Loading...

MCP server for VictoriaMetrics, enabling data ingestion, querying, and management via PromQL, with potential risks from write operations.
Boost this tool
Subscribe to listing upgrades or segmented pushes.
MCP server for VictoriaMetrics, enabling data ingestion, querying, and management via PromQL, with potential risks from write operations.
This MCP server provides both read and write access to a VictoriaMetrics database. While querying is generally safe, write operations pose a moderate risk if not properly controlled. Ensure proper input validation and resource limits to mitigate potential abuse.
Performance depends heavily on the complexity of the PromQL queries and the size of the VictoriaMetrics dataset. Indexing and data partitioning are crucial for optimal performance.
Cost is primarily determined by the VictoriaMetrics instance's resource consumption (CPU, memory, disk). Complex queries and high data ingestion rates can increase costs.
npx -y @smithery/cli install @yincongcyincong/victoriametrics-mcp-server --client claude{
"mcpServers": {
"victoriametrics": {
"command": "npx",
"args": [
"-y",
"@yincongcyincong/victoriametrics-mcp-server"
],
"env": {
"VM_URL": "",
"VM_SELECT_URL": "",
"VM_INSERT_URL": ""
}
}
}
}VM_URLvm_data_writeWrites time series data to VictoriaMetrics.
Allows data injection, potentially overwriting or corrupting existing data.
vm_prometheus_writeImports Prometheus exposition format data into VictoriaMetrics.
Enables importing potentially malicious or incorrect data.
vm_query_rangeQueries time series data over a specified time range.
Read-only operation; no data modification.
vm_queryQueries the current value of a time series.
Read-only operation; no data modification.
vm_labelsRetrieves all unique label names.
Read-only operation; no data modification.
vm_label_valuesRetrieves all unique values for a specific label.
Read-only operation; no data modification.
Environment Variable
cloud
This MCP server provides both read and write access to a VictoriaMetrics database. While querying is generally safe, write operations pose a moderate risk if not properly controlled. Ensure proper input validation and resource limits to mitigate potential abuse.
Autonomy is limited by the capabilities exposed through the MCP tools. The agent can query and write data, but cannot perform administrative tasks or execute arbitrary code.
Production Tip
Implement rate limiting and query complexity limits to prevent resource exhaustion and ensure stability.
Authentication is based on environment variables (VM_URL, VM_SELECT_URL, VM_INSERT_URL). Ensure these are securely managed.
Implement query timeouts and complexity limits in your VictoriaMetrics instance.
No, it is not recommended. Expose it only within a trusted network and consider using a reverse proxy with authentication.
Monitor the resource usage (CPU, memory) of the Node.js process running the server.
No, there is no explicit delete functionality exposed in the provided documentation. However, data can be overwritten using write operations.
The documentation does not specify supported versions. Compatibility should be tested with your specific VictoriaMetrics deployment.
Errors from VictoriaMetrics are likely propagated back to the client. Implement proper error handling in your client application.