Agent-readable docs index: /llms.txt. Full docs in one file: /llms-full.txt. Download /docs.zip to grep all markdown files locally.

Reference Rates

Reference rates blend the same market across platforms (and across rolling contracts) into one continuous benchmark. Rate prices are 0-100 percent (same scale as market prices), with a spread capturing the dispersion across sources.
For volume_weighted_average, each source's effective weight is its configured or roll weight multiplied by cumulative contract/share count: Σ(price_i · weight_i · count_i) / Σ(weight_i · count_i). Other supported methodologies are simple_average and median.

List rates

curl "http://localhost:7101/api/v1/rates"

Rate detail

Returns the source markets, their weights, and recent price changes.
curl "http://localhost:7101/api/v1/rates/adj_redp"

Rate price history

curl "http://localhost:7101/api/v1/rates/adj_redp/prices?interval=1hour&limit=24"

Rolling-contract chains

Chains stitch successive contracts together (e.g. the 2024 then 2028 presidential contract) so a benchmark survives expiries. The roll_schedule blends adjacent contracts during the roll window.
A chain-backed rate is consumed like any other rate: there are no separate chain read endpoints. The blended series comes from the rate's own price history, and sources=true returns the per-leg roll weights for each point:
# Rolling rate price history, with per-leg roll weights curl "http://localhost:7101/api/v1/rates/adj_redp/prices?interval=1hour&limit=24&sources=true"
Chain roll config (contracts, schedule) is managed via the superuser chain endpoints. See the API Reference for full schemas.