chronulusai/chronulus

Best practices for using the Chronulus MCP server to produce time-series forecasts with the NormalizedForecaster — demand, foot traffic, share of volume, seasonal weights, occupancy, or any quantity over a future horizon of hours, days or weeks, with no historical data required. Use whenever a user asks Claude to forecast, project, predict a trend over time, or plot a forecast with Chronulus in Claude.ai or Claude Code. Covers writing the session (situation + task), designing one reusable input_data_model, setting the horizon, reusing one agent across many items, rescaling the 0–1 forecast into real units with y_min/y_max (and invert_scale), showing the result with render_forecast_scorecard instead of hand-building a chart, and the optional on-demand risk scorecard.

Best practices for using the Chronulus MCP server to produce reusable, calibrated probability estimates for high-value, intelligence, and prediction-market-style questions (Kalshi, Polymarket, and similar) in sports, elections, politics, culture, crypto, commodities, climate, economics, mentions, finance, and tech & science. Handles binary markets ("will A beat B?", "will X happen by date D?", "will Y exceed a threshold?") and markets with more than two outcomes: a 3-way soccer line, an ordinal ladder (cut/hold/hike), or an exclusive-winner field (awards, tournaments, multi-candidate races). Use whenever setting up or running Chronulus predictions in Claude.ai or Claude Code. Covers reusing one session + agent per category and bet type, dual-framing (run each question twice with the framing swapped, average the Beta parameters) to cancel order bias, batching, Dirichlet reconciliation, and per-category guides, with detailed sport workflows (football, soccer, tennis, volleyball, basketball, e-sports).