spark.exposed

Privacy

A July Sparkscan sample exposed Spark identities in Lightning entries

A sanitized unauthenticated mainnet sample returned Spark-side addresses and raw identity keys for every sampled Lightning record, together with exact amount and timestamp fields. The address is a durable wallet user ID even when an integrating app never shows it to the user. 1234

4 sourcesEvidence reviewed through September 11, 2026

Historical observation checked for source support on September 11. The complete original experiment or product runtime was not reproduced; its reported status must not be read as current status.

What the evidence shows

Technical and product details

  1. Documented behavior

    In a sanitized read-only mainnet observation on 2026-07-14, an unauthenticated request to Sparkscan's latest-transactions endpoint returned 100 records. All 35 Lightning records contained the Spark-side address and raw identity public key, alongside exact amount and timestamp fields. 1

    Supporting material shown on this page

  2. Documented behavior

    Sparkscan documents that endpoint as returning network transactions “across all addresses,” with transaction type, status, exact amount, from/to identifiers, and creation and update times. 2

    Supporting material shown on this page

  3. Documented behavior

    Spark describes a Spark address as a user ID derived from the wallet identity key and says the same wallet always produces the same address. Spark separately says some Lightning-only applications never show that address to their users. 34

    Supporting material shown on this page

    Spark addressingOfficial documentation · docs.spark.money · checked 2026-07-14

Interpretation

Why this matters for Spark's trust model

Hiding the Spark address in a wallet interface does not hide it on the public data surface. A payer or merchant who knows a Lightning amount and time can search for a matching record, discover the otherwise-hidden durable Spark pseudonym, and then monitor any history that remains publicly queryable. 1234

This section states the trust-model consequence of the cited evidence. Conditional or inferred claims remain bounded by the next section.

Scope control

What this does not establish

  • The observation does not establish that the feed is complete or that any sampled user had enabled privacy mode. The amount-and-time identity match is a correlation method, not proof that every match is unique. Fully enforced privacy can still suppress covered Bitcoin queries when every relevant participant is private. The sample is not a complete-network census or a new September observation. Amount/time matching need not be unique and differs from decoding an identity embedded in an invoice. 14

Sources and excerpts

Primary sources

The relevant details and available source-code excerpts appear alongside the claims above. This list preserves the complete original-source trail for independent verification.

1

sanitized mainnet observation

Reproducible audit manifest · Review of Spark's trust model · Pinned or archival · checked 2026-07-14

Open original source
2

Sparkscan transaction API

Official documentation · docs.sparkscan.io · Live source · checked 2026-07-14

Open original source
3

Spark addressing

Official documentation · docs.spark.money · Live source · checked 2026-07-14

Open original source
4

Spark privacy mode

Official documentation · docs.spark.money · Live source · checked 2026-07-14

Open original source

How sources, absence findings, and limitations were evaluated