SuperAPI AI Directory

vikingdb-mcp-server

VikingDB integration: collection and index info, vector storage and search.

Open website

About this tool

VikingDB MCP server is an MCP server for VikingDB, used to store and search information in VikingDB. VikingDB is a high-performance vector database developed by ByteDance; refer to the official documentation https://www.volcengine.com/docs/84313/1254444 for usage.

Main capabilities: This server implements the following tools: vikingdb-colleciton-intro (introduces VikingDB collections), vikingdb-index-intro (introduces VikingDB indexes), vikingdb-upsert-information (writes information to VikingDB for later use), vikingdb-search-information (searches for information in VikingDB).

Configuration items include: vikingdb_host (VikingDB server host), vikingdb_region (region), vikingdb_ak (Access Key), vikingdb_sk (Secret Key), collection_name (collection name), index_name (index name).

Usage: It can be automatically installed for Claude Desktop via Smithery; the configuration file is ~/Library/Application Support/Claude/claude_desktop_config.json on MacOS and %APPDATA%/Claude/claude_desktop_config.json on Windows. The repository also includes instructions for development/unpublished server configuration and published server configuration, as well as the build and release process: sync dependencies and update the lock file, build package distributions (generate source and wheel distributions in the dist/ directory), publish to PyPI (PyPI credentials must be set via environment variables or command arguments; for Token use --token or UV_PUBLISH_TOKEN, for username/password use --username / UV_PUBLISH_USERNAME and --password / UV_PUBLISH_PASSWORD).

Debugging: Because the MCP server runs over stdio, debugging can be difficult. MCP Inspector is recommended; it can be started via an npm command, and after startup it displays a URL that can be accessed in a browser to begin debugging.

Limitations: The material does not provide more details about performance, pricing, platform support, or specific usage steps.