MCP Server MCP

io.github.n24q02m/mnemo-mcp

By github 👁 1 views ▲ 0 votes

Persistent AI memory with hybrid search (FTS5 + semantic) and cross-machine sync.

Homepage Verified source GitHub

Install

uvx mnemo-mcp

Configuration Example

{
  "remotes": [],
  "packages": [
    {
      "registryType": "pypi",
      "identifier": "mnemo-mcp",
      "version": "2.7.0",
      "runtimeHint": "uvx",
      "transport": {
        "type": "stdio"
      },
      "environmentVariables": [
        {
          "description": "Provider API keys (format: PROVIDER_API_KEY:key,...). Select models per task with EMBEDDING_MODELS / RERANK_MODELS / LLM_MODELS (CSV provider/model, order = litellm fallback); provider is inferred from the model prefix. Empty embedding/rerank chain falls back to the built-in local Qwen3 model.",
          "isSecret": true,
          "name": "API_KEYS"
        }
      ]
    },
    {
      "registryType": "oci",
      "identifier": "docker.io/n24q02m/mnemo-mcp:latest",
      "runtimeHint": "docker",
      "transport": {
        "type": "stdio"
      },
      "environmentVariables": [
        {
          "description": "Provider API keys (format: PROVIDER_API_KEY:key,...). Select models per task with EMBEDDING_MODELS / RERANK_MODELS / LLM_MODELS (CSV provider/model, order = litellm fallback); provider is inferred from the model prefix. Empty embedding/rerank chain falls back to the built-in local Qwen3 model.",
          "isSecret": true,
          "name": "API_KEYS"
        }
      ]
    }
  ]
}

mcp model-context-protocol pypi oci