tasks)
-
-| Strategy | RuvLTRA | Qwen Base |
-|----------|---------|-----------|
-| Embedding Only | 45% | 40% |
-| Keyword-First (Hybrid) | **100%** | 95% |
-
-### Inference Performance (M4 Pro)
-
-| Operation | Performance |
-|-----------|-------------|
-| Inference | 88-135 tok/s |
-| Flash Attention | 320us (seq=2048) |
-| HNSW Search | 17-62us |
-| SONA Adapt | <1ms |
-| RLM Query | 50-200ms |
-
-### SIMD Optimizations
-
-- AVX2/AVX-512 on x86_64
-- NEON on ARM64
-- 4-8x speedup on vector operations
+| **RuvLTRA Models** | Purpose-built 0.5B & 3B models for Claude Flow |
+| **Task-Specific LoRA** | 5 pre-trained adapters (coder, researcher, security, architect, reviewer) |
+| **HuggingFace Hub** | Download/upload models directly |
+| **Adapter Merging** | TIES, DARE, SLERP strategies |
+| **HNSW Routing** | 150x faster semantic matching |
+| **Evaluation Harness** | SWE-Bench testing with 5 ablation modes |
+| **Auto-Dimension** | HNSW auto-detects model embedding size |
+| **mistral-rs Backend** | Production serving with PagedAttention, X-LoRA, ISQ (5-10x concurrent users) |
 
 ## CLI Usage
 
@@ -226,101 +66,6 @@ ruvllm bench ./models/model.gguf
 ruvllm eval --model ./models/model.gguf --subset lite --max-tasks 50
 ```
 
-## Training
-
-### Routing Dataset Generation
-
-Generate training data for agent routing:
-
-```bash
-# Generate routing dataset
-node scripts/training/routing-dataset.js
-
-# Output: 381 examples, 793 contrastive pairs
-```
-
-### Contrastive Fine-tuning
-
-Fine-tune models with LoRA adapters:
-
-```typescript
-import { ContrastiveTrainer, ContrastivePair } from '@ruvector/ruvllm';
-
-const trainer = new ContrastiveTrainer({
-  modelPath: './models/base.gguf',
-  loraRank: 8,
-  loraAlpha: 16,
-  learningRate: 1e-4,
-});
-
-// Training pairs from routing dataset
-const pairs: ContrastivePair[] = [
-  {
-    anchor: 'Fix the authentication bug in login.ts',
-    positive: 'coder',
-    negative: 'researcher',
-  },
-  // ... more pairs
-];
-
-await trainer.train(pairs, { epochs: 10, batchSize: 32 });
-await trainer.save('./adapters/routing-lora');
-```
-
-### Training Scripts
-
-Located in `scripts/training/`:
-
-| Script | Description |
-|--------|-------------|
-| `routing-dataset.js` | Generate 381 routing examples |
-| `claude-code-synth.js` | Synthetic data generation |
-| `contrastive-finetune.js` | LoRA fine-tuning pipeline |
-
-## Model Links
-
-### HuggingFace Repository
-
-**URL**: [https://huggingface.co/ruv/ruvltra](https://huggingface.co/ruv/ruvltra)
-
-### Available Models
-
-| Model | File | Size | Purpose |
-|-------|------|------|---------|
-| RuvLTRA Claude Code 0.5B | `ruvltra-claude-code-0.5b-q4_k_m.gguf` | ~400MB | Agent routing (100% with hybrid) |
-| RuvLTRA Small 0.5B | `ruvltra-0.5b-q4_k_m.gguf` | ~400MB | General embeddings |
-| RuvLTRA Medium 3B | `ruvltra-3b-q4_k_m.gguf` | ~2GB | Full LLM inference |
-
-### Download Models
-
-```bash
-# Using CLI
-ruvllm download ruv/ruvltra
-
-# Using HuggingFace CLI
-huggingface-cli download ruv/ruvltra ruvltra-claude-code-0.5b-q4_k_m.gguf
-
-# Programmatic download
-import { downloadModel } from '@ruvector/ruvllm';
-await downloadModel('ruv/ruvltra', { quantization: 'q4_k_m' });
-```
-
-### Environment Variables
-
-```bash
-# HuggingFace authentication (any of these)
-HF_TOKEN=hf_xxx
-HUGGING_FACE_HUB_TOKEN=hf_xxx
-HUGGINGFACE_API_KEY=hf_xxx
-```
-
-### Local Model Storage
-
-```bash
-~/.ruvllm/models/       # Downloaded GGUF models
-~/.ruvllm/training/     # Training data and configs
-```
-
 ## API Reference
 
 ### RuvLLM Class
@@ -338,10 +83,6 @@ class RuvLLM {
   // Load a model
   loadModel(path: string): Promise<void>;
 
-  // Memory operations
-  addMemory(text: string, metadata?: Record<string, unknown>): number;
-  searchMemory(query: string, topK?: number): MemoryResult[];
-
   // Get SONA learning stats
   sonaStats(): SonaStats | null;
 
@@ -350,32 +91,6 @@ class RuvLLM {
 }
 ```
 
-### RlmController Class
-
-```typescript
-class RlmController {
-  constructor(config?: RlmConfig, engine?: RuvLLM);
-
-  // Query with recursive retrieval
-  query(input: string): Promise<RlmAnswer>;
-
-  // Stream query
-  queryStream(input: string): AsyncGenerator<StreamToken>;
-
-  // Memory management
-  addMemory(text: string, metadata?: Record<string, unknown>): Promise<string>;
-  searchMemory(query: string, topK?: number): Promise<MemorySpan[]>;
-
-  // Cache management
-  clearCache(): void;
-  getCacheStats(): { size: number; entries: number };
-
-  // Configuration
-  updateConfig(config: Partial<RlmConfig>): void;
-  getConfig(): Required<RlmConfig>;
-}
-```
-
 ### Configuration
 
 ```typescript
@@ -422,6 +137,16 @@ const attended = simd.flashAttention(query, key, value, scale);
 simd.rmsNorm(hidden, weights, epsilon);
 ```
 
+## Performance (M4 Pro)
+
+| Operation | Performance |
+|-----------|-------------|
+| Inference | 88-135 tok/s |
+| Flash Attention | 320µs (seq=2048) |
+| HNSW Search | 17-62µs |
+| SONA Adapt | <1ms |
+| Evaluation | 5 ablation modes |
+
 ## Evaluation Harness
 
 Run model evaluations with SWE-Bench integration:
@@ -502,11 +227,11 @@ const response = await llm.query('Write production code');
 
 | Platform | Architecture | Status |
 |----------|--------------|--------|
-| macOS | arm64 (M1-M4) | Full support |
-| macOS | x64 | Supported |
-| Linux | x64 | Supported |
-| Linux | arm64 | Supported |
-| Windows | x64 | Supported |
+| macOS | arm64 (M1-M4) | ✅ Full support |
+| macOS | x64 | ✅ Supported |
+| Linux | x64 | ✅ Supported |
+| Linux | arm64 | ✅ Supported |
+| Windows | x64 | ✅ Supported |
 
 ## Related Packages
 
@@ -517,7 +242,6 @@ const response = await llm.query('Write production code');
 ## Links
 
 - [GitHub Repository](https://github.com/ruvnet/ruvector)
-- [HuggingFace Models](https://huggingface.co/ruv/ruvltra)
 - [API Documentation](https://docs.rs/ruvllm)
 - [Crate (Rust)](https://crates.io/crates/ruvllm)
 

package/dist/cjs/index.d.ts

@@ -62,7 +62,7 @@ export * from './training';
 export * from './contrastive';
 export * from './models';
 export * from './benchmarks';
-export * from './rlm';
+export * from './intelligence';
 export { version, hasSimdSupport } from './native';
 export { RuvLLM as default } from './engine';
 //# sourceMappingURL=index.d.ts.map

package/dist/cjs/index.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../../src/index.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GAkDG;AAGH,cAAc,SAAS,CAAC;AAGxB,cAAc,UAAU,CAAC;AAGzB,cAAc,QAAQ,CAAC;AAGvB,cAAc,WAAW,CAAC;AAG1B,cAAc,aAAa,CAAC;AAG5B,cAAc,QAAQ,CAAC;AAGvB,cAAc,aAAa,CAAC;AAG5B,cAAc,QAAQ,CAAC;AAGvB,cAAc,UAAU,CAAC;AAGzB,cAAc,YAAY,CAAC;AAG3B,cAAc,eAAe,CAAC;AAG9B,cAAc,UAAU,CAAC;AAGzB,cAAc,cAAc,CAAC;AAG7B,cAAc,OAAO,CAAC;AAGtB,OAAO,EAAE,OAAO,EAAE,cAAc,EAAE,MAAM,UAAU,CAAC;AAGnD,OAAO,EAAE,MAAM,IAAI,OAAO,EAAE,MAAM,UAAU,CAAC"}
+{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../../src/index.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GAkDG;AAGH,cAAc,SAAS,CAAC;AAGxB,cAAc,UAAU,CAAC;AAGzB,cAAc,QAAQ,CAAC;AAGvB,cAAc,WAAW,CAAC;AAG1B,cAAc,aAAa,CAAC;AAG5B,cAAc,QAAQ,CAAC;AAGvB,cAAc,aAAa,CAAC;AAG5B,cAAc,QAAQ,CAAC;AAGvB,cAAc,UAAU,CAAC;AAGzB,cAAc,YAAY,CAAC;AAG3B,cAAc,eAAe,CAAC;AAG9B,cAAc,UAAU,CAAC;AAGzB,cAAc,cAAc,CAAC;AAG7B,cAAc,gBAAgB,CAAC;AAG/B,OAAO,EAAE,OAAO,EAAE,cAAc,EAAE,MAAM,UAAU,CAAC;AAGnD,OAAO,EAAE,MAAM,IAAI,OAAO,EAAE,MAAM,UAAU,CAAC"}

package/dist/cjs/index.js

@@ -92,8 +92,8 @@ __exportStar(require("./contrastive"), exports);
 __exportStar(require("./models"), exports);
 // Benchmarks for Claude Code use cases
 __exportStar(require("./benchmarks"), exports);
-// RLM - Retrieval Language Model
-__exportStar(require("./rlm"), exports);
+// External Intelligence Providers (ADR-043)
+__exportStar(require("./intelligence"), exports);
 // Native bindings utilities
 var native_1 = require("./native");
 Object.defineProperty(exports, "version", { enumerable: true, get: function () { return native_1.version; } });
@@ -101,4 +101,4 @@ Object.defineProperty(exports, "hasSimdSupport", { enumerable: true, get: functi
 // Default export
 var engine_1 = require("./engine");
 Object.defineProperty(exports, "default", { enumerable: true, get: function () { return engine_1.RuvLLM; } });
-//# sourceMappingURL=data:application/json;base64,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
+//# sourceMappingURL=data:application/json;base64,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

package/dist/cjs/intelligence.d.ts

@@ -0,0 +1,136 @@
+/**
+ * External Intelligence Providers for SONA Learning (ADR-043)
+ *
+ * TypeScript bindings for the IntelligenceProvider trait, enabling
+ * external systems to feed quality signals into RuvLLM's learning loops.
+ *
+ * @example
+ * ```typescript
+ * import { IntelligenceLoader, FileSignalProvider, QualitySignal } from '@ruvector/ruvllm';
+ *
+ * const loader = new IntelligenceLoader();
+ * loader.registerProvider(new FileSignalProvider('./signals.json'));
+ *
+ * const { signals, errors } = loader.loadAllSignals();
+ * console.log(`Loaded ${signals.length} signals`);
+ * ```
+ */
+/**
+ * A quality signal from an external system.
+ *
+ * Represents one completed task with quality assessment data
+ * that can feed into SONA trajectories, the embedding classifier,
+ * and model router calibration.
+ */
+export interface QualitySignal {
+    /** Unique identifier for this signal */
+    id: string;
+    /** Human-readable task description (used for embedding generation) */
+    taskDescription: string;
+    /** Execution outcome */
+    outcome: 'success' | 'partial_success' | 'failure';
+    /** Composite quality score (0.0 - 1.0) */
+    qualityScore: number;
+    /** Optional human verdict */
+    humanVerdict?: 'approved' | 'rejected';
+    /** Optional structured quality factors for detailed analysis */
+    qualityFactors?: QualityFactors;
+    /** ISO 8601 timestamp of task completion */
+    completedAt: string;
+}
+/**
+ * Granular quality factor breakdown.
+ *
+ * Not all providers will have all factors. Undefined fields mean
+ * "not assessed" (distinct from 0.0, which means "assessed as zero").
+ */
+export interface QualityFactors {
+    acceptanceCriteriaMet?: number;
+    testsPassing?: number;
+    noRegressions?: number;
+    lintClean?: number;
+    typeCheckClean?: number;
+    followsPatterns?: number;
+    contextRelevance?: number;
+    reasoningCoherence?: number;
+    executionEfficiency?: number;
+}
+/**
+ * Quality weight overrides from a provider.
+ *
+ * Weights should sum to approximately 1.0.
+ */
+export interface ProviderQualityWeights {
+    taskCompletion: number;
+    codeQuality: number;
+    process: number;
+}
+/**
+ * Error from a single provider during batch loading.
+ */
+export interface ProviderError {
+    providerName: string;
+    message: string;
+}
+/**
+ * Result from a single provider during grouped loading.
+ */
+export interface ProviderResult {
+    providerName: string;
+    signals: QualitySignal[];
+    weights?: ProviderQualityWeights;
+}
+/**
+ * Interface for external systems that supply quality signals to RuvLLM.
+ *
+ * Implement this interface and register with IntelligenceLoader.
+ */
+export interface IntelligenceProvider {
+    /** Human-readable name for this provider */
+    name(): string;
+    /** Load quality signals from this provider's data source */
+    loadSignals(): QualitySignal[];
+    /** Optional quality weight overrides */
+    qualityWeights?(): ProviderQualityWeights | undefined;
+}
+/**
+ * Built-in file-based intelligence provider.
+ *
+ * Reads quality signals from a JSON file. This is the default provider
+ * for non-Rust integrations that write signal files.
+ */
+export declare class FileSignalProvider implements IntelligenceProvider {
+    private readonly filePath;
+    constructor(filePath: string);
+    name(): string;
+    loadSignals(): QualitySignal[];
+    qualityWeights(): ProviderQualityWeights | undefined;
+}
+/**
+ * Aggregates quality signals from multiple registered providers.
+ *
+ * If no providers are registered, loadAllSignals returns empty arrays
+ * with zero overhead.
+ */
+export declare class IntelligenceLoader {
+    private providers;
+    /** Register an external intelligence provider */
+    registerProvider(provider: IntelligenceProvider): void;
+    /** Returns the number of registered providers */
+    get providerCount(): number;
+    /** Returns the names of all registered providers */
+    get providerNames(): string[];
+    /**
+     * Load signals from all registered providers.
+     *
+     * Non-fatal: if a provider fails, its error is captured but
+     * other providers continue loading.
+     */
+    loadAllSignals(): {
+        signals: QualitySignal[];
+        errors: ProviderError[];
+    };
+    /** Load signals grouped by provider with weight overrides */
+    loadGrouped(): ProviderResult[];
+}
+//# sourceMappingURL=intelligence.d.ts.map

package/dist/cjs/intelligence.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"intelligence.d.ts","sourceRoot":"","sources":["../../src/intelligence.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;GAgBG;AAEH;;;;;;GAMG;AACH,MAAM,WAAW,aAAa;IAC5B,wCAAwC;IACxC,EAAE,EAAE,MAAM,CAAC;IACX,sEAAsE;IACtE,eAAe,EAAE,MAAM,CAAC;IACxB,wBAAwB;IACxB,OAAO,EAAE,SAAS,GAAG,iBAAiB,GAAG,SAAS,CAAC;IACnD,0CAA0C;IAC1C,YAAY,EAAE,MAAM,CAAC;IACrB,6BAA6B;IAC7B,YAAY,CAAC,EAAE,UAAU,GAAG,UAAU,CAAC;IACvC,gEAAgE;IAChE,cAAc,CAAC,EAAE,cAAc,CAAC;IAChC,4CAA4C;IAC5C,WAAW,EAAE,MAAM,CAAC;CACrB;AAED;;;;;GAKG;AACH,MAAM,WAAW,cAAc;IAC7B,qBAAqB,CAAC,EAAE,MAAM,CAAC;IAC/B,YAAY,CAAC,EAAE,MAAM,CAAC;IACtB,aAAa,CAAC,EAAE,MAAM,CAAC;IACvB,SAAS,CAAC,EAAE,MAAM,CAAC;IACnB,cAAc,CAAC,EAAE,MAAM,CAAC;IACxB,eAAe,CAAC,EAAE,MAAM,CAAC;IACzB,gBAAgB,CAAC,EAAE,MAAM,CAAC;IAC1B,kBAAkB,CAAC,EAAE,MAAM,CAAC;IAC5B,mBAAmB,CAAC,EAAE,MAAM,CAAC;CAC9B;AAED;;;;GAIG;AACH,MAAM,WAAW,sBAAsB;IACrC,cAAc,EAAE,MAAM,CAAC;IACvB,WAAW,EAAE,MAAM,CAAC;IACpB,OAAO,EAAE,MAAM,CAAC;CACjB;AAED;;GAEG;AACH,MAAM,WAAW,aAAa;IAC5B,YAAY,EAAE,MAAM,CAAC;IACrB,OAAO,EAAE,MAAM,CAAC;CACjB;AAED;;GAEG;AACH,MAAM,WAAW,cAAc;IAC7B,YAAY,EAAE,MAAM,CAAC;IACrB,OAAO,EAAE,aAAa,EAAE,CAAC;IACzB,OAAO,CAAC,EAAE,sBAAsB,CAAC;CAClC;AAED;;;;GAIG;AACH,MAAM,WAAW,oBAAoB;IACnC,4CAA4C;IAC5C,IAAI,IAAI,MAAM,CAAC;IACf,4DAA4D;IAC5D,WAAW,IAAI,aAAa,EAAE,CAAC;IAC/B,wCAAwC;IACxC,cAAc,CAAC,IAAI,sBAAsB,GAAG,SAAS,CAAC;CACvD;AAED;;;;;GAKG;AACH,qBAAa,kBAAmB,YAAW,oBAAoB;IAC7D,OAAO,CAAC,QAAQ,CAAC,QAAQ,CAAS;gBAEtB,QAAQ,EAAE,MAAM;IAI5B,IAAI,IAAI,MAAM;IAId,WAAW,IAAI,aAAa,EAAE;IA8B9B,cAAc,IAAI,sBAAsB,GAAG,SAAS;CAiBrD;AAgBD;;;;;GAKG;AACH,qBAAa,kBAAkB;IAC7B,OAAO,CAAC,SAAS,CAA8B;IAE/C,iDAAiD;IACjD,gBAAgB,CAAC,QAAQ,EAAE,oBAAoB,GAAG,IAAI;IAItD,iDAAiD;IACjD,IAAI,aAAa,IAAI,MAAM,CAE1B;IAED,oDAAoD;IACpD,IAAI,aAAa,IAAI,MAAM,EAAE,CAE5B;IAED;;;;;OAKG;IACH,cAAc,IAAI;QAAE,OAAO,EAAE,aAAa,EAAE,CAAC;QAAC,MAAM,EAAE,aAAa,EAAE,CAAA;KAAE;IAmBvE,6DAA6D;IAC7D,WAAW,IAAI,cAAc,EAAE;CAehC"}