KosmoKrator

data

Milvus MCP Integration for OpenAI Agents SDK

Connect Milvus to OpenAI Agents SDK through the local KosmoKrator MCP gateway with scoped tools, credentials, and write policy.

7 functions 5 read 2 write API key auth
Submit Milvus feedback View package source

Connect Milvus to OpenAI Agents SDK

Attach KosmoKrator integration tools to OpenAI Agents SDK workflows through a local MCP gateway.

Start the KosmoKrator MCP gateway locally and point the OpenAI Agents SDK MCP tool at that process or wrapper. The gateway is local, scoped to this integration, and starts with --write=deny so OpenAI Agents can inspect read-capable tools without receiving write access by default.

Milvus MCP Config for OpenAI Agents SDK

Use headless JSON commands for CI-style execution and MCP for agent tool discovery.

{
  "mcpServers": {
    "kosmokrator-milvus": {
      "type": "stdio",
      "command": "kosmo",
      "args": [
        "mcp:serve",
        "--integration=milvus",
        "--write=deny"
      ]
    }
  }
}

Run the Gateway Manually

kosmokrator mcp:serve --integration=milvus --write=deny

Why Use KosmoKrator Here

Scoped tools

Expose only Milvus instead of a broad multi-service tool list.

Local credentials

Reuse credentials already configured for the KosmoKrator CLI and Lua runtime.

Write policy

Start read-only, then opt into ask or allow for trusted workspaces.

Milvus Tools Visible to OpenAI Agents

OpenAI Agents sees stable MCP tool names generated from the Milvus integration catalog.

MCP toolSource functionTypeDescription
integration__milvus__milvus_list_collections milvus.milvus_list_collections Read List all vector collections in Milvus. Returns collection names and details that can be used for further operations.
integration__milvus__milvus_get_collection milvus.milvus_get_collection Read Get details of a specific Milvus collection by its name, including schema and description.
integration__milvus__milvus_create_collection milvus.milvus_create_collection Write Create a new vector collection in Milvus. A collection requires a name and the embedding dimension size.
integration__milvus__milvus_insert_documents milvus.milvus_insert_documents Write Insert documents with embedding vectors into a Milvus collection. Each document requires a vector and an optional ID.
integration__milvus__milvus_search_documents milvus.milvus_search_documents Read Search for similar documents in a Milvus collection using a query vector. Returns the most similar documents ranked by distance or similarity.
integration__milvus__milvus_get_collection_stats milvus.milvus_get_collection_stats Read Get statistics for a Milvus collection, including row count and index information.
integration__milvus__milvus_get_health milvus.milvus_get_health Read Check the health status of the Milvus vector database server. Returns health and version information.

Related Milvus Pages

Milvus CLIUse direct JSON commands instead of MCP discovery. Milvus MCPFull gateway reference and MCP tool names. KosmoKrator for OpenAI AgentsClient-level gateway setup.