ZenLedger + Hyperliquid: reconciled totals on an unreconciled record
updated november 2026 · ~5 min · basisproof research · not tax advice
Reconciliation is not coverage
ZenLedger's strength is reconciliation — taking what the exchanges send and making the columns agree. That works when the inputs are complete. Hyperliquid's API shape makes that assumption dangerous: the venue's fill endpoint retains roughly the last 30 hours, so an importer can reconcile a perfectly consistent record that is also a small fraction of your actual year.
The honest caveat: integrations evolve constantly, and ZenLedger may use more than one data path for Hyperliquid. Check what your own import contains — this page gives you the tools to do that, whatever the current sync behavior is.
The three ways a sync loses data silently
Truncated fills. The venue only serves ~30 hours of fills per endpoint. An importer that reads it gets a valid, successful, deeply incomplete response — nothing to flag, nothing to warn about.
A phantom-zero funding line. Funding settles hourly on its own endpoint outside fills. If the import only reads trades, funding simply doesn't exist in the record — and on an account that held positions overnight, that's a missing taxable stream, not a clean zero.
Missing markets and inflated flows. Builder-dex markets (HIP-3 venues like xyz:, flx:) carried over half of perp volume at their mid-2026 peak yet live outside the standard perp universe. Meanwhile, sub-account and spot↔perp transfers — moves between your own wallets — can surface as deposits and withdrawals, inflating both sides of the ledger.
Verify before you reconcile
Run three checks on the imported record before trusting any report built on it. Transaction count: plausible against how much you actually traded? Funding line: exactly zero on a perps account that held positions? Symbols: do the builder-dex markets you traded appear anywhere? A failure on any of these means the sync lost data upstream — and re-syncing just re-reads the same endpoint.
Reconciling a truncated record produces confident, internally consistent, wrong numbers. The totals agree with each other because they all descended from the same partial input.
What complete actually requires
Complete fills live on a chain indexer — the venue endpoint cannot serve them past its retention window. Funding and ledger flows come from separate full-history endpoints. And before any of it feeds a filing, the rebuilt record should close an independent account-value identity against the venue's own series.
The free check recomputes your record fill by fill and prints the totals — run it on your address, then compare against your sync. Agreement is evidence. A gap names the stream that was dropped. The paid report produces the corrected artifacts — complete fills, funding, ledger, per-disposal basis records — that a preparer or a manual ZenLedger CSV import can consume.
Common questions
?Does ZenLedger support Hyperliquid?
ZenLedger supports a broad range of exchanges and wallets. But a working sync and a complete record are different things — the verification checks above tell you which you have.
?My reports reconcile — doesn't that mean the data is right?
Reconciliation proves the imported rows agree with each other. It says nothing about rows that never arrived — a truncated fill stream reconciles perfectly.
?Can corrected data go back into ZenLedger?
ZenLedger accepts manual CSV imports, so a complete recomputed export can replace the sync's partial record. The fix is upstream — a source that serves full history.
Sources & method
See what the venue API isn't showing you — free
The fill endpoint forgets everything older than ~30 hours. The free gap check measures what fraction of your record it actually serves — seconds, not minutes.
Measure my gap