{
  "markdown": "[![smithery badge](https://smithery.ai/badge/OjasKord/local-model-suitability-mcp)](https://smithery.ai/servers/OjasKord/local-model-suitability-mcp)\n\n# Local Model Suitability MCP\n\n[![ToolRank](https://toolrank.dev/badge/dominant.svg)](https://toolrank.dev/ranking)\n\n**Cloud inference is expensive. Everything that can run locally should.**\n\nThis MCP server tells your agent — before every cloud API call — whether the task can be handled by a local model instead. Route to Ollama, LM Studio, or llama.cpp when you can. Only pay for cloud when you must.\n\n## The Tool\n\n### `check_local_viability`\n\nCall this BEFORE every cloud inference call. If verdict is `LOCAL`, skip the cloud call entirely and route to your local model. Only use cloud when this tool returns `CLOUD`.\n\n**Inputs:**\n| Field | Required | Description |\n|---|---|---|\n| `task` | ✅ | The exact task you are about to send to a cloud model |\n| `quality_threshold` | Optional | `PRODUCTION` (default) / `PROTOTYPE` / `BEST_EFFORT` |\n| `data_sensitivity` | Optional | `PUBLIC` (default) / `INTERNAL` / `CONFIDENTIAL` |\n\n`CONFIDENTIAL` forces `LOCAL` regardless of task complexity — data never leaves the machine.\n\n**Response:**\n```json\n{\n  \"verdict\": \"LOCAL\",\n  \"confidence\": \"HIGH\",\n  \"reason\": \"Simple text summarisation — no reasoning depth required. Any 7B+ local model handles this well.\",\n  \"estimated_cost_saving\": \"$0.002-0.008 saved per call at claude-sonnet pricing\",\n  \"recommended_local_models\": [\"llama3.2:8b\", \"mistral-7b\", \"phi3:medium\"],\n  \"cloud_justified_reason\": null,\n  \"analysis_type\": \"AI-powered cost routing — NOT a simple lookup\"\n}\n```\n\n## Data Sources\n\n- AI reasoning: Anthropic Claude (claude-sonnet) — cost routing analysis\n- No external data sources — pure AI reasoning\n\n## Pricing\n\n| Plan | Calls | Price |\n|---|---|---|\n| Free | 20/month | $0 |\n| Starter | 500-call bundle | $20 |\n| Pro | 2,000-call bundle | $70 |\n\n[Subscribe at kordagencies.com](https://kordagencies.com)\n\n## Setup\n\n```json\n{\n  \"mcpServers\": {\n    \"local-model-suitability\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"local-model-suitability-mcp\"],\n      \"env\": {\n        \"ANTHROPIC_API_KEY\": \"your-key\",\n        \"API_KEY\": \"your-lms-api-key-for-paid-tier\"\n      }\n    }\n  }\n}\n```\n\nFree tier requires no API key — tracked by IP.\n\n## Harness Integration\n\n### Claude Code / Claude Desktop (.mcp.json)\n```json\n{\n  \"mcpServers\": {\n    \"local-model-suitability\": {\n      \"type\": \"http\",\n      \"url\": \"https://local-model-suitability-mcp-production.up.railway.app\"\n    }\n  }\n}\n```\n\n### LangChain (Python)\n```python\nfrom langchain_mcp_adapters.client import MultiServerMCPClient\nclient = MultiServerMCPClient({\n    \"local-model-suitability\": {\n        \"url\": \"https://local-model-suitability-mcp-production.up.railway.app\",\n        \"transport\": \"http\"\n    }\n})\ntools = await client.get_tools()\n```\n\n### OpenAI Agents SDK (Python)\n```python\nfrom agents import Agent, HostedMCPTool\nagent = Agent(\n    name=\"Assistant\",\n    tools=[HostedMCPTool(tool_config={\n        \"type\": \"mcp\",\n        \"server_label\": \"local-model-suitability\",\n        \"server_url\": \"https://local-model-suitability-mcp-production.up.railway.app\",\n        \"require_approval\": \"never\"\n    })]\n)\n```\n\n### LangGraph\nSame as LangChain above — langchain-mcp-adapters works with LangGraph natively.\n\n## Legal\n\nResults are for cost-optimisation guidance only and do not constitute technical advice. Full terms: [kordagencies.com/terms.html](https://kordagencies.com/terms.html)\n",
  "bytes": 3500,
  "sha": "31e176cb40e01d62e3de8815af000fa71cb5b0c789bb42901a3cd198d50a00f6",
  "repo_slug": "ojaskord/local-model-suitability-mcp",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_ojaskord_local_model_suitabili_eaa17e92/readme"
}