The npm package 'checkout-test-utils' version 35.1.3 contains malicious code that automatically downloads and executes attacker-controlled binaries from obfuscated Cloudflare Workers hosts. The package disguises its payload download and execution to evade detection, writing binaries to temporary files and running them without verification. Any user installing or requiring this package will inadvertently execute this malicious code on their host system.
The npm package 'checkout-use-boolean' version 35.4.5 contains malicious code that executes upon module import. It loads a secondary script that downloads a platform-specific binary from obfuscated worker domains, using DNS TXT record fallback channels to evade detection. The payload is saved to temporary directories with disguised filenames, made executable, and launched detached via system shell commands. This behavior triggers automatically when the package is required by any other module.
The npm package 'claims-auth-api-adapters' version 35.3.3 contains malicious code that unconditionally loads a runtime module which fetches and executes an attacker-controlled binary payload. This payload is retrieved from obfuscated Cloudflare workers.dev hosts and fallback DNS TXT records, then saved and executed on the host system, masquerading as telemetry. Any process requiring this package will execute this malicious code, leading to remote code execution on the host.
The npm package 'checkout-use-disable-interaction' version 35.4.9 contains malicious code that downloads and executes a platform-specific native binary from obfuscated Cloudflare Workers subdomains and fallback DNS TXT records. The payload is saved to system temporary directories with filenames mimicking legitimate system artifacts and executed detached from the main process. The package README falsely advertises a benign microservice module, while the actual code performs unauthorized code execution and persistence by periodically re-running the dropper every ~6 hours.
The npm package 'checkout-use-notifications' version 35.6.1 contains malicious code that, upon requiring the package, loads a script which fetches a platform-specific binary payload from obfuscated Cloudflare Workers hosts. The payload is delivered using a DNS TXT-record fallback channel and is saved to temporary directories under disguised filenames before execution. This behavior is characteristic of a malicious dropper designed to evade detection and persist on the system.
The npm package 'claims-auth-domain' version 35.5.4 contains malicious code that executes at import time. It loads a script that downloads a binary payload from attacker-controlled infrastructure, stages it on the host with a disguised filename, sets executable permissions, and runs it detached. The payload delivery uses obfuscated domain names and falls back to reconstructing the binary from DNS TXT records if HTTPS delivery fails. The code masquerades as telemetry and respects certain environment variables to evade detection. This results in full host code execution under attacker control upon loading the package.
The npm package claims-format-date version 35.6.5 contains malicious code that auto-executes a binary dropper upon being required. It downloads platform-specific binaries from obfuscated Cloudflare Workers hosts and fallback domains, writes them to temporary directories under disguised names, sets executable permissions, and executes them detached from the main process. The package uses runtime string concatenation to evade static analysis and includes a second dropper disguised as an analytics SDK. This package mimics legitimate date-formatting utilities but performs remote payload execution immediately on load.
The npm package 'claims-get-error-string' version 35.5.9 contains malicious code that downloads and executes a platform-specific binary from hardcoded Cloudflare Workers hosts. The binary is saved to temporary directories with innocuous names, given executable permissions, and executed detached from the main process. Additionally, a DNS-TXT covert channel is used as a fallback method to reassemble and deliver the payload from sequential TXT records under the domain wel1.ru. A secondary telemetry module with similar dropper logic is included but not invoked in this version. There is no hash pinning or signature verification for the downloaded payload, increasing the risk of malicious activity.
The npm package claims-microgen-templates version 35.7.1 contains malicious code that loads a platform-specific binary from obfuscated Cloudflare Workers hosts and fallback domains. The payload is saved to temporary directories with randomized filenames, made executable, and run detached via shell commands. This behavior is unrelated to the package's stated purpose and lacks any integrity verification or publisher validation.
The npm package 'claims-person-api-adapters' version 35.2.4 contains malicious code that executes on import. It reconstructs obfuscated network destinations to download and execute attacker-controlled binaries without verification. The payload is saved under misleading filenames and executed detached on the host system, enabling remote code execution. This behavior is hidden behind a seemingly benign package name and uses evasion techniques such as URL reconstruction and DNS-TXT fallback channels.
The npm package claims-page-maker-types version 35.5.1 contains malicious code that, upon requiring the package, loads a module which fetches and executes a platform-specific binary from hardcoded external Cloudflare worker domains and a fallback DNS TXT record. The binary is saved to a disguised temporary path, given executable permissions, and spawned detached from the main process. The code attempts to evade detection by assembling host strings from fragments and mimicking system diagnostic filenames. There is no integrity verification of the downloaded payload, and the external hosts are unrelated to the package publisher. Execution is throttled by a local stamp file to limit repeated runs.
The npm package claims-patch-url version 35.3.9 contains malicious code that downloads and executes a binary payload from multiple Cloudflare Worker hosts. The package's main API is an empty shell class, with the actual functionality being a dropper that writes and executes a binary under disguised filenames in temporary directories. The payload download uses covert DNS-TXT fallback channels and runtime hostname assembly to evade detection. Environment variables are checked to gate execution, and the package appears to be a typosquat or decoy with no legitimate dependencies.
The npm package 'claims-payout-offer-domain' version 35.6.1 contains malicious code that, upon being required, downloads and executes an opaque binary payload from obfuscated external sources. The payload is fetched without integrity verification and executed with elevated permissions, potentially compromising the host system.
The npm package 'claims-policy-details-domain' version 35.9.3 contains malicious code that downloads and executes platform-specific binaries from anonymous Cloudflare worker subdomains. It uses a DNS TXT record covert channel fallback to retrieve payload chunks, which are then saved to temporary directories and executed with elevated permissions. The package is likely a dependency confusion lure with minimal legitimate functionality, designed primarily as a dropper for malicious payloads.
The npm package 'claims-person-domain' version 35.5.9 contains malicious code that downloads and executes opaque binaries from multiple Cloudflare Workers endpoints and a DNS-TXT covert channel fallback. The binaries are saved under disguised filenames in temporary directories, given executable permissions, and executed detached from the main process. The package also includes a parallel dropper in telemetry.js that decodes and executes a base64 payload similarly. The obfuscation techniques and disguised filenames mimic diagnostic tools and serve no legitimate purpose.
The npm package 'claims-use-visual-viewport-resizing' version 35.2.8 contains malicious code that, upon requiring the package, loads a helper script which automatically executes a function to fetch and run a binary payload. This payload is retrieved from dynamically constructed hostnames designed to evade static detection and is saved to temporary directories under misleading filenames before being executed on both POSIX and Windows systems.
The npm package 'clients-structure' version 35.9.8 contains malicious code that, upon require(), loads a platform-specific script which fetches and executes a binary payload from anonymous Cloudflare Workers subdomains and fallback DNS-TXT records. The payload is reconstructed from base64 chunks, saved to a temporary path with executable permissions, and executed without any hash or signature verification. The malicious activity uses runtime-assembled hostnames and employs a TTL marker and environment variable facade as a cover story. This behavior indicates a supply chain compromise with arbitrary code execution risk.
The npm package 'cobrowsing-decorators' version 35.2.9 contains malicious code that, upon import, downloads and executes an opaque binary from obfuscated Cloudflare Workers endpoints and fallback domains. The binary is saved with disguised filenames and executed without signature or hash verification. The package's telemetry component also contains redundant code to drop and execute the payload. This behavior is unrelated to the package's declared purpose and represents a supply chain compromise.
The npm package claims-use-base-analytics@35.4.9 is identified as a malicious native-binary dropper disguised as an analytics wrapper. It fetches and executes a platform-specific binary from attacker-controlled Cloudflare Workers or, if that fails, reconstructs and executes a payload via a DNS-over-TXT covert channel. The package uses obfuscation techniques and respects environment variables to evade detection. Its advertised analytics functionality does not justify the presence of such malicious behavior.
The npm package 'claims-types' version 35.8.3 contains malicious code that downloads and executes binaries from Cloudflare Workers hosts or via a DNS TXT record covert channel. The malicious payload is saved under disguised filenames in temporary directories, made executable, and spawned detached from the main process. This behavior is designed to evade static detection by assembling hostnames and API calls dynamically and using cover stories like 'analytics' and 'telemetry'. A secondary dropper module with similar behavior is present but not actively used in this version.