[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6EX3EYHFD0H9EMQ1T04R3AX":3,"skill-same:01M6EX3EYHFD0H9EMQ1T04R3AX":81},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":23,"publishedAt":44,"linkOnly":45,"markdownContent":46,"markdownBody":47,"frontMatter":48,"refName":53,"commitSha":54,"skillDirectory":55,"forksCount":56,"repositoryHtmlUrl":57,"repositoryHomepageUrl":58,"ownerHtmlUrl":59,"ownerBlogUrl":6,"ownerDescription":60,"updatedAt":61,"install":62,"resources":74},"01M6EX3EYHFD0H9EMQ1T04R3AX","flow-nexus-neural",null,"Train and deploy neural networks in distributed E2B sandboxes with Flow Nexus","ComeOnOliver\u002Fskillshub","ComeOnOliver","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F43215937?v=4","01M6E5JNFECJAAVFE0W7SMWWD2","Haoran Shu","01M6E5JNNDQS7AJZDYFQ8TPZCK",66,"MIT","https:\u002F\u002Fgithub.com\u002FComeOnOliver\u002Fskillshub\u002Fblob\u002Fmain\u002Fskills\u002Faiskillstore\u002Fmarketplace\u002Fruvnet\u002Fflow-nexus-neural\u002FSKILL.md",[18],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000011","运维与部署","SCENARIO","使用场景",[24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43],"flow nexus neural","flow","Neural","Nexus","deep-learning","sandboxes","e2b-sandboxes","flow-nexus","distributed-training","e2b","machine","deep","neural-networks","distributed","learning","networks","machine-learning","training","ml","neural networks","2026-11-10 10:12:30",false,"---\nname: flow-nexus-neural\ndescription: Train and deploy neural networks in distributed E2B sandboxes with Flow Nexus\nversion: 1.0.0\ncategory: ai-ml\ntags:\n  - neural-networks\n  - distributed-training\n  - machine-learning\n  - deep-learning\n  - flow-nexus\n  - e2b-sandboxes\nrequires_auth: true\nmcp_server: flow-nexus\n---\n\n# Flow Nexus Neural Networks\n\nDeploy, train, and manage neural networks in distributed E2B sandbox environments. Train custom models with multiple architectures (feedforward, LSTM, GAN, transformer) or use pre-built templates from the marketplace.\n\n## Prerequisites\n\n```bash\n# Add Flow Nexus MCP server\nclaude mcp add flow-nexus npx flow-nexus@latest mcp start\n\n# Register and login\nnpx flow-nexus@latest register\nnpx flow-nexus@latest login\n```\n\n## Core Capabilities\n\n### 1. Single-Node Neural Training\n\nTrain neural networks with custom architectures and configurations.\n\n**Available Architectures:**\n- `feedforward` - Standard fully-connected networks\n- `lstm` - Long Short-Term Memory for sequences\n- `gan` - Generative Adversarial Networks\n- `autoencoder` - Dimensionality reduction\n- `transformer` - Attention-based models\n\n**Training Tiers:**\n- `nano` - Minimal resources (fast, limited)\n- `mini` - Small models\n- `small` - Standard models\n- `medium` - Complex models\n- `large` - Large-scale training\n\n#### Example: Train Custom Classifier\n\n```javascript\nmcp__flow-nexus__neural_train({\n  config: {\n    architecture: {\n      type: \"feedforward\",\n      layers: [\n        { type: \"dense\", units: 256, activation: \"relu\" },\n        { type: \"dropout\", rate: 0.3 },\n        { type: \"dense\", units: 128, activation: \"relu\" },\n        { type: \"dropout\", rate: 0.2 },\n        { type: \"dense\", units: 64, activation: \"relu\" },\n        { type: \"dense\", units: 10, activation: \"softmax\" }\n      ]\n    },\n    training: {\n      epochs: 100,\n      batch_size: 32,\n      learning_rate: 0.001,\n      optimizer: \"adam\"\n    },\n    divergent: {\n      enabled: true,\n      pattern: \"lateral\", \u002F\u002F quantum, chaotic, associative, evolutionary\n      factor: 0.5\n    }\n  },\n  tier: \"small\",\n  user_id: \"your_user_id\"\n})\n```\n\n#### Example: LSTM for Time Series\n\n```javascript\nmcp__flow-nexus__neural_train({\n  config: {\n    architecture: {\n      type: \"lstm\",\n      layers: [\n        { type: \"lstm\", units: 128, return_sequences: true },\n        { type: \"dropout\", rate: 0.2 },\n        { type: \"lstm\", units: 64 },\n        { type: \"dense\", units: 1, activation: \"linear\" }\n      ]\n    },\n    training: {\n      epochs: 150,\n      batch_size: 64,\n      learning_rate: 0.01,\n      optimizer: \"adam\"\n    }\n  },\n  tier: \"medium\"\n})\n```\n\n#### Example: Transformer Architecture\n\n```javascript\nmcp__flow-nexus__neural_train({\n  config: {\n    architecture: {\n      type: \"transformer\",\n      layers: [\n        { type: \"embedding\", vocab_size: 10000, embedding_dim: 512 },\n        { type: \"transformer_encoder\", num_heads: 8, ff_dim: 2048 },\n        { type: \"global_average_pooling\" },\n        { type: \"dense\", units: 128, activation: \"relu\" },\n        { type: \"dense\", units: 2, activation: \"softmax\" }\n      ]\n    },\n    training: {\n      epochs: 50,\n      batch_size: 16,\n      learning_rate: 0.0001,\n      optimizer: \"adam\"\n    }\n  },\n  tier: \"large\"\n})\n```\n\n### 2. Model Inference\n\nRun predictions on trained models.\n\n```javascript\nmcp__flow-nexus__neural_predict({\n  model_id: \"model_abc123\",\n  input: [\n    [0.5, 0.3, 0.2, 0.1],\n    [0.8, 0.1, 0.05, 0.05],\n    [0.2, 0.6, 0.15, 0.05]\n  ],\n  user_id: \"your_user_id\"\n})\n```\n\n**Response:**\n```json\n{\n  \"predictions\": [\n    [0.12, 0.85, 0.03],\n    [0.89, 0.08, 0.03],\n    [0.05, 0.92, 0.03]\n  ],\n  \"inference_time_ms\": 45,\n  \"model_version\": \"1.0.0\"\n}\n```\n\n### 3. Template Marketplace\n\nBrowse and deploy pre-trained models from the marketplace.\n\n#### List Available Templates\n\n```javascript\nmcp__flow-nexus__neural_list_templates({\n  category: \"classification\", \u002F\u002F timeseries, regression, nlp, vision, anomaly, generative\n  tier: \"free\", \u002F\u002F or \"paid\"\n  search: \"sentiment\",\n  limit: 20\n})\n```\n\n**Response:**\n```json\n{\n  \"templates\": [\n    {\n      \"id\": \"sentiment-analysis-v2\",\n      \"name\": \"Sentiment Analysis Classifier\",\n      \"description\": \"Pre-trained BERT model for sentiment analysis\",\n      \"category\": \"nlp\",\n      \"accuracy\": 0.94,\n      \"downloads\": 1523,\n      \"tier\": \"free\"\n    },\n    {\n      \"id\": \"image-classifier-resnet\",\n      \"name\": \"ResNet Image Classifier\",\n      \"description\": \"ResNet-50 for image classification\",\n      \"category\": \"vision\",\n      \"accuracy\": 0.96,\n      \"downloads\": 2341,\n      \"tier\": \"paid\"\n    }\n  ]\n}\n```\n\n#### Deploy Template\n\n```javascript\nmcp__flow-nexus__neural_deploy_template({\n  template_id: \"sentiment-analysis-v2\",\n  custom_config: {\n    training: {\n      epochs: 50,\n      learning_rate: 0.0001\n    }\n  },\n  user_id: \"your_user_id\"\n})\n```\n\n### 4. Distributed Training Clusters\n\nTrain large models across multiple E2B sandboxes with distributed computing.\n\n#### Initialize Cluster\n\n```javascript\nmcp__flow-nexus__neural_cluster_init({\n  name: \"large-model-cluster\",\n  architecture: \"transformer\", \u002F\u002F transformer, cnn, rnn, gnn, hybrid\n  topology: \"mesh\", \u002F\u002F mesh, ring, star, hierarchical\n  consensus: \"proof-of-learning\", \u002F\u002F byzantine, raft, gossip\n  daaEnabled: true, \u002F\u002F Decentralized Autonomous Agents\n  wasmOptimization: true\n})\n```\n\n**Response:**\n```json\n{\n  \"cluster_id\": \"cluster_xyz789\",\n  \"name\": \"large-model-cluster\",\n  \"status\": \"initializing\",\n  \"topology\": \"mesh\",\n  \"max_nodes\": 100,\n  \"created_at\": \"2025-10-19T10:30:00Z\"\n}\n```\n\n#### Deploy Worker Nodes\n\n```javascript\n\u002F\u002F Deploy parameter server\nmcp__flow-nexus__neural_node_deploy({\n  cluster_id: \"cluster_xyz789\",\n  node_type: \"parameter_server\",\n  model: \"large\",\n  template: \"nodejs\",\n  capabilities: [\"parameter_management\", \"gradient_aggregation\"],\n  autonomy: 0.8\n})\n\n\u002F\u002F Deploy worker nodes\nmcp__flow-nexus__neural_node_deploy({\n  cluster_id: \"cluster_xyz789\",\n  node_type: \"worker\",\n  model: \"xl\",\n  role: \"worker\",\n  capabilities: [\"training\", \"inference\"],\n  layers: [\n    { type: \"transformer_encoder\", num_heads: 16 },\n    { type: \"feed_forward\", units: 4096 }\n  ],\n  autonomy: 0.9\n})\n\n\u002F\u002F Deploy aggregator\nmcp__flow-nexus__neural_node_deploy({\n  cluster_id: \"cluster_xyz789\",\n  node_type: \"aggregator\",\n  model: \"large\",\n  capabilities: [\"gradient_aggregation\", \"model_synchronization\"]\n})\n```\n\n#### Connect Cluster Topology\n\n```javascript\nmcp__flow-nexus__neural_cluster_connect({\n  cluster_id: \"cluster_xyz789\",\n  topology: \"mesh\" \u002F\u002F Override default if needed\n})\n```\n\n#### Start Distributed Training\n\n```javascript\nmcp__flow-nexus__neural_train_distributed({\n  cluster_id: \"cluster_xyz789\",\n  dataset: \"imagenet\", \u002F\u002F or custom dataset identifier\n  epochs: 100,\n  batch_size: 128,\n  learning_rate: 0.001,\n  optimizer: \"adam\", \u002F\u002F sgd, rmsprop, adagrad\n  federated: true \u002F\u002F Enable federated learning\n})\n```\n\n**Federated Learning Example:**\n```javascript\nmcp__flow-nexus__neural_train_distributed({\n  cluster_id: \"cluster_xyz789\",\n  dataset: \"medical_images_distributed\",\n  epochs: 200,\n  batch_size: 64,\n  learning_rate: 0.0001,\n  optimizer: \"adam\",\n  federated: true, \u002F\u002F Data stays on local nodes\n  aggregation_rounds: 50,\n  min_nodes_per_round: 5\n})\n```\n\n#### Monitor Cluster Status\n\n```javascript\nmcp__flow-nexus__neural_cluster_status({\n  cluster_id: \"cluster_xyz789\"\n})\n```\n\n**Response:**\n```json\n{\n  \"cluster_id\": \"cluster_xyz789\",\n  \"status\": \"training\",\n  \"nodes\": [\n    {\n      \"node_id\": \"node_001\",\n      \"type\": \"parameter_server\",\n      \"status\": \"active\",\n      \"cpu_usage\": 0.75,\n      \"memory_usage\": 0.82\n    },\n    {\n      \"node_id\": \"node_002\",\n      \"type\": \"worker\",\n      \"status\": \"active\",\n      \"training_progress\": 0.45\n    }\n  ],\n  \"training_metrics\": {\n    \"current_epoch\": 45,\n    \"total_epochs\": 100,\n    \"loss\": 0.234,\n    \"accuracy\": 0.891\n  }\n}\n```\n\n#### Run Distributed Inference\n\n```javascript\nmcp__flow-nexus__neural_predict_distributed({\n  cluster_id: \"cluster_xyz789\",\n  input_data: JSON.stringify([\n    [0.1, 0.2, 0.3],\n    [0.4, 0.5, 0.6]\n  ]),\n  aggregation: \"ensemble\" \u002F\u002F mean, majority, weighted, ensemble\n})\n```\n\n#### Terminate Cluster\n\n```javascript\nmcp__flow-nexus__neural_cluster_terminate({\n  cluster_id: \"cluster_xyz789\"\n})\n```\n\n### 5. Model Management\n\n#### List Your Models\n\n```javascript\nmcp__flow-nexus__neural_list_models({\n  user_id: \"your_user_id\",\n  include_public: true\n})\n```\n\n**Response:**\n```json\n{\n  \"models\": [\n    {\n      \"model_id\": \"model_abc123\",\n      \"name\": \"Custom Classifier v1\",\n      \"architecture\": \"feedforward\",\n      \"accuracy\": 0.92,\n      \"created_at\": \"2025-10-15T14:20:00Z\",\n      \"status\": \"trained\"\n    },\n    {\n      \"model_id\": \"model_def456\",\n      \"name\": \"LSTM Forecaster\",\n      \"architecture\": \"lstm\",\n      \"mse\": 0.0045,\n      \"created_at\": \"2025-10-18T09:15:00Z\",\n      \"status\": \"training\"\n    }\n  ]\n}\n```\n\n#### Check Training Status\n\n```javascript\nmcp__flow-nexus__neural_training_status({\n  job_id: \"job_training_xyz\"\n})\n```\n\n**Response:**\n```json\n{\n  \"job_id\": \"job_training_xyz\",\n  \"status\": \"training\",\n  \"progress\": 0.67,\n  \"current_epoch\": 67,\n  \"total_epochs\": 100,\n  \"current_loss\": 0.234,\n  \"estimated_completion\": \"2025-10-19T12:45:00Z\"\n}\n```\n\n#### Performance Benchmarking\n\n```javascript\nmcp__flow-nexus__neural_performance_benchmark({\n  model_id: \"model_abc123\",\n  benchmark_type: \"comprehensive\" \u002F\u002F inference, throughput, memory, comprehensive\n})\n```\n\n**Response:**\n```json\n{\n  \"model_id\": \"model_abc123\",\n  \"benchmarks\": {\n    \"inference_latency_ms\": 12.5,\n    \"throughput_qps\": 8000,\n    \"memory_usage_mb\": 245,\n    \"gpu_utilization\": 0.78,\n    \"accuracy\": 0.92,\n    \"f1_score\": 0.89\n  },\n  \"timestamp\": \"2025-10-19T11:00:00Z\"\n}\n```\n\n#### Create Validation Workflow\n\n```javascript\nmcp__flow-nexus__neural_validation_workflow({\n  model_id: \"model_abc123\",\n  user_id: \"your_user_id\",\n  validation_type: \"comprehensive\" \u002F\u002F performance, accuracy, robustness, comprehensive\n})\n```\n\n### 6. Publishing and Marketplace\n\n#### Publish Model as Template\n\n```javascript\nmcp__flow-nexus__neural_publish_template({\n  model_id: \"model_abc123\",\n  name: \"High-Accuracy Sentiment Classifier\",\n  description: \"Fine-tuned BERT model for sentiment analysis with 94% accuracy\",\n  category: \"nlp\",\n  price: 0, \u002F\u002F 0 for free, or credits amount\n  user_id: \"your_user_id\"\n})\n```\n\n#### Rate a Template\n\n```javascript\nmcp__flow-nexus__neural_rate_template({\n  template_id: \"sentiment-analysis-v2\",\n  rating: 5,\n  review: \"Excellent model! Achieved 95% accuracy on my dataset.\",\n  user_id: \"your_user_id\"\n})\n```\n\n## Common Use Cases\n\n### Image Classification with CNN\n\n```javascript\n\u002F\u002F Initialize cluster for large-scale image training\nconst cluster = await mcp__flow-nexus__neural_cluster_init({\n  name: \"image-classification-cluster\",\n  architecture: \"cnn\",\n  topology: \"hierarchical\",\n  wasmOptimization: true\n})\n\n\u002F\u002F Deploy worker nodes\nawait mcp__flow-nexus__neural_node_deploy({\n  cluster_id: cluster.cluster_id,\n  node_type: \"worker\",\n  model: \"large\",\n  capabilities: [\"training\", \"data_augmentation\"]\n})\n\n\u002F\u002F Start training\nawait mcp__flow-nexus__neural_train_distributed({\n  cluster_id: cluster.cluster_id,\n  dataset: \"custom_images\",\n  epochs: 100,\n  batch_size: 64,\n  learning_rate: 0.001,\n  optimizer: \"adam\"\n})\n```\n\n### NLP Sentiment Analysis\n\n```javascript\n\u002F\u002F Use pre-built template\nconst deployment = await mcp__flow-nexus__neural_deploy_template({\n  template_id: \"sentiment-analysis-v2\",\n  custom_config: {\n    training: {\n      epochs: 30,\n      batch_size: 16\n    }\n  }\n})\n\n\u002F\u002F Run inference\nconst result = await mcp__flow-nexus__neural_predict({\n  model_id: deployment.model_id,\n  input: [\"This product is amazing!\", \"Terrible experience.\"]\n})\n```\n\n### Time Series Forecasting\n\n```javascript\n\u002F\u002F Train LSTM model\nconst training = await mcp__flow-nexus__neural_train({\n  config: {\n    architecture: {\n      type: \"lstm\",\n      layers: [\n        { type: \"lstm\", units: 128, return_sequences: true },\n        { type: \"dropout\", rate: 0.2 },\n        { type: \"lstm\", units: 64 },\n        { type: \"dense\", units: 1 }\n      ]\n    },\n    training: {\n      epochs: 150,\n      batch_size: 64,\n      learning_rate: 0.01,\n      optimizer: \"adam\"\n    }\n  },\n  tier: \"medium\"\n})\n\n\u002F\u002F Monitor progress\nconst status = await mcp__flow-nexus__neural_training_status({\n  job_id: training.job_id\n})\n```\n\n### Federated Learning for Privacy\n\n```javascript\n\u002F\u002F Initialize federated cluster\nconst cluster = await mcp__flow-nexus__neural_cluster_init({\n  name: \"federated-medical-cluster\",\n  architecture: \"transformer\",\n  topology: \"mesh\",\n  consensus: \"proof-of-learning\",\n  daaEnabled: true\n})\n\n\u002F\u002F Deploy nodes across different locations\nfor (let i = 0; i \u003C 5; i++) {\n  await mcp__flow-nexus__neural_node_deploy({\n    cluster_id: cluster.cluster_id,\n    node_type: \"worker\",\n    model: \"large\",\n    autonomy: 0.9\n  })\n}\n\n\u002F\u002F Train with federated learning (data never leaves nodes)\nawait mcp__flow-nexus__neural_train_distributed({\n  cluster_id: cluster.cluster_id,\n  dataset: \"medical_records_distributed\",\n  epochs: 200,\n  federated: true,\n  aggregation_rounds: 100\n})\n```\n\n## Architecture Patterns\n\n### Feedforward Networks\nBest for: Classification, regression, simple pattern recognition\n```javascript\n{\n  type: \"feedforward\",\n  layers: [\n    { type: \"dense\", units: 256, activation: \"relu\" },\n    { type: \"dropout\", rate: 0.3 },\n    { type: \"dense\", units: 128, activation: \"relu\" },\n    { type: \"dense\", units: 10, activation: \"softmax\" }\n  ]\n}\n```\n\n### LSTM Networks\nBest for: Time series, sequences, forecasting\n```javascript\n{\n  type: \"lstm\",\n  layers: [\n    { type: \"lstm\", units: 128, return_sequences: true },\n    { type: \"lstm\", units: 64 },\n    { type: \"dense\", units: 1 }\n  ]\n}\n```\n\n### Transformers\nBest for: NLP, attention mechanisms, large-scale text\n```javascript\n{\n  type: \"transformer\",\n  layers: [\n    { type: \"embedding\", vocab_size: 10000, embedding_dim: 512 },\n    { type: \"transformer_encoder\", num_heads: 8, ff_dim: 2048 },\n    { type: \"global_average_pooling\" },\n    { type: \"dense\", units: 2, activation: \"softmax\" }\n  ]\n}\n```\n\n### GANs\nBest for: Generative tasks, image synthesis\n```javascript\n{\n  type: \"gan\",\n  generator_layers: [...],\n  discriminator_layers: [...]\n}\n```\n\n### Autoencoders\nBest for: Dimensionality reduction, anomaly detection\n```javascript\n{\n  type: \"autoencoder\",\n  encoder_layers: [\n    { type: \"dense\", units: 128, activation: \"relu\" },\n    { type: \"dense\", units: 64, activation: \"relu\" }\n  ],\n  decoder_layers: [\n    { type: \"dense\", units: 128, activation: \"relu\" },\n    { type: \"dense\", units: input_dim, activation: \"sigmoid\" }\n  ]\n}\n```\n\n## Best Practices\n\n1. **Start Small**: Begin with `nano` or `mini` tiers for experimentation\n2. **Use Templates**: Leverage marketplace templates for common tasks\n3. **Monitor Training**: Check status regularly to catch issues early\n4. **Benchmark Models**: Always benchmark before production deployment\n5. **Distributed Training**: Use clusters for large models (>1B parameters)\n6. **Federated Learning**: Use for privacy-sensitive data\n7. **Version Models**: Publish successful models as templates for reuse\n8. **Validate Thoroughly**: Use validation workflows before deployment\n\n## Troubleshooting\n\n### Training Stalled\n```javascript\n\u002F\u002F Check cluster status\nconst status = await mcp__flow-nexus__neural_cluster_status({\n  cluster_id: \"cluster_id\"\n})\n\n\u002F\u002F Terminate and restart if needed\nawait mcp__flow-nexus__neural_cluster_terminate({\n  cluster_id: \"cluster_id\"\n})\n```\n\n### Low Accuracy\n- Increase epochs\n- Adjust learning rate\n- Add regularization (dropout)\n- Try different optimizer\n- Use data augmentation\n\n### Out of Memory\n- Reduce batch size\n- Use smaller model tier\n- Enable gradient accumulation\n- Use distributed training\n\n## Related Skills\n\n- `flow-nexus-sandbox` - E2B sandbox management\n- `flow-nexus-swarm` - AI swarm orchestration\n- `flow-nexus-workflow` - Workflow automation\n\n## Resources\n\n- Flow Nexus Docs: https:\u002F\u002Fflow-nexus.ruv.io\u002Fdocs\n- Neural Network Guide: https:\u002F\u002Fflow-nexus.ruv.io\u002Fdocs\u002Fneural\n- Template Marketplace: https:\u002F\u002Fflow-nexus.ruv.io\u002Ftemplates\n- API Reference: https:\u002F\u002Fflow-nexus.ruv.io\u002Fapi\n\n---\n\n**Note**: Distributed training requires authentication. Register at https:\u002F\u002Fflow-nexus.ruv.io or use `npx flow-nexus@latest register`.\n","# Flow Nexus Neural Networks\n\nDeploy, train, and manage neural networks in distributed E2B sandbox environments. Train custom models with multiple architectures (feedforward, LSTM, GAN, transformer) or use pre-built templates from the marketplace.\n\n## Prerequisites\n\n```bash\n# Add Flow Nexus MCP server\nclaude mcp add flow-nexus npx flow-nexus@latest mcp start\n\n# Register and login\nnpx flow-nexus@latest register\nnpx flow-nexus@latest login\n```\n\n## Core Capabilities\n\n### 1. Single-Node Neural Training\n\nTrain neural networks with custom architectures and configurations.\n\n**Available Architectures:**\n- `feedforward` - Standard fully-connected networks\n- `lstm` - Long Short-Term Memory for sequences\n- `gan` - Generative Adversarial Networks\n- `autoencoder` - Dimensionality reduction\n- `transformer` - Attention-based models\n\n**Training Tiers:**\n- `nano` - Minimal resources (fast, limited)\n- `mini` - Small models\n- `small` - Standard models\n- `medium` - Complex models\n- `large` - Large-scale training\n\n#### Example: Train Custom Classifier\n\n```javascript\nmcp__flow-nexus__neural_train({\n  config: {\n    architecture: {\n      type: \"feedforward\",\n      layers: [\n        { type: \"dense\", units: 256, activation: \"relu\" },\n        { type: \"dropout\", rate: 0.3 },\n        { type: \"dense\", units: 128, activation: \"relu\" },\n        { type: \"dropout\", rate: 0.2 },\n        { type: \"dense\", units: 64, activation: \"relu\" },\n        { type: \"dense\", units: 10, activation: \"softmax\" }\n      ]\n    },\n    training: {\n      epochs: 100,\n      batch_size: 32,\n      learning_rate: 0.001,\n      optimizer: \"adam\"\n    },\n    divergent: {\n      enabled: true,\n      pattern: \"lateral\", \u002F\u002F quantum, chaotic, associative, evolutionary\n      factor: 0.5\n    }\n  },\n  tier: \"small\",\n  user_id: \"your_user_id\"\n})\n```\n\n#### Example: LSTM for Time Series\n\n```javascript\nmcp__flow-nexus__neural_train({\n  config: {\n    architecture: {\n      type: \"lstm\",\n      layers: [\n        { type: \"lstm\", units: 128, return_sequences: true },\n        { type: \"dropout\", rate: 0.2 },\n        { type: \"lstm\", units: 64 },\n        { type: \"dense\", units: 1, activation: \"linear\" }\n      ]\n    },\n    training: {\n      epochs: 150,\n      batch_size: 64,\n      learning_rate: 0.01,\n      optimizer: \"adam\"\n    }\n  },\n  tier: \"medium\"\n})\n```\n\n#### Example: Transformer Architecture\n\n```javascript\nmcp__flow-nexus__neural_train({\n  config: {\n    architecture: {\n      type: \"transformer\",\n      layers: [\n        { type: \"embedding\", vocab_size: 10000, embedding_dim: 512 },\n        { type: \"transformer_encoder\", num_heads: 8, ff_dim: 2048 },\n        { type: \"global_average_pooling\" },\n        { type: \"dense\", units: 128, activation: \"relu\" },\n        { type: \"dense\", units: 2, activation: \"softmax\" }\n      ]\n    },\n    training: {\n      epochs: 50,\n      batch_size: 16,\n      learning_rate: 0.0001,\n      optimizer: \"adam\"\n    }\n  },\n  tier: \"large\"\n})\n```\n\n### 2. Model Inference\n\nRun predictions on trained models.\n\n```javascript\nmcp__flow-nexus__neural_predict({\n  model_id: \"model_abc123\",\n  input: [\n    [0.5, 0.3, 0.2, 0.1],\n    [0.8, 0.1, 0.05, 0.05],\n    [0.2, 0.6, 0.15, 0.05]\n  ],\n  user_id: \"your_user_id\"\n})\n```\n\n**Response:**\n```json\n{\n  \"predictions\": [\n    [0.12, 0.85, 0.03],\n    [0.89, 0.08, 0.03],\n    [0.05, 0.92, 0.03]\n  ],\n  \"inference_time_ms\": 45,\n  \"model_version\": \"1.0.0\"\n}\n```\n\n### 3. Template Marketplace\n\nBrowse and deploy pre-trained models from the marketplace.\n\n#### List Available Templates\n\n```javascript\nmcp__flow-nexus__neural_list_templates({\n  category: \"classification\", \u002F\u002F timeseries, regression, nlp, vision, anomaly, generative\n  tier: \"free\", \u002F\u002F or \"paid\"\n  search: \"sentiment\",\n  limit: 20\n})\n```\n\n**Response:**\n```json\n{\n  \"templates\": [\n    {\n      \"id\": \"sentiment-analysis-v2\",\n      \"name\": \"Sentiment Analysis Classifier\",\n      \"description\": \"Pre-trained BERT model for sentiment analysis\",\n      \"category\": \"nlp\",\n      \"accuracy\": 0.94,\n      \"downloads\": 1523,\n      \"tier\": \"free\"\n    },\n    {\n      \"id\": \"image-classifier-resnet\",\n      \"name\": \"ResNet Image Classifier\",\n      \"description\": \"ResNet-50 for image classification\",\n      \"category\": \"vision\",\n      \"accuracy\": 0.96,\n      \"downloads\": 2341,\n      \"tier\": \"paid\"\n    }\n  ]\n}\n```\n\n#### Deploy Template\n\n```javascript\nmcp__flow-nexus__neural_deploy_template({\n  template_id: \"sentiment-analysis-v2\",\n  custom_config: {\n    training: {\n      epochs: 50,\n      learning_rate: 0.0001\n    }\n  },\n  user_id: \"your_user_id\"\n})\n```\n\n### 4. Distributed Training Clusters\n\nTrain large models across multiple E2B sandboxes with distributed computing.\n\n#### Initialize Cluster\n\n```javascript\nmcp__flow-nexus__neural_cluster_init({\n  name: \"large-model-cluster\",\n  architecture: \"transformer\", \u002F\u002F transformer, cnn, rnn, gnn, hybrid\n  topology: \"mesh\", \u002F\u002F mesh, ring, star, hierarchical\n  consensus: \"proof-of-learning\", \u002F\u002F byzantine, raft, gossip\n  daaEnabled: true, \u002F\u002F Decentralized Autonomous Agents\n  wasmOptimization: true\n})\n```\n\n**Response:**\n```json\n{\n  \"cluster_id\": \"cluster_xyz789\",\n  \"name\": \"large-model-cluster\",\n  \"status\": \"initializing\",\n  \"topology\": \"mesh\",\n  \"max_nodes\": 100,\n  \"created_at\": \"2025-10-19T10:30:00Z\"\n}\n```\n\n#### Deploy Worker Nodes\n\n```javascript\n\u002F\u002F Deploy parameter server\nmcp__flow-nexus__neural_node_deploy({\n  cluster_id: \"cluster_xyz789\",\n  node_type: \"parameter_server\",\n  model: \"large\",\n  template: \"nodejs\",\n  capabilities: [\"parameter_management\", \"gradient_aggregation\"],\n  autonomy: 0.8\n})\n\n\u002F\u002F Deploy worker nodes\nmcp__flow-nexus__neural_node_deploy({\n  cluster_id: \"cluster_xyz789\",\n  node_type: \"worker\",\n  model: \"xl\",\n  role: \"worker\",\n  capabilities: [\"training\", \"inference\"],\n  layers: [\n    { type: \"transformer_encoder\", num_heads: 16 },\n    { type: \"feed_forward\", units: 4096 }\n  ],\n  autonomy: 0.9\n})\n\n\u002F\u002F Deploy aggregator\nmcp__flow-nexus__neural_node_deploy({\n  cluster_id: \"cluster_xyz789\",\n  node_type: \"aggregator\",\n  model: \"large\",\n  capabilities: [\"gradient_aggregation\", \"model_synchronization\"]\n})\n```\n\n#### Connect Cluster Topology\n\n```javascript\nmcp__flow-nexus__neural_cluster_connect({\n  cluster_id: \"cluster_xyz789\",\n  topology: \"mesh\" \u002F\u002F Override default if needed\n})\n```\n\n#### Start Distributed Training\n\n```javascript\nmcp__flow-nexus__neural_train_distributed({\n  cluster_id: \"cluster_xyz789\",\n  dataset: \"imagenet\", \u002F\u002F or custom dataset identifier\n  epochs: 100,\n  batch_size: 128,\n  learning_rate: 0.001,\n  optimizer: \"adam\", \u002F\u002F sgd, rmsprop, adagrad\n  federated: true \u002F\u002F Enable federated learning\n})\n```\n\n**Federated Learning Example:**\n```javascript\nmcp__flow-nexus__neural_train_distributed({\n  cluster_id: \"cluster_xyz789\",\n  dataset: \"medical_images_distributed\",\n  epochs: 200,\n  batch_size: 64,\n  learning_rate: 0.0001,\n  optimizer: \"adam\",\n  federated: true, \u002F\u002F Data stays on local nodes\n  aggregation_rounds: 50,\n  min_nodes_per_round: 5\n})\n```\n\n#### Monitor Cluster Status\n\n```javascript\nmcp__flow-nexus__neural_cluster_status({\n  cluster_id: \"cluster_xyz789\"\n})\n```\n\n**Response:**\n```json\n{\n  \"cluster_id\": \"cluster_xyz789\",\n  \"status\": \"training\",\n  \"nodes\": [\n    {\n      \"node_id\": \"node_001\",\n      \"type\": \"parameter_server\",\n      \"status\": \"active\",\n      \"cpu_usage\": 0.75,\n      \"memory_usage\": 0.82\n    },\n    {\n      \"node_id\": \"node_002\",\n      \"type\": \"worker\",\n      \"status\": \"active\",\n      \"training_progress\": 0.45\n    }\n  ],\n  \"training_metrics\": {\n    \"current_epoch\": 45,\n    \"total_epochs\": 100,\n    \"loss\": 0.234,\n    \"accuracy\": 0.891\n  }\n}\n```\n\n#### Run Distributed Inference\n\n```javascript\nmcp__flow-nexus__neural_predict_distributed({\n  cluster_id: \"cluster_xyz789\",\n  input_data: JSON.stringify([\n    [0.1, 0.2, 0.3],\n    [0.4, 0.5, 0.6]\n  ]),\n  aggregation: \"ensemble\" \u002F\u002F mean, majority, weighted, ensemble\n})\n```\n\n#### Terminate Cluster\n\n```javascript\nmcp__flow-nexus__neural_cluster_terminate({\n  cluster_id: \"cluster_xyz789\"\n})\n```\n\n### 5. Model Management\n\n#### List Your Models\n\n```javascript\nmcp__flow-nexus__neural_list_models({\n  user_id: \"your_user_id\",\n  include_public: true\n})\n```\n\n**Response:**\n```json\n{\n  \"models\": [\n    {\n      \"model_id\": \"model_abc123\",\n      \"name\": \"Custom Classifier v1\",\n      \"architecture\": \"feedforward\",\n      \"accuracy\": 0.92,\n      \"created_at\": \"2025-10-15T14:20:00Z\",\n      \"status\": \"trained\"\n    },\n    {\n      \"model_id\": \"model_def456\",\n      \"name\": \"LSTM Forecaster\",\n      \"architecture\": \"lstm\",\n      \"mse\": 0.0045,\n      \"created_at\": \"2025-10-18T09:15:00Z\",\n      \"status\": \"training\"\n    }\n  ]\n}\n```\n\n#### Check Training Status\n\n```javascript\nmcp__flow-nexus__neural_training_status({\n  job_id: \"job_training_xyz\"\n})\n```\n\n**Response:**\n```json\n{\n  \"job_id\": \"job_training_xyz\",\n  \"status\": \"training\",\n  \"progress\": 0.67,\n  \"current_epoch\": 67,\n  \"total_epochs\": 100,\n  \"current_loss\": 0.234,\n  \"estimated_completion\": \"2025-10-19T12:45:00Z\"\n}\n```\n\n#### Performance Benchmarking\n\n```javascript\nmcp__flow-nexus__neural_performance_benchmark({\n  model_id: \"model_abc123\",\n  benchmark_type: \"comprehensive\" \u002F\u002F inference, throughput, memory, comprehensive\n})\n```\n\n**Response:**\n```json\n{\n  \"model_id\": \"model_abc123\",\n  \"benchmarks\": {\n    \"inference_latency_ms\": 12.5,\n    \"throughput_qps\": 8000,\n    \"memory_usage_mb\": 245,\n    \"gpu_utilization\": 0.78,\n    \"accuracy\": 0.92,\n    \"f1_score\": 0.89\n  },\n  \"timestamp\": \"2025-10-19T11:00:00Z\"\n}\n```\n\n#### Create Validation Workflow\n\n```javascript\nmcp__flow-nexus__neural_validation_workflow({\n  model_id: \"model_abc123\",\n  user_id: \"your_user_id\",\n  validation_type: \"comprehensive\" \u002F\u002F performance, accuracy, robustness, comprehensive\n})\n```\n\n### 6. Publishing and Marketplace\n\n#### Publish Model as Template\n\n```javascript\nmcp__flow-nexus__neural_publish_template({\n  model_id: \"model_abc123\",\n  name: \"High-Accuracy Sentiment Classifier\",\n  description: \"Fine-tuned BERT model for sentiment analysis with 94% accuracy\",\n  category: \"nlp\",\n  price: 0, \u002F\u002F 0 for free, or credits amount\n  user_id: \"your_user_id\"\n})\n```\n\n#### Rate a Template\n\n```javascript\nmcp__flow-nexus__neural_rate_template({\n  template_id: \"sentiment-analysis-v2\",\n  rating: 5,\n  review: \"Excellent model! Achieved 95% accuracy on my dataset.\",\n  user_id: \"your_user_id\"\n})\n```\n\n## Common Use Cases\n\n### Image Classification with CNN\n\n```javascript\n\u002F\u002F Initialize cluster for large-scale image training\nconst cluster = await mcp__flow-nexus__neural_cluster_init({\n  name: \"image-classification-cluster\",\n  architecture: \"cnn\",\n  topology: \"hierarchical\",\n  wasmOptimization: true\n})\n\n\u002F\u002F Deploy worker nodes\nawait mcp__flow-nexus__neural_node_deploy({\n  cluster_id: cluster.cluster_id,\n  node_type: \"worker\",\n  model: \"large\",\n  capabilities: [\"training\", \"data_augmentation\"]\n})\n\n\u002F\u002F Start training\nawait mcp__flow-nexus__neural_train_distributed({\n  cluster_id: cluster.cluster_id,\n  dataset: \"custom_images\",\n  epochs: 100,\n  batch_size: 64,\n  learning_rate: 0.001,\n  optimizer: \"adam\"\n})\n```\n\n### NLP Sentiment Analysis\n\n```javascript\n\u002F\u002F Use pre-built template\nconst deployment = await mcp__flow-nexus__neural_deploy_template({\n  template_id: \"sentiment-analysis-v2\",\n  custom_config: {\n    training: {\n      epochs: 30,\n      batch_size: 16\n    }\n  }\n})\n\n\u002F\u002F Run inference\nconst result = await mcp__flow-nexus__neural_predict({\n  model_id: deployment.model_id,\n  input: [\"This product is amazing!\", \"Terrible experience.\"]\n})\n```\n\n### Time Series Forecasting\n\n```javascript\n\u002F\u002F Train LSTM model\nconst training = await mcp__flow-nexus__neural_train({\n  config: {\n    architecture: {\n      type: \"lstm\",\n      layers: [\n        { type: \"lstm\", units: 128, return_sequences: true },\n        { type: \"dropout\", rate: 0.2 },\n        { type: \"lstm\", units: 64 },\n        { type: \"dense\", units: 1 }\n      ]\n    },\n    training: {\n      epochs: 150,\n      batch_size: 64,\n      learning_rate: 0.01,\n      optimizer: \"adam\"\n    }\n  },\n  tier: \"medium\"\n})\n\n\u002F\u002F Monitor progress\nconst status = await mcp__flow-nexus__neural_training_status({\n  job_id: training.job_id\n})\n```\n\n### Federated Learning for Privacy\n\n```javascript\n\u002F\u002F Initialize federated cluster\nconst cluster = await mcp__flow-nexus__neural_cluster_init({\n  name: \"federated-medical-cluster\",\n  architecture: \"transformer\",\n  topology: \"mesh\",\n  consensus: \"proof-of-learning\",\n  daaEnabled: true\n})\n\n\u002F\u002F Deploy nodes across different locations\nfor (let i = 0; i \u003C 5; i++) {\n  await mcp__flow-nexus__neural_node_deploy({\n    cluster_id: cluster.cluster_id,\n    node_type: \"worker\",\n    model: \"large\",\n    autonomy: 0.9\n  })\n}\n\n\u002F\u002F Train with federated learning (data never leaves nodes)\nawait mcp__flow-nexus__neural_train_distributed({\n  cluster_id: cluster.cluster_id,\n  dataset: \"medical_records_distributed\",\n  epochs: 200,\n  federated: true,\n  aggregation_rounds: 100\n})\n```\n\n## Architecture Patterns\n\n### Feedforward Networks\nBest for: Classification, regression, simple pattern recognition\n```javascript\n{\n  type: \"feedforward\",\n  layers: [\n    { type: \"dense\", units: 256, activation: \"relu\" },\n    { type: \"dropout\", rate: 0.3 },\n    { type: \"dense\", units: 128, activation: \"relu\" },\n    { type: \"dense\", units: 10, activation: \"softmax\" }\n  ]\n}\n```\n\n### LSTM Networks\nBest for: Time series, sequences, forecasting\n```javascript\n{\n  type: \"lstm\",\n  layers: [\n    { type: \"lstm\", units: 128, return_sequences: true },\n    { type: \"lstm\", units: 64 },\n    { type: \"dense\", units: 1 }\n  ]\n}\n```\n\n### Transformers\nBest for: NLP, attention mechanisms, large-scale text\n```javascript\n{\n  type: \"transformer\",\n  layers: [\n    { type: \"embedding\", vocab_size: 10000, embedding_dim: 512 },\n    { type: \"transformer_encoder\", num_heads: 8, ff_dim: 2048 },\n    { type: \"global_average_pooling\" },\n    { type: \"dense\", units: 2, activation: \"softmax\" }\n  ]\n}\n```\n\n### GANs\nBest for: Generative tasks, image synthesis\n```javascript\n{\n  type: \"gan\",\n  generator_layers: [...],\n  discriminator_layers: [...]\n}\n```\n\n### Autoencoders\nBest for: Dimensionality reduction, anomaly detection\n```javascript\n{\n  type: \"autoencoder\",\n  encoder_layers: [\n    { type: \"dense\", units: 128, activation: \"relu\" },\n    { type: \"dense\", units: 64, activation: \"relu\" }\n  ],\n  decoder_layers: [\n    { type: \"dense\", units: 128, activation: \"relu\" },\n    { type: \"dense\", units: input_dim, activation: \"sigmoid\" }\n  ]\n}\n```\n\n## Best Practices\n\n1. **Start Small**: Begin with `nano` or `mini` tiers for experimentation\n2. **Use Templates**: Leverage marketplace templates for common tasks\n3. **Monitor Training**: Check status regularly to catch issues early\n4. **Benchmark Models**: Always benchmark before production deployment\n5. **Distributed Training**: Use clusters for large models (>1B parameters)\n6. **Federated Learning**: Use for privacy-sensitive data\n7. **Version Models**: Publish successful models as templates for reuse\n8. **Validate Thoroughly**: Use validation workflows before deployment\n\n## Troubleshooting\n\n### Training Stalled\n```javascript\n\u002F\u002F Check cluster status\nconst status = await mcp__flow-nexus__neural_cluster_status({\n  cluster_id: \"cluster_id\"\n})\n\n\u002F\u002F Terminate and restart if needed\nawait mcp__flow-nexus__neural_cluster_terminate({\n  cluster_id: \"cluster_id\"\n})\n```\n\n### Low Accuracy\n- Increase epochs\n- Adjust learning rate\n- Add regularization (dropout)\n- Try different optimizer\n- Use data augmentation\n\n### Out of Memory\n- Reduce batch size\n- Use smaller model tier\n- Enable gradient accumulation\n- Use distributed training\n\n## Related Skills\n\n- `flow-nexus-sandbox` - E2B sandbox management\n- `flow-nexus-swarm` - AI swarm orchestration\n- `flow-nexus-workflow` - Workflow automation\n\n## Resources\n\n- Flow Nexus Docs: https:\u002F\u002Fflow-nexus.ruv.io\u002Fdocs\n- Neural Network Guide: https:\u002F\u002Fflow-nexus.ruv.io\u002Fdocs\u002Fneural\n- Template Marketplace: https:\u002F\u002Fflow-nexus.ruv.io\u002Ftemplates\n- API Reference: https:\u002F\u002Fflow-nexus.ruv.io\u002Fapi\n\n---\n\n**Note**: Distributed training requires authentication. Register at https:\u002F\u002Fflow-nexus.ruv.io or use `npx flow-nexus@latest register`.\n",{"name":5,"tags":49,"version":50,"category":51,"mcp_server":31,"description":7,"requires_auth":52},[36,32,40,28,31,30],"1.0.0","ai-ml",true,"main","def8531e65114c0fca8fb8551c1871ee0eed705c","skills\u002Faiskillstore\u002Fmarketplace\u002Fruvnet\u002Fflow-nexus-neural",24,"https:\u002F\u002Fgithub.com\u002FComeOnOliver\u002Fskillshub","https:\u002F\u002Fskillshub.wtf","https:\u002F\u002Fgithub.com\u002FComeOnOliver","AI & Blockchain Engineer\r\n","2026-06-24 22:07:19",{"source":8,"skillName":5,"commands":63,"prompt":73,"zipAvailable":52},[64,67,70],{"tool":65,"command":66},"npx","npx skills add ComeOnOliver\u002Fskillshub --skill flow-nexus-neural",{"tool":68,"command":69},"bunx","bunx skills add ComeOnOliver\u002Fskillshub --skill flow-nexus-neural",{"tool":71,"command":72},"pnpm","pnpm dlx skills add ComeOnOliver\u002Fskillshub --skill flow-nexus-neural","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FComeOnOliver\u002Fskillshub\u002Fblob\u002Fmain\u002Fskills\u002Faiskillstore\u002Fmarketplace\u002Fruvnet\u002Fflow-nexus-neural\u002FSKILL.md",[75],{"path":76,"resourceType":77,"resourceTypeDesc":78,"sizeBytes":79,"sourceHtmlUrl":80},"skill-report.json","REFERENCE","参考文档",21471,"https:\u002F\u002Fgithub.com\u002FComeOnOliver\u002Fskillshub\u002Fblob\u002Fmain\u002Fskills\u002Faiskillstore\u002Fmarketplace\u002Fruvnet\u002Fflow-nexus-neural\u002Fskill-report.json",[82,94,104,113,119,126],{"skillNo":83,"name":84,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":85,"categories":86,"keywords":90,"publishedAt":44},"01M6FKJN13GMG4KY57J3QD8GY5","model-hierarchy-skill","https:\u002F\u002Fgithub.com\u002FComeOnOliver\u002Fskillshub\u002Fblob\u002Fmain\u002Fskills\u002Fzscole\u002Fmodel-hierarchy-skill\u002Fmodel-hierarchy-skill\u002FSKILL.md",[87],{"categoryNo":88,"categoryName":89,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发",[91,92,93],"model","model hierarchy","model hierarchy skill",{"skillNo":95,"name":96,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":97,"categories":98,"keywords":100,"publishedAt":44},"01M6FKJJJ6GCAMR4PPPRD12M5X","zoom-mcp","https:\u002F\u002Fgithub.com\u002FComeOnOliver\u002Fskillshub\u002Fblob\u002Fmain\u002Fskills\u002Fzoom\u002Fskills\u002Fzoom-mcp\u002FSKILL.md",[99],{"categoryNo":88,"categoryName":89,"taxonomyType":21,"taxonomyTypeDesc":22},[101,102,103],"MCP","zoom","zoom mcp",{"skillNo":105,"name":106,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":107,"categories":108,"keywords":109,"publishedAt":44},"01M6FKJE3KYT3KPEZMEEHYKA6K","zoom-apps-sdk","https:\u002F\u002Fgithub.com\u002FComeOnOliver\u002Fskillshub\u002Fblob\u002Fmain\u002Fskills\u002Fzoom\u002Fskills\u002Fzoom-apps-sdk\u002FSKILL.md",[],[110,111,112,102],"sdk","zoom apps sdk","apps",{"skillNo":114,"name":115,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":116,"categories":117,"keywords":118,"publishedAt":44},"01M6FKJ9EDH9Q5VWDF8HAJRSP0","windows","https:\u002F\u002Fgithub.com\u002FComeOnOliver\u002Fskillshub\u002Fblob\u002Fmain\u002Fskills\u002Fzoom\u002Fskills\u002Fwindows\u002FSKILL.md",[],[115],{"skillNo":120,"name":121,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":122,"categories":123,"keywords":125,"publishedAt":44},"01M6FKJ666BEEFC3AE9GJ72Q93","whiteboard","https:\u002F\u002Fgithub.com\u002FComeOnOliver\u002Fskillshub\u002Fblob\u002Fmain\u002Fskills\u002Fzoom\u002Fskills\u002Fwhiteboard\u002FSKILL.md",[124],{"categoryNo":88,"categoryName":89,"taxonomyType":21,"taxonomyTypeDesc":22},[121],{"skillNo":127,"name":128,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":129,"categories":130,"keywords":131,"publishedAt":44},"01M6FKJ21WFCDGTJQ2Q91T9DVW","websockets","https:\u002F\u002Fgithub.com\u002FComeOnOliver\u002Fskillshub\u002Fblob\u002Fmain\u002Fskills\u002Fzoom\u002Fskills\u002Fwebsockets\u002FSKILL.md",[],[128]]