Skip to main content
This recipe shows you how to build a discovery board that follows launchpad tokens through their lifecycle: newly launched, close to graduating, and graduated. POST /v1/discovery returns all three columns in one request.

1. Request the columns

Each column takes its own sortBy, order, limit and filters. Discovery boards documents every filter and sort field. The graduating and graduated columns also apply default minimums, such as 10 holders. Your filters can raise these minimums but cannot lower them.

2. Render the response

The response is a list with one entry per column. The deltas array of each column contains the whole ranked list. Draw the items in rank order. data is the full token object, so each card can show the following fields.

3. Refresh the board

Poll the endpoint on an interval. An interval of 2 to 5 seconds suits a fast-moving newPairs column and stays well within your rate limit. To receive changes as they happen instead, subscribe to spot.tokens with a filter close to the column, such as { "chains": ["solana"], "stages": ["about_to_graduate"], "holderCount": { "min": 10 }, "liquidityUsd": { "min": "1000" }, "marketCapUsd": { "min": "5000" } }. The stream does not apply the column defaults, so include them in your filter. A token is sent each time it changes and is removed with an exit frame. The stream does not rank tokens, so sort the set in your client and seed it from one discovery read. When you redraw, match cards by chain + address so that a token that changes rank animates instead of flickering. A token that leaves a column’s ranked list is no longer in that column, and a token that graduates appears in graduated. Occasionally, a card’s windowed stats are zero in one response while the token’s full record catches up. When this happens, keep the card’s previous values; the next refresh corrects them.

Filter presets

These values are starting points, not recommendations. Tune them for your users.

Multiple chains

To cover more than one chain, send several slugs in chains, for example ["solana", "bsc", "robinhood"]. Each column then ranks tokens from all listed chains as one set. Use data.chain to label each card with its chain.