Use this file to discover all available pages before exploring further.
Lighter is a separate top-level venue family. Use /v1/lighter/* routes for Lighter instruments, order books, trades, funding, open interest, and L3 depth.
The first Lighter request confirms that the key, base URL, Lighter namespace, symbol, and response envelope are working together. It does not confirm that every Lighter data family, time window, L3 route, or replay job is available to the same key.
route, symbol, response timestamp, meta.request_id, status code
Next branch
trades, candles, funding, OI, L3, replay, or reconstruction
Keep this packet separate from Hyperliquid core, Spot, HIP-3, and HIP-4 packets. A generic BTC string is not enough context for logs or generated clients.
Do not route Lighter symbols through Hyperliquid examples. Do not assume Lighter L3 and Hyperliquid L4 are the same primitive. Keep venue family in logs, storage, and generated code so later joins or comparisons can distinguish sources.
Use this path when the product specifically needs Lighter data rather than a generic crypto price. The first route should confirm that Lighter is the correct venue family, then the implementation can switch to trades, candles, funding, open interest, L3, replay, or SDK reconstruction depending on the job.
Use L2 order-book routes when aggregated price-level depth answers the question. Use L3 when the workflow needs order-level state, queue behavior, or reconstruction-oriented history. Treat L3 as a different data family in storage, tests, and exports; do not merge it into generic order-book output without labeling the depth level.
Keep the first working request as a fixture. When the client expands, compare every new route to that fixture for auth, envelope shape, symbol handling, and request-ID logging. That makes Lighter-specific regressions easier to catch.If the next workflow is L3 or replay, add gap handling before using the result in production.
Last modified on May 18, 2026
Was this page helpful?
⌘I
Assistant
Responses are generated using AI and may contain mistakes.