Flagged and tainted Bitcoin: sources, cases and what it means for Provenance

Summary

Provenance fills a gap between free sanctions checkers and enterprise compliance platforms: a local, open-source purity test that bitcoin holders can verify themselves. Five findings support the case, and each is detailed with sources below.

1. Purity is a judgment, not a protocol property. Bitcoin does not mark coins as clean or tainted. A purity test compares a coin's history with published lists and reports the exposure it finds.
2. The jeweller's assay is the right model. A fineness out of 1000, a karat figure and a dated certificate turn a history check into a record a counterparty can file.
3. Flagged bitcoin is large and documented. The US forfeiture of 127,271 BTC in the Prince Group case is the biggest example, and sanctions lists already carry Bitcoin addresses.
4. Treasuries keep coins moving. 199 public companies held about 1.264M BTC in June 2026, after a 2025 boom and a 2026 pullback. Sales, mergers and fund transfers keep raising questions about where coins came from.
5. The market leaves a gap. Enterprise platforms are centralised and priced for compliance teams, and free tools mostly offer exact-match checks. I found no local, open-source, boutique equivalent, though the search was not exhaustive.

The concept: purity testing for Bitcoin

A purity test asks one question: what is this coin's history, and does any of it touch a flagged source?

Digital gold raises the stakes. Bitcoin is called digital gold because supply is capped at 21 million coins. Scarcity makes history matter. Coins that are frozen, seized or tied up in legal disputes are not freely available, and anyone buying at size wants to know what they are getting.

The jeweller's assay is the model. A hallmark does not make gold purer. It makes purity trusted, because a respected assayer measured it and stamped the result. Gold fineness is counted in parts per thousand, and 24 karat is pure gold. Provenance borrows that language: a fineness out of 1000, a karat figure and a dated certificate.

Taint is defined by tracing rules, not by the protocol. Bitcoin tracks quantities of unspent outputs, not individual named coins (CoinDesk). Different tracing methods give different answers. The strictest, called the poison method, marks every output of a transaction as tainted if any input was (arXiv). That is why a published methodology matters as much as the list.

How a BTC purity test works

A test takes three steps on the user's own machine: trace the history, test it against the taint list, and export a certificate.

The fineness score and exposure list come from the taint list, which the app checks against its Merkle root before use. Labels are kept for internal records and can be exported for wallet tools.

Taint lists: three tiers of source

A taint list is only as strong as its sources, so each entry should carry its tier, its evidence and its date.

In Provenance, the taint list (the vault) is House of Bitcoin's own data and sits outside the open-source repository. The app checks it against its Merkle root before use, which proves the list was not altered in transit, not that its contents are complete. Publishing the methodology, source categories and last-updated date is what closes that trust gap.

Cases: the largest pools of flagged BTC

Six cases hold or touched most of the large, well-documented flagged bitcoin, and several pools are now in government custody.

Bitcoin is a shrinking share of current illicit flow. Chainalysis recorded about $154B of illicit inflows in 2025, two-thirds tied to sanctioned entities, and stablecoins made up 84% of illicit volume (MEXC). The opportunity for a bitcoin tool is the large historical pools above, not every new theft.

The Bitcoin treasury boom of 2025 and 2026

Public-company bitcoin holdings grew by roughly half between mid-2025 and mid-2026, then came under stress, and both phases create questions about where coins came from.

Trackers count differently, so the trend is indicative, not exact. Strategy alone held about 69% of the pool in May 2026.

2026 brought a pullback. Bitcoin fell 48.7% from its October 2025 peak to about $64,600 by 26 July 2026. At least 20 public treasury companies cut, paused or unwound accumulation, and Strategy's market-to-net-asset-value (mNAV) fell below 1 for the first time (DexTools). MARA disclosed selling 15,133 BTC for about $1.1B in March 2026 (KuCoin).

Why the need persists.

• Coins keep changing hands. Sales, forced liquidations and mergers, such as the April 2026 proposal to merge Twenty One with Strike and Elektron Energy (CoinMarketCap), put large stacks in front of new buyers who will ask about their history.
• Verification is becoming normal. Block launched a proof-of-reserves dashboard with third-party cryptographic verification in April 2026 (Spark).
• Family offices want compliance support. A BNY survey found 74% invested in or exploring crypto (Crypto Briefing), while JPMorgan Private Bank's 2026 report found 89% hold none (Wealth Professional). The surveys differ in sample, but Ocorian reports that 70% of family offices considering digital assets struggle to find outside help with regulatory compliance and reporting (Funds Society).
• Compliance officers worry about third parties. In a survey of 550 chief compliance officers, 75% cited due diligence and oversight of third parties as a concern (DACFP).

No source I found measures demand for taint checks specifically, so that link is an inference to validate with real users.

Market landscape: centralised Bitcoin compliance tools versus a local, verifiable one

Existing tools are centralised and built for compliance departments, which leaves room for a local tool that individuals and family offices can verify themselves.

The enterprise market is real. Chainalysis was estimated at about $190M in annual recurring revenue in 2023, with roughly two-thirds from regulators and law enforcement and pricing from about $10K per seat (Sacra). TRM Labs raised a $70M Series C in February 2026 (CB Insights). Free sanctions checks are already common from Chainalysis, TRM and Scorechain, so a plain list match has little standalone value.

The Provenance column reflects the product's own website. The privacy and verifiability rows for other vendors describe typical hosted, proprietary tools, and I did not audit each one. I found no boutique, local, open-source equivalent, but the search was not exhaustive.

Who needs tainted BTC checker tools

Three groups share one need: to look at a coin's history first, privately, before a bank, auditor or the press asks.

Discretion is the selling point. Because searches run locally, a family office can examine its own coins without telling a vendor which addresses it holds. The 2025 and 2026 treasury data above suggests this audience is growing, though demand for this exact check still needs testing with real users.

Open source and the sharing ethos

Open source matters here because the product's promise is that people can verify it, in the same spirit as running your own bitcoin node.

• Verifiable code. Anyone can read what the app does. It is read-only, and it never creates, signs or broadcasts a transaction.
• Local by design. Searches stay on the user's machine, so a holder's addresses are not revealed to a vendor.
• Shared baseline. The free app gives individuals the explorer, taint checks, certificates and exports, in keeping with bitcoin's culture of open tools.
• Verified data. The taint list is checked against its Merkle root before use. The list itself stays House of Bitcoin's data, outside the repository, and is the basis of the paid Bitcoin Intelligence tier.

The honest limit is that open code does not make the list open. Trust in the results will come from a published methodology, named source categories, visible coverage and evidence for each flag. Open question for the authors: name the alliance members and set out how outside contributions to the taint list are accepted.

Limits and risks

A purity result is a dated test against a list, so its credibility depends on how plainly its limits are stated.

• Not a guarantee. Results depend on list coverage, and official lists lag new thefts. A high fineness is not proof a coin is clean.
• Method-dependent. Tracing rules change the answer, and strict methods can spread taint widely. Criminals can also send tainted coins to innocent addresses (arXiv).
• Not a credit score or compliance advice. The tool tests an address, transaction or output. It does not rate people or entities.
• Naming risk. Publicly labelling a named company's coins as tainted can cause legal and reputational harm if the data is wrong. Use graded exposure, show evidence and offer a correction route.
• Premium is unproven. One CoinDesk analysis found miners had not met buyers paying extra for "virgin" coins (CoinDesk), and the claims of 2% to 20% premiums I found came from promotional content.
• Licences. Check the terms of every list before ingesting it, especially Chainabuse and vendor labels.
• Demand is an inference. The treasury and family-office data show a growing, compliance-conscious audience, not proof that people will pay for this exact check.

This document analyses public data. It is not legal, tax or compliance advice.