Skip to main content
AVOID.NET
← Cozy V2reviewed 2026-11-01 · 7 claims checked

Fact-check findings

What an automated fact-checker found when it re-read Cozy V2 against the sources the page cites. Only the most recent review is shown.

Read this first

These findings are produced by an automated reviewer, and its results vary between runs: the same page, checked three times on the same day, came back with 15%, 20% and 34% of its claims disputed, mostly because each run extracted a different number of claims. Treat what follows as leads, not rulings.

“Disputed” means the reviewer could not reconcile the claim with the evidence it cited. It does not mean the claim is false. “Unverifiable” means no reachable source settled it either way.

Nothing here changes the page on its own. A proposed correction is applied only after a human moderator approves it; until then the page reads as it did when reviewed.

disputed

1 claim

The reviewer could not reconcile the claim with the evidence it cited. This is a lead, not a ruling that the claim is false.

  1. #1[disputed][awaiting moderator]in the timeline
    “2025-08-30”
    reviewerVerichains published its post-mortem on 2025-08-30The Verichains post is dated October 15, 2025, not August 30, 2025. Note timeline[6] and timeline[7] both carry the quoted date value; target is the Verichains entry (index 7).
    Proposed correction (not yet applied)
    2025-10-15

unverifiable

2 claims

No source the reviewer could reach confirms or contradicts the claim.

  1. #2[unverifiable][awaiting moderator]in section: August 2025 Exploit — Protocol Logic Vulnerability
    “Following the exploit, the attacker bridged the stolen funds to an Ethereum mainnet address and subsequently deposited them into Tornado Cash, a transaction mixing service, in an apparent attempt to obscure the on-chain trail.”
    reviewerAttacker bridged to Ethereum and deposited into Tornado Cash, attributed to the Verichains/Decurity reportingThe only readable cited source (Verichains) is silent on laundering; the Decurity tweet was blocked. Not scored.
  2. #5[unverifiable][awaiting moderator]in section: Funding, Team, and Background
    “The company was incorporated and operates out of Seattle, Washington.”
    reviewerCompany incorporated and operates out of Seattle, WashingtonNot in the CoinDesk source; Crunchbase not checked.

partially supported

2 claims

The cited evidence supports part of the claim but not all of it.

  1. #6[partially supported][awaiting moderator]in section: Protocol Risk Factors
    “The protocol is restricted to Optimism and has not expanded to additional chains.”
    reviewerNo public statement from Cozy team / no remediation / no regulatory actions; Cozy has not expanded beyond OptimismPlausible, but the page is out of date given a second exploit in Sept 2026; TVL figure is also stale.
  2. #7[partially supported][awaiting moderator]in section: Protocol Overview
    “As of mid-2025, the protocol's total value locked stood at approximately $184,931 on OP Mainnet, according to DeFiLlama.”
    reviewerTVL approx $184,931 mid-2025Figure cannot be verified exactly and is now superseded by a near-total TVL collapse in 2026 (stale).

confirmed

2 claims

The cited evidence supports the claim as written.

  1. #3[confirmed][no action needed]in section: August 2025 Exploit — Protocol Logic Vulnerability
    “On August 29, 2025, Cozy V2 was exploited on OP Mainnet through a critical authorization flaw in its withdrawal system, resulting in losses of approximately $427,000.”
    reviewerExploit on Aug 29, 2025, ~$427,000 loss, missing caller check in completeWithdraw, receiver_ not validated, victim redemption ~$376,661 ID 6 initiated Aug 28Core technical details match Verichains and a secondary report.
  2. #4[confirmed][no action needed]in section: Funding, Team, and Background
    “Cozy Finance raised a $2 million seed round announced on September 3, 2020, led by Electric Capital.”
    reviewer2020 seed round $2M led by Electric Capital with named investors; Sheng and Dousti backgroundsConfirmed.
How this fits together. The reviewer reads the published page and its cited sources and records one finding per claim. A human moderator decides whether each proposed correction is applied; those decisions, and the score changes they cause, appear in the audit log. Earlier review runs are not shown here; only the latest reflects the page as it stands.