This walks the whole path once, end to end. Everything else in these docs is a
detail of one of these steps.
Before that: your first three steps
When you confirm your address you get a personal organization, a personal
research space, and in it a Getting started project with one page. The
page shows a real cited value: the debt repaid in the most recent Aave v3
liquidation on record at the moment the page was made, read from the
warehouse and cited exactly as you would cite it yourself. (If the warehouse
was unreachable at that moment the page says so, and the citation is added
when it is back.) Delete the project whenever you like.
The dashboard carries a checklist of three steps until each is done or you
put it away. Every step ticks itself off from a record, not from a click:
- Open a transaction and cite a value. From any transaction in the
explorer, Cite lists the decoded fields of its logs; choose one and
you get a{{cite:N}}token to paste into a page. - Create a research page from a template. Question, method, data,
findings: the structure a result needs to be checkable. Choose the
template when you create the page. - Save a pair to your favourites. Find a pair in Prices and star it;
favourites sit at the top of the dashboard.
1. Find the market
Open Prices. Choose a base and a quote token (say WETH and USDC) and
optionally a venue. You get every pool that trades that pair, ranked by
activity, with the latest close oriented as quote per base.
There are over 600,000 markets. The finder ranks them; it does not list them.
If a pair returns nothing, that pair has no pool we have ingested, and the page
says so rather than showing an empty chart.
2. Read the series
Open a pool. You can set the range (7 days to a custom window), the periodicity
(hourly, daily, weekly), overlay up to three other venues for the same pair, and
show the spread between them as a percentage.
Prices are measured from on-chain events (V2 reserves, V3 sqrtPriceX96), not
taken from an aggregator. The methodology note on each page says which.
3. Take the data
Two ways out:
- CSV from the chart page. Needs the Data subscription.
- A citation. Below the chart, “cite this series” pins the exact bytes and
gives you a SHA-256 hash plus a public URL. Put the hash in your paper; anyone
can resolve it at/citations/<hash>and get the identical file.
The citation is the one to prefer. A CSV in a supplementary folder is a claim
about data; a hash is a check on it.
4. Keep your work
Research holds projects, pages and notes. Save a transaction or a block into
a page and annotate individual values. Notes attach either to one saved item or
to a value (an address, a topic, a hash), and then follow that value wherever
it appears on the platform.
What to read next
- How prices are computed: before you cite anything.
- Citations and reproducibility
- Research pages and notes