vikingdb
a mcp server for vikingdb store and search
claude mcp add --transport stdio kashiwabyte-vikingdb-mcp-server uv --directory dir to mcp-server-vikingdb run mcp-server-vikingdb --vikingdb-host your host --vikingdb-region your region --vikingdb-ak your access key --vikingdb-sk your secret key --collection-name your collection name --index-name your index name
How to use
This VikingDB MCP server exposes tools to interact with VikingDB storage and search capabilities. The available tools include: vikingdb-colleciton-intro for introducing the VikingDB collection, vikingdb-index-intro for discussing the VikingDB index, vikingdb-upsert-information to upsert data into VikingDB for later retrieval, and vikingdb-search-information to search information within VikingDB. To use these tools, configure the server with your VikingDB host, region, and credentials, along with the target collection and index names. Once running, you can invoke the specific tool commands to understand the collection/index, insert data, and perform searches against VikingDB. For debugging and development, you can use the MCP Inspector tool to attach to the running MCP server and inspect the stdio-based I/O flow.
How to install
Prerequisites:
- Python with uv (or uvx-based tooling) installed on your system.
- Access to VikingDB service with host, region, and credentials (AK/SK).
- Optional: Smithery CLI for easy installation.
Install via Smithery (recommended for Claude Desktop users):
npx -y @smithery/cli install mcp-server-vikingdb --client claude
Manual installation via UV (development):
- Ensure your environment has UV installed.
- Prepare a directory for the MCP server configuration and run script.
- Start the server using the UV configuration provided (see mcp_config).
To run via UVX (published configuration):
uvx mcp-server-vikingdb --vikingdb-host your_host --vikingdb-region your_region --vikingdb-ak your_access_key --vikingdb-sk your_secret_key --collection-name your_collection_name --index-name your_index_name
Note: If you are publishing to PyPI or distributing the Python package, ensure you have your PyPI credentials configured if you package and publish via UV_PUBLISH_TOKEN or UV_PUBLISH_USERNAME/UV_PUBLISH_PASSWORD.
Additional notes
Tips and notes:
- The MCP server runs over stdio; debugging can be challenging. Use the MCP Inspector for a better debugging experience. You can launch the Inspector with the provided npm command to attach to the UV/UVX server run, which will expose a browser-based debugging URL.
- Replace placeholders like your host, region, access key, secret key, collection name, and index name with your actual VikingDB configuration.
- If you encounter environment-related issues, ensure the necessary environment variables or CLI flags are set for authentication with VikingDB and that network access to VikingDB is permitted.
- The configuration supports separate operational modes (development with uv and published with uvx). Choose the one that matches how you distribute or run the server.
- When debugging, consider mounting a directory for --directory in UV to organize logs and runtime data.
Related MCP Servers
mcp-vegalite
MCP server from isaacwasserman/mcp-vegalite-server
github-chat
A Model Context Protocol (MCP) for analyzing and querying GitHub repositories using the GitHub Chat API.
nautex
MCP server for guiding Coding Agents via end-to-end requirements to implementation plan pipeline
pagerduty
PagerDuty's official local MCP (Model Context Protocol) server which provides tools to interact with your PagerDuty account directly from your MCP-enabled client.
futu-stock
mcp server for futuniuniu stock
mcp -boilerplate
Boilerplate using one of the 'better' ways to build MCP Servers. Written using FastMCP