Home - Coinspect Security
ill bloom investigation

Ill Bloom: Investigating a Wallet Generation Vulnerability During Active Exploitation

Senior Security Consultant
Wallets

In May 2026, Coinspect began investigating a wallet-drain incident affecting users across multiple networks. What was initially attributed to a single wallet product led to the discovery of a broader wallet-generation vulnerability that had remained hidden for years.

The vulnerability had existed for more than a decade, making it difficult to determine how widely the vulnerable implementation had spread, and attackers were already exploiting it while our investigation was underway. During the investigation, we were able to identify vulnerable wallets before they were drained and worked with relevant parties to help protect affected users despite the practical limitations of responding to an incident of this kind.

This article explains how the investigation developed, the challenges of responding during an active exploitation, and why the findings were disclosed in stages.

We named this investigation Ill Bloom. We created a dedicated research site documenting the technical disclosure, the methodologies used throughout the investigation, and the findings from each stage of the research.

The Root Cause Was Only the Beginning

Identifying the vulnerable implementation was only the beginning. Confirming the root cause explained how vulnerable recovery phrases could be generated, but it did not provide a list of exposed accounts or reveal the scale of the impact.

The complete technical analysis of the vulnerability is documented in Ill Bloom Technical Disclosure: The CryptoJS Randomness Vulnerability.

We generated candidate recovery phrases using the vulnerable random number generation function, derived their corresponding addresses, and compared them against blockchain data to identify addresses with transaction history. This allowed us to distinguish the recovery phrases that had been used in real blockchain accounts.

This process required significant computational resources and was performed progressively across multiple networks, derivation paths, and wallet configurations as the investigation expanded. Differences between affected applications, including entropy sizes, mnemonic languages, derivation paths, and supported networks, introduced additional search configurations.

Identifying Affected Wallet Applications

Identifying vulnerable addresses with on-chain activity did not reveal which wallet applications had generated the affected recovery phrases.

On the other hand, identifying a project that used the vulnerable code was not sufficient by itself. We needed to confirm that the weak randomness was actually used during wallet generation.

We combined on-chain analysis with public source-code research, dependency tracing, and reverse engineering of historical application versions. On-chain activity helped narrow down which wallet applications could have generated the vulnerable addresses, while code analysis was required to confirm that those applications had actually used the vulnerable generation process.

This analysis also required reviewing historical releases, as some wallets may have used the vulnerable implementation only during certain periods over the years. This was especially challenging for closed-source or discontinued wallet applications whose relevant versions were no longer available for inspection.

This work confirmed several affected applications, including active wallets that released fixes and discontinued wallets without any remediation. We document the attribution process and supporting evidence in Identifying the Wallets Behind Vulnerable Recovery Phrases.

The Challenge of Protecting Unknown Victims

Responding during active exploitation created competing priorities. We needed to identify affected wallet applications, discover exposed blockchain accounts, and work with relevant parties while avoiding the release of technical details that could enable further exploitation.

Identifying an exposed account did not reveal the identity of its owner or provide a direct way to help protect its funds. The challenge was therefore not only to locate exposed accounts, but also to establish channels for notifying affected users and helping them take action.

To help users determine whether they were affected, Coinspect also created a public address checker as part of the Ill Bloom research site. The checker allowed users to verify whether their addresses matched the vulnerable address sets identified during the investigation and provided guidance on the next steps.

During the investigation, we reproduced the vulnerability, analyzed public blockchain activity, expanded the known vulnerable address sets, researched wallet implementations, and contacted emergency-response organizations. We shared vulnerable addresses and supporting evidence with relevant parties to help assess exposure and, where possible, support user notification and recovery attempts.

These response efforts also influenced how and when the findings were publicly disclosed.

Disclosure Considerations

Coinspect carefully considered the risks of publishing technical details about this vulnerability. The weakness was already publicly known and had been exploited to drain accounts, but continued downstream reuse of the vulnerable implementation made disclosure relevant for identifying historical exposure. The underlying CryptoJS randomness vulnerability had existed for twelve years, had been fixed six years earlier, and had been publicly discussed multiple times in GitHub issues.

This was not a new zero-day in an actively maintained library. CryptoJS has been effectively unmaintained since 2023, and some downstream wallet-related packages continued to rely on old, unmaintained code that used insecure randomness for seed generation.

Because exploitation has already been observed, withholding the findings would not remove the threat. Disclosure helps affected projects and users identify vulnerable dependency paths, assess historical exposure, and migrate funds away from wallets generated with insecure randomness.

Learn More

The full technical disclosure, investigation methodology, affected wallet analysis, and ongoing research updates are available at illbloom.org.