{"id":46076066,"url":"https://github.com/ry-ops/k3s-mcp-server","last_synced_at":"2026-03-01T14:35:06.706Z","repository":{"id":329170066,"uuid":"1115771020","full_name":"ry-ops/k3s-mcp-server","owner":"ry-ops","description":"Model Context Protocol server for K3s cluster management - kubectl operations for Claude","archived":false,"fork":false,"pushed_at":"2026-02-03T01:10:26.000Z","size":77,"stargazers_count":1,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-02-06T00:10:02.376Z","etag":null,"topics":["ai","claude","k3s","kubernetes","mcp","mcp-server","python"],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/ry-ops.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2025-12-13T14:18:27.000Z","updated_at":"2026-02-03T15:47:19.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/ry-ops/k3s-mcp-server","commit_stats":null,"previous_names":["ry-ops/k3s-mcp-server"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/ry-ops/k3s-mcp-server","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ry-ops%2Fk3s-mcp-server","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ry-ops%2Fk3s-mcp-server/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ry-ops%2Fk3s-mcp-server/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ry-ops%2Fk3s-mcp-server/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/ry-ops","download_url":"https://codeload.github.com/ry-ops/k3s-mcp-server/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ry-ops%2Fk3s-mcp-server/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":29971001,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-03-01T14:11:48.712Z","status":"ssl_error","status_checked_at":"2026-03-01T14:11:48.352Z","response_time":124,"last_error":"SSL_connect returned=1 errno=0 peeraddr=140.82.121.5:443 state=error: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["ai","claude","k3s","kubernetes","mcp","mcp-server","python"],"created_at":"2026-03-01T14:35:06.035Z","updated_at":"2026-03-01T14:35:06.688Z","avatar_url":"https://github.com/ry-ops.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# K3s MCP Server\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"assets/cortex-fabric.svg\" alt=\"Cortex Platform - AI-Native K3s Infrastructure\" width=\"700\"/\u003e\n\u003c/p\u003e\n\n\u003cp align=\"center\"\u003e\n  \u003ca href=\"https://www.python.org/downloads/\"\u003e\u003cimg src=\"https://img.shields.io/badge/python-3.10+-blue.svg\" alt=\"Python\"\u003e\u003c/a\u003e\n  \u003ca href=\"https://github.com/astral-sh/uv\"\u003e\u003cimg src=\"https://img.shields.io/badge/uv-latest-green.svg\" alt=\"uv\"\u003e\u003c/a\u003e\n  \u003ca href=\"https://modelcontextprotocol.io/\"\u003e\u003cimg src=\"https://img.shields.io/badge/MCP-1.0-purple.svg\" alt=\"MCP\"\u003e\u003c/a\u003e\n  \u003ca href=\"https://k3s.io/\"\u003e\u003cimg src=\"https://img.shields.io/badge/K3s-Adopter-orange.svg\" alt=\"K3s\"\u003e\u003c/a\u003e\n  \u003ca href=\"LICENSE\"\u003e\u003cimg src=\"https://img.shields.io/badge/License-MIT-yellow.svg\" alt=\"License: MIT\"\u003e\u003c/a\u003e\n\u003c/p\u003e\n\nA Model Context Protocol (MCP) server for managing Kubernetes (K3s) clusters. This server enables AI assistants like Claude to manage K3s clusters through natural language, and serves as the foundation for **Cortex Platform** - an AI-native infrastructure orchestration system.\n\n## Cortex Platform: AI-Native Infrastructure on K3s\n\nThis project is part of the **Cortex Platform**, a production system that demonstrates advanced K3s usage patterns:\n\n- **7-Layer Serverless Fabric** with KEDA auto-scaling (0→1 pods on demand)\n- **AI-powered query routing** with multi-tier classification\n- **Self-healing infrastructure** through MCP-based automation\n- **Vector memory (Qdrant)** for learning from operational patterns\n- **Dynamic worker pools** managed by AI agents\n\n### Architecture Overview\n\n```\n┌─────────────────────────────────────────────────────────────────────────────┐\n│                           CORTEX PLATFORM                                   │\n│                        AI-Native K3s Infrastructure                         │\n├─────────────────────────────────────────────────────────────────────────────┤\n│                                                                             │\n│   ┌─────────────────────────────────────────────────────────────────────┐   │\n│   │                    K3s CLUSTER (7 nodes)                            │   │\n│   │                    Talos Linux + etcd HA                            │   │\n│   │                                                                     │   │\n│   │   ┌─────────────────────────────────────────────────────────────┐   │   │\n│   │   │              UNIFI LAYER FABRIC                             │   │   │\n│   │   │         Serverless AI Network Operations                    │   │   │\n│   │   │                                                             │   │   │\n│   │   │   USER QUERY                                                │   │   │\n│   │   │       │                                                     │   │   │\n│   │   │       ▼                                                     │   │   │\n│   │   │   ┌───────────────────┐    ┌─────────────────┐              │   │   │\n│   │   │   │ CORTEX ACTIVATOR  │───▶│  CORTEX QDRANT  │              │   │   │\n│   │   │   │   (Always On)     │    │  (Always On)    │              │   │   │\n│   │   │   │   Query Router    │    │  Vector Memory  │              │   │   │\n│   │   │   │   128MB, 2 pods   │    │  512MB, 5Gi PVC │              │   │   │\n│   │   │   └─────────┬─────────┘    └─────────────────┘              │   │   │\n│   │   │             │                                               │   │   │\n│   │   │             │ 4-Tier Routing Cascade                        │   │   │\n│   │   │             │ 2. Similarity Search (\u003c50ms)                  │   │   │\n│   │   │             │ 3. Lightweight Classifier (~5s cold)          │   │   │\n│   │   │             │ 4. Full SLM Reasoning (~12s cold)             │   │   │\n│   │   │             ▼                                               │   │   │\n│   │   │   ┌─────────────────────────────────────────────────────┐   │   │   │\n│   │   │   │            REASONING LAYERS (Scale 0→1)             │   │   │   │\n│   │   │   │                                                     │   │   │   │\n│   │   │   │   ┌──────────────────┐  ┌──────────────────────┐    │   │   │   │\n│   │   │   │   │ reasoning-       │  │ reasoning-slm        │    │   │   │   │\n│   │   │   │   │ classifier       │  │                      │    │   │   │   │\n│   │   │   │   │ Qwen2 0.5B       │  │ Phi-3 3.8B           │    │   │   │   │\n│   │   │   │   │ ~5s cold start   │  │ ~12s cold start      │    │   │   │   │\n│   │   │   │   │ 400MB warm       │  │ 2.5GB warm           │    │   │   │   │\n│   │   │   │   └──────────────────┘  └──────────────────────┘    │   │   │   │\n│   │   │   └─────────────────────────────────────────────────────┘   │   │   │\n│   │   │             │                                               │   │   │\n│   │   │             ▼                                               │   │   │\n│   │   │   ┌─────────────────────────────────────────────────────┐   │   │   │\n│   │   │   │            EXECUTION LAYERS (Scale 0→1)             │   │   │   │\n│   │   │   │                                                     │   │   │   │\n│   │   │   │   ┌──────────────────┐  ┌──────────────────────┐    │   │   │   │\n│   │   │   │   │ execution-       │  │ execution-           │    │   │   │   │\n│   │   │   │   │ execution-       │  │ execution-           │    │   │   │   │\n│   │   │   │   │ execution-       │  │ execution-           │    │   │   │   │\n│   │   │   │   │ unifi-api        │  │ unifi-ssh            │    │   │   │   │\n│   │   │   │   │ unifi-api        │  │ unifi-ssh            │    │   │   │   │\n│   │   │   │   │ Primary          │  │ Failover             │    │   │   │   │\n│   │   │   │   │ ~3s cold start   │  │ ~3s cold start       │    │   │   │   │\n│   │   │   │   └──────────────────┘  └──────────────────────┘    │   │   │   │\n│   │   │   └─────────────────────────────────────────────────────┘   │   │   │\n│   │   │             │                                               │   │   │\n│   │   │             ▼                                               │   │   │\n│   │   │   ┌───────────────────┐                                     │   │   │\n│   │   │   │ CORTEX TELEMETRY  │                                     │   │   │\n│   │   │   │ Metrics + Learning│                                     │   │   │\n│   │   │   │ Scale 0→1         │                                     │   │   │\n│   │   │   └───────────────────┘                                     │   │   │  \n│   │   └─────────────────────────────────────────────────────────────┘   │   │\n│   │                                                                     │   │\n│   │   ┌─────────────────────────────────────────────────────────────┐   │   │\n│   │   │                    MCP SERVERS                              │   │   │\n│   │   │                                                             │   │   │\n│   │   │   ┌──────────────┐  ┌──────────────┐  ┌──────────────┐      │   │   │\n│   │   │   │ k3s-mcp      │  │ talos-mcp    │  │ proxmox-mcp  │      │   │   │\n│   │   │   │ (this repo)  │  │              │  │              │      │   │   │\n│   │   │   │ Cluster ops  │  │ Node mgmt    │  │ VM lifecycle │      │   │   │\n│   │   │   └──────────────┘  └──────────────┘  └──────────────┘      │   │   │\n│   │   └─────────────────────────────────────────────────────────────┘   │   │\n│   │                                                                     │   │\n│   │   ┌─────────────────────────────────────────────────────────────┐   │   │\n│   │   │              DYNAMIC WORKER POOLS                           │   │   │\n│   │   │         (Managed by AI Resource Manager)                    │   │   │\n│   │   │                                                             │   │   │\n│   │   │   Permanent: 3-10 nodes (always running)                    │   │   │\n│   │   │   Burst:     0-20 nodes (TTL-based cleanup)                 │   │   │\n│   │   │   Spot:      0-15 nodes (70% cost savings)                  │   │   │\n│   │   │   GPU:       0-5 nodes  (special hardware taints)           │   │   │\n│   │   └─────────────────────────────────────────────────────────────┘   │   │\n│   └─────────────────────────────────────────────────────────────────────┘   │\n│                                                                             │\n│   Memory Profile:                                                           │\n│   • Idle:         640MB  (Activator + Qdrant only)                          │\n│   • Simple query: 1GB    (+ execution layer)                                │\n│   • Complex:      4GB    (+ SLM reasoning)                                  │\n│   • Savings:      85%+   vs always-on architecture                          │\n│                                                                             │\n└─────────────────────────────────────────────────────────────────────────────┘\n```\n\n### The 7 Layers\n\n| Layer | Component | Purpose | Memory | Cold Start | Scale |\n|-------|-----------|---------|--------|------------|-------|\n| 1 | **cortex-activator** | Query routing \u0026 orchestration | 128MB | Always on | 2 replicas |\n| 2 | **cortex-qdrant** | Vector memory \u0026 RAG | 512MB | Always on | 1 replica |\n| 3 | **reasoning-classifier** | Fast intent classification | 400MB | ~5s | 0→1 |\n| 4 | **reasoning-slm** | Full reasoning (Phi-3) | 2.5GB | ~12s | 0→1 |\n| 5 | **execution-unifi-api** | Primary API operations | 200MB | ~3s | 0→2 |\n| 6 | **execution-unifi-ssh** | Failover \u0026 diagnostics | 100MB | ~3s | 0→1 |\n| 7 | **cortex-telemetry** | Metrics \u0026 learning pipeline | 128MB | ~2s | 0→1 |\n\n### Cortex Activator: Intelligent Query Router\n\nThe Cortex Activator is the brain of the system - a lightweight service that routes queries through a 4-tier cascade:\n\n```\nQuery: \"Block the client with MAC aa:bb:cc:dd:ee:ff\"\n         │\n         ▼\n┌─────────────────────────────────────────────────────────────┐\n│ TIER 1: Keyword Pattern Match (\u003c10ms)                       │\n│                                                             │\n│ Pattern: \"(block|unblock).*client\" → MATCH                  │\n│ Confidence: 95%                                             │\n│ Action: Route directly to execution-unifi-api               │\n└─────────────────────────────────────────────────────────────┘\n         │\n         │ (If no match, continue to Tier 2)\n         ▼\n┌─────────────────────────────────────────────────────────────┐\n│ TIER 2: Qdrant Similarity Search (\u003c50ms)                    │\n│                                                             │\n│ Query embedding → Search past successful routes             │\n│ If similar query succeeded before → Reuse routing           │\n│ Learning: Skip expensive LLM classification                 │\n└─────────────────────────────────────────────────────────────┘\n         │\n         │ (If similarity \u003c 92%, continue to Tier 3)\n         ▼\n┌─────────────────────────────────────────────────────────────┐\n│ TIER 3: Lightweight Classifier (~5s cold start)             │\n│                                                             │\n│ Model: Qwen2-0.5B (quantized)                               │\n│ Use: Ambiguous queries needing quick classification         │\n│ KEDA: Scales from 0→1 on demand                             │\n└─────────────────────────────────────────────────────────────┘\n         │\n         │ (If complex investigation needed, continue to Tier 4)\n         ▼\n┌─────────────────────────────────────────────────────────────┐\n│ TIER 4: Full SLM Reasoning (~12s cold start)                │\n│                                                             │\n│ Model: Phi-3-mini-4k-instruct (3.8B, quantized)             │\n│ Use: Multi-step reasoning, complex troubleshooting          │\n│ KEDA: Scales from 0→1 on demand                             │\n└─────────────────────────────────────────────────────────────┘\n```\n\n### KEDA Serverless Scaling\n\nLayers scale from 0 to 1 based on Prometheus metrics:\n\n```yaml\n# reasoning-slm KEDA configuration\nkeda:\n  minReplicaCount: 0      # Scale to zero when idle\n  maxReplicaCount: 1\n  cooldownPeriod: 300     # Scale down after 5 min idle\n  trigger:\n    type: prometheus\n    query: sum(cortex_activator_pending_requests{layer=\"reasoning-slm\"})\n    threshold: \"1\"        # Wake if ANY pending request\n```\n\n**Activation flow:**\n1. Query arrives → Activator increments `pending_requests` gauge\n2. KEDA detects metric \u003e threshold → Scales deployment 0→1\n3. Pod starts → Health probe passes → Pod ready\n4. Request processed → Response sent\n5. 5 minutes idle → Cooldown triggers → Scales back to 0\n\n### Adaptive Intelligence (Phase 4)\n\nQuery complexity scoring (0-100) determines execution mode:\n\n| Complexity | Score | Mode | Resources |\n|------------|-------|------|-----------|\n| SIMPLE | 0-25 | Direct execution | Activator only |\n| MODERATE | 26-50 | Basic classification | + Classifier |\n| COMPLEX | 51-75 | Full reasoning | + SLM |\n| EXPERT | 76-100 | Escalation | Human review |\n\n**Auto-escalation triggers:**\n- Low confidence (\u003c50%) → Escalate mode\n- Previous similar queries failed → Escalate\n- Timeout (\u003e30s agent, \u003e60s hybrid) → Escalate\n\n---\n\n## K3s MCP Server Features\n\nThis MCP server provides the foundation for AI-driven cluster management:\n\n### Pod Management\n- List pods across namespaces with label selectors\n- Get pod logs with tail and container selection\n- Execute commands in pods\n- Restart pods (delete and recreate)\n\n### Deployment Management\n- List and describe deployments\n- Scale deployments up or down\n- Get deployment status and replica counts\n\n### Service Management\n- List services and endpoints\n- View service ports and selectors\n- Check service types (ClusterIP, NodePort, LoadBalancer)\n\n### Node Management\n- List all cluster nodes\n- Get node status and resources\n- View node capacity and allocatable resources\n- Check node conditions (Ready, MemoryPressure, etc.)\n\n### Resource Management\n- Apply YAML manifests\n- Delete resources (pods, deployments, services)\n- List namespaces\n- Get cluster information\n\n## Quick Start\n\n```bash\n# 1. Install uv (if not already installed)\ncurl -LsSf https://astral.sh/uv/install.sh | sh\n\n# 2. Clone this repository\ngit clone https://github.com/ry-ops/k3s-mcp-server.git\ncd k3s-mcp-server\n\n# 3. Run setup script\nchmod +x setup.sh\n./setup.sh\n\n# 4. Set environment variables\nexport KUBECONFIG=\"$HOME/.kube/config\"\n\n# 5. Test the server\nuv run k3s-mcp-server\n\n# 6. Configure Claude Desktop and restart\n```\n\nSee [QUICKSTART.md](QUICKSTART.md) for detailed instructions.\n\n## Installation\n\n### Prerequisites\n\n- Python 3.10 or higher\n- `uv` package manager\n- K3s cluster with kubeconfig access\n- Kubeconfig file\n\n### Setup\n\n```bash\n# Install uv\ncurl -LsSf https://astral.sh/uv/install.sh | sh\n\n# Clone this repository\ngit clone https://github.com/ry-ops/k3s-mcp-server.git\ncd k3s-mcp-server\n\n# Run setup script (creates structure and installs dependencies)\n./setup.sh\n\n# Or manually:\nuv sync\n```\n\n## Configuration\n\n### Environment Variables\n\n| Variable | Description | Default |\n|----------|-------------|---------|\n| `KUBECONFIG` | Path to kubeconfig file | `~/.kube/config` |\n| `K3S_DEFAULT_NAMESPACE` | Default namespace | `default` |\n| `K3S_DEBUG` | Enable debug logging | `false` |\n\n### Claude Desktop Configuration\n\n**MacOS**: `~/Library/Application Support/Claude/claude_desktop_config.json`\n**Windows**: `%APPDATA%/Claude/claude_desktop_config.json`\n\n```json\n{\n  \"mcpServers\": {\n    \"k3s\": {\n      \"command\": \"uv\",\n      \"args\": [\n        \"--directory\",\n        \"/absolute/path/to/k3s-mcp-server\",\n        \"run\",\n        \"k3s-mcp-server\"\n      ],\n      \"env\": {\n        \"KUBECONFIG\": \"/path/to/.kube/config\"\n      }\n    }\n  }\n}\n```\n\n## Available Tools\n\n### Pod Tools\n\n| Tool | Description | Parameters |\n|------|-------------|------------|\n| `get_pods` | List pods in namespace or cluster-wide | `namespace`, `labels` |\n| `get_logs` | Get pod logs | `pod_name`, `namespace`, `container`, `tail_lines` |\n| `restart_pod` | Restart a pod by deleting it | `name`, `namespace` |\n| `execute_command` | Execute command in pod | `pod_name`, `namespace`, `command`, `container` |\n\n### Deployment Tools\n\n| Tool | Description | Parameters |\n|------|-------------|------------|\n| `get_deployments` | List all deployments | `namespace` |\n| `get_deployment` | Get specific deployment details | `name`, `namespace` |\n| `scale_deployment` | Scale deployment replicas | `name`, `namespace`, `replicas` |\n\n### Cluster Tools\n\n| Tool | Description | Parameters |\n|------|-------------|------------|\n| `get_nodes` | List all nodes with resources | - |\n| `get_services` | List all services | `namespace` |\n| `get_cluster_info` | Get cluster version and summary | - |\n| `get_namespaces` | List all namespaces | - |\n| `apply_manifest` | Apply YAML manifest | `manifest_yaml`, `namespace` |\n| `delete_resource` | Delete a resource | `kind`, `name`, `namespace` |\n\n## Example Usage\n\nOnce configured, ask Claude to interact with your K3s cluster:\n\n```\n\"List all pods in the cortex-system namespace\"\n\n\"What's the status of the cortex-activator deployment?\"\n\n\"Scale reasoning-slm to 1 replica\"\n\n\"Show me logs from the cortex-qdrant pod\"\n\n\"Which nodes are in my cluster and what's their capacity?\"\n\n\"Apply this deployment manifest: [YAML content]\"\n```\n\n## Project Structure\n\n```\nk3s-mcp-server/\n├── src/\n│   └── k3s_mcp_server/\n│       ├── __init__.py       # Package initialization\n│       └── server.py         # Main server implementation\n├── docs/\n│   ├── ARCHITECTURE.md       # Detailed architecture docs\n│   └── CORTEX_INTEGRATION.md # Cortex Platform integration\n├── pyproject.toml            # Project configuration\n├── uv.lock                   # Locked dependencies\n├── setup.sh                  # Automated setup script\n├── README.md                 # This file\n└── QUICKSTART.md             # Quick setup guide\n```\n\n## K3s Features Used\n\nThis project demonstrates production usage of K3s features:\n\n| Feature | Usage |\n|---------|-------|\n| **Namespaces** | Multi-tenant isolation (cortex-system, cortex-mcp, cortex-unifi) |\n| **Deployments** | All workloads with rolling updates |\n| **KEDA** | Serverless 0→1 scaling for reasoning/execution layers |\n| **Helm** | Package management for all components |\n| **RBAC** | Fine-grained service account permissions |\n| **PVCs** | Persistent storage for Qdrant vector database |\n| **ConfigMaps/Secrets** | Configuration management |\n| **Health Probes** | Liveness, readiness, startup probes |\n| **Resource Limits** | CPU/memory requests and limits |\n| **Pod Anti-Affinity** | HA distribution across nodes |\n| **ArgoCD** | GitOps continuous deployment |\n\n## Security Considerations\n\n- **Kubeconfig Security**: Keep your kubeconfig file secure (chmod 600)\n- **RBAC**: Ensure appropriate permissions for the kubeconfig user\n- **Network Access**: Secure network connectivity to K3s API server\n- **Audit**: K8s API server logs all actions for auditing\n\n## Troubleshooting\n\n### Connection Errors\n- Verify `KUBECONFIG` path is correct and file exists\n- Check network connectivity to K3s server\n- Test with: `kubectl --kubeconfig /path/to/config get nodes`\n\n### Authentication Errors\n- Verify kubeconfig contains valid credentials\n- Check if certificates are valid and not expired\n\n### Tools Not Showing in Claude\n- Verify absolute path in Claude config\n- Check that config file is valid JSON\n- Restart Claude Desktop completely\n\n### Debug Mode\n```bash\nexport K3S_DEBUG=true\nuv run k3s-mcp-server\n```\n\n## Roadmap\n\n- [ ] ConfigMap and Secret management\n- [ ] PersistentVolume and PVC operations\n- [ ] Ingress management\n- [ ] Job and CronJob support\n- [ ] StatefulSet operations\n- [ ] HorizontalPodAutoscaler (HPA) configuration\n- [ ] Helm chart deployment support\n- [ ] KEDA ScaledObject management\n\n## Dependencies\n\n- **mcp** (\u003e=1.0.0): Model Context Protocol SDK\n- **kubernetes** (\u003e=29.0.0): Official Python client for Kubernetes\n- **pyyaml** (\u003e=6.0): YAML parser for manifest handling\n\n## Contributing\n\nContributions welcome! Areas for improvement:\n- Additional resource types\n- Better error handling\n- Performance optimizations\n- Documentation improvements\n- Test coverage\n\n## License\n\nMIT\n\n## Related Projects\n\n- [Model Context Protocol](https://modelcontextprotocol.io/)\n- [K3s - Lightweight Kubernetes](https://k3s.io/)\n- [KEDA - Kubernetes Event-driven Autoscaling](https://keda.sh/)\n- [Qdrant - Vector Database](https://qdrant.tech/)\n- [uv - Python Package Manager](https://github.com/astral-sh/uv)\n\n---\n\n**Part of the Cortex Platform** - AI-native infrastructure orchestration on K3s\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fry-ops%2Fk3s-mcp-server","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fry-ops%2Fk3s-mcp-server","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fry-ops%2Fk3s-mcp-server/lists"}