Connect your AI agent to Spavik
Give your agent tools to evaluate data, train models and make predictions through the Model Context Protocol.
MCP server
https://mcp.spavik.co - Transport
- Streamable HTTP
- Authentication
- API key
Hosted by Spavik. No local server to install.
Connect your MCP client
Claude Code
Export your Spavik API key as SPAVIK_API_KEY, then run this command to add the remote server to Claude Code.
Terminal
claude mcp add --transport http spavik https://mcp.spavik.co \
--header "X-Spavik-Api-Key: $SPAVIK_API_KEY"From data to predictions
Give your agent CSV content with a header or JSON rows. The remote server accepts the data itself, rather than a local file path.
Evaluate and train
Check whether your data can predict the target, then create a model.
evaluate_datatrain_model
Predict and forecast
Request predictions from a model, or backtest and forecast a time series.
predictbacktest_forecastforecast
Report and improve
Submit observed outcomes, review performance and refresh the model.
submit_outcomesmodel_performancerefresh_model
Connect your workspace
Create an API key, configure your client and start using Spavik from your agent.