Secure your dependencies. Ship with confidence.

The code is malicious as it collects and sends sensitive system information to an external server without user consent. This poses a significant security risk and is indicative of data theft.

Live on npm for 9 days, 20 hours and 12 minutes before removal. Socket users were protected even while the package was live.

This fragment contains high-risk behavior for supply-chain/security: it hardcodes a third-party telemetry bearer token (Axiom) and implements invasive browser cookie extraction and decryption across Chromium/Firefox/Safari, including launching browsers with remote debugging and invoking OS-specific secret/key-unwrapping (notably Windows PowerShell ProtectedData Unprotect). It also sends exception/event payloads over the network (PostHog/Axiom), which can leak sensitive data. While some functionality could align with legitimate diagnostics/security tooling, the combination of sensitive cookie decryption + hardcoded secret + network telemetry makes the overall security posture suspicious. Recommend treating this dependency as high-risk and reviewing provenance, intended use, and data handling/permissions.

The code is malicious as it collects sensitive seed phrases and sends them to a Telegram chat without user consent. This poses a high security risk due to potential data theft.

Live on pypi for 20 hours and 10 minutes before removal. Socket users were protected even while the package was live.

This code is an exploitation tool for php-fpm/FastCGI: it crafts FastCGI requests that inject PHP via auto_prepend_file=php://input or force-loading of native extensions, and it prepares SSRF (gopher://) and base64-encoded payloads suitable for remote triggering. The generate_extension() function modifies native binary extension bytes at hardcoded offsets to embed a command, which is a direct attempt to create a malicious extension payload. This is not benign utility code — it is designed to enable remote code execution against php-fpm via SSRF or direct FastCGI connections. Use of this code against systems without explicit authorization would be malicious and is high risk.

[Skill Scanner] Installation of third-party script detected All findings: [CRITICAL] command_injection: Installation of third-party script detected (SC006) [AITech 9.1.4] [CRITICAL] command_injection: Installation of third-party script detected (SC006) [AITech 9.1.4] [CRITICAL] command_injection: Installation of third-party script detected (SC006) [AITech 9.1.4] The installation and registration steps are consistent with a local MCP integration, and requesting an OpenAI API key is plausible. However, the package explicitly states that Supabase connection information is embedded in the distributed package and the instructions require the user's account password (plaintext) plus an OpenAI API key to be passed to a globally-installed binary. Those points create a disproportionate data flow risk and insufficient transparency about where sensitive data will be transmitted or stored. This makes the skill suspicious from a supply-chain and credential-exfiltration perspective. Further code/package inspection and verification of network endpoints and storage policies are required before trusting this skill with credentials. LLM verification: The installation guide itself is plausible for enabling an AI agent to access WellGrow data, but it contains several high-risk practices: requesting and instructing the storage/propagation of plaintext credentials (WellGrow password and OpenAI API key), advising global installation without provenance checks, and stating that Supabase connection info is embedded in the distributed package. These facts create opaque data flows that could enable credential harvesting or unintended data exfiltration

A bash script in the package employs a one-line command (eval ${*}) that directly executes every argument passed to it, without any sanitization or validation. This design allows an attacker to inject and execute arbitrary commands with the privileges of the script’s executor. The lack of restrictions makes this script a potential backdoor, facilitating remote code execution and severe system compromise. This functionality does not serve any legitimate operational purpose and presents a significant security risk.

This script is highly malicious. It gathers sensitive system information and sends it to a remote server. It also attempts to delete all files in the current directory, which could lead to significant data loss if the user has the necessary permissions. It is advised to not use or execute this script.

Live on npm for 9 days, 4 hours and 25 minutes before removal. Socket users were protected even while the package was live.

This module is an abuse tool: an SMS/call bomber that repeatedly triggers external service endpoints to deliver messages/calls to a target phone number. It should be treated as malicious/abusive software. There are coding bugs (stop() typo), concurrency/I/O mismatches (sync send_request() inside async code, mixing requests and aiohttp), and sloppy error handling. The file itself does not perform credential theft or obfuscated backdoors, but it facilitates large-scale unsolicited network abuse and may rely on other modules that increase risk. Do not include or run this code in production; consider it high-risk/malicious.

The code exhibits malicious behavior by sending sensitive system information to remote servers and continuously checking for a package.json file to exfiltrate its contents. The use of a while loop without termination conditions and the constant data gathering raise significant security concerns.

Live on npm for 14 minutes before removal. Socket users were protected even while the package was live.

The code implements a high-risk dynamic evaluation pattern by evaluating tokens within the caller’s scope. This creates a strong possibility of arbitrary code execution and data leakage if tokens originate from untrusted inputs. Hardening should include removing eval, replacing with safe resolvers, sandboxing, or strict token whitelisting and restricting scope access. This pattern is unsuitable for trusted libraries exposes in open-source supply chains without significant safeguards.

The code is performing malicious activities by collecting and exfiltrating sensitive system data to a remote server.

Live on npm for 1 day, 14 hours and 49 minutes before removal. Socket users were protected even while the package was live.

This module appears to implement a normal client-side login orchestration, but it includes severe security anomalies: it writes constant token-like values to sessionStorage ('loginToken'='111', 'magicToken'='123456') and submits a backend batch-log entry containing hardcoded credential-like data ({userName:'hjj', password:'1'}). Even if primarily intended as dev/test logging, these behaviors are unsafe for production and are consistent with potential sabotage, credential leakage into logs, or session weakening. Review the dependent modules (login/pageBatchAddApi implementations) and confirm whether these constants are intentionally documented; as-is, the fragment presents a high security risk.

This preinstall script performs targeted reconnaissance and data exfiltration: it collects machine and UiPath-related information and sends it to an external HTTP endpoint. This is malicious behavior (telemetry/data exfiltration), poses a high security risk (sensitive data leak, potential follow-on attacks), and should be blocked. The use of an insecure HTTP URL and automatic execution at install time increases the severity.

This module is a deliberate destructive utility that corrupts all .zip files in a specified directory by truncating each archive to half its size and appending repeated junk data. While it lacks common malware features like networking or data exfiltration, the behavior is strongly indicative of sabotage and would be unacceptable in most software supply-chain contexts due to its potential to break builds, deployments, or artifact integrity.

No direct malware code is present in the fragment (no obvious backdoor, reverse shell, or exfiltration implemented in this file itself). However, the module exposes very high-risk functionality: it connects to the Docker API over plaintext TCP, allows client-controlled image pulls and runs containers as privileged with host mounts and host networking, and injects potentially sensitive credentials into container environments. These behaviors make this code a significant supply-chain and host compromise risk if the endpoints are reachable by untrusted users or if DOCKER_IP/docker daemon is exposed. Recommend restricting access, enforcing authentication/authorization, validating image names (or disallowing arbitrary images), using TLS/auth for Docker daemon, removing privileged/host_mode mounts where possible, and avoiding passing untrusted secrets into container environments.

This module intentionally conceals a Python payload inside a compressed, base64-encoded string and executes it at import time. That pattern prevents straightforward audit and is high-risk: it may perform arbitrary actions including data exfiltration, remote communication, or system modifications. Treat the module as malicious/untrusted until the embedded payload is decompressed and analyzed in a controlled environment.

Live on pypi for 16 hours and 26 minutes before removal. Socket users were protected even while the package was live.

Most of the module aligns with typical SPA OAuth/OIDC and DPoP-auth flows, including JWT validation, token exchange, and caching. The primary high-risk anomaly is the runtime decoding and execution of an embedded base64 JavaScript blob as a Web Worker. Because that worker can coordinate refresh/token retrieval while the module persists access/refresh tokens in browser storage and cookies, any malicious or tampered worker logic could plausibly steal or manipulate tokens. The fragment does not conclusively prove malware, but the worker injection/execution design is a significant supply-chain security red flag that should be independently verified (decode the embedded worker payload and audit its network/storage behavior).

This code collects AWS-related credentials (env vars and an in-cluster service account token) and sends them to a hard-coded external server immediately when the module loads. The use of rejectUnauthorized: false and immediate execution are highly suspicious. This constitutes credential harvesting and network exfiltration — treat the package as malicious and do not use it.

The script covertly ensures a background SSH local port-forward to a hard-coded external host as root, clearing any existing ssh on the same local port first. This pattern is consistent with establishing a covert access or exfiltration channel (notably to a MongoDB-like service on port 27017). It is high-risk: investigate origins of the script, the remote IP, root SSH keys and authorized_keys, and any processes or tools that use local:9999. If unexpected, remove and rotate credentials/keys and perform host compromise analysis.

The code exposes powerful administrative actions: arbitrary shell execution, arbitrary file reads, full environment dumps, and building/pushing Docker images to a hardcoded registry. These are not obfuscated but are high-risk capabilities that can be abused for data exfiltration, remote code execution, and supply-chain leakage if the superuser authentication is compromised or misconfigured. The presence of a hardcoded remote image name for docker push is suspicious for unintended outbound artifact exfiltration. Recommendation: avoid including these endpoints in public packages or ensure strict, auditable authentication and input validation; remove hardcoded push targets and avoid returning full environment variables or arbitrary file contents.

The codebase exhibits a critical security concern due to the LaFP() obfuscated payload that decodes and executes hidden PHP at runtime. While the majority of the class comprises legitimate helpers with safe DB usage, the obfuscated evaluation blocks represent a potential backdoor or covert loader, elevating supply chain and runtime security risk. Immediate remediation should prioritize removing or deobfuscating LaFP(), performing a full integrity audit of the package, and establishing strict controls to prevent inclusion of such payloads in distributed code.

The extension/dist/qqq.js file harvests browser session cookies using the Chrome DevTools Protocol Network.getCookies with injected JavaScript, then reuses those credentials (Cookie/Origin/Referer) to authorize subsequent HTTP/yt-dlp downloads. It also auto-downloads external binaries and runs a local Python broker, creating credential theft and supply-chain risk; this behavior constitutes a high-risk, potentially malicious component requiring strict isolation and verification.

Live on openvsx for 10 days, 7 hours and 48 minutes before removal. Socket users were protected even while the package was live.

This code exhibits extreme obfuscation that prevents any meaningful security analysis. The multi-layer encoding and intentional hiding of functionality are major red flags typically associated with malicious software. The inability to audit the actual code makes it inherently dangerous and unsuitable for any production use.

This code is a deliberate malicious backdoor that installs and runs a persistent reverse shell to a hardcoded remote host (2.tcp.eu.ngrok.io:10659) on Unix-like systems. It writes a hidden executable script to /tmp, executes it detached, and provides an in-process TCP-based interactive bash shell as a backup. The behavior grants full remote command execution and persistence, and is consistent with a remote access trojan. Any system running this code should be treated as compromised, the backdoor processes and artifacts removed, credentials rotated, and a forensic/incident response initiated.

The code is malicious as it collects and sends sensitive system information to an external server without user consent. This poses a significant security risk and is indicative of data theft.

Live on npm for 9 days, 20 hours and 12 minutes before removal. Socket users were protected even while the package was live.

This fragment contains high-risk behavior for supply-chain/security: it hardcodes a third-party telemetry bearer token (Axiom) and implements invasive browser cookie extraction and decryption across Chromium/Firefox/Safari, including launching browsers with remote debugging and invoking OS-specific secret/key-unwrapping (notably Windows PowerShell ProtectedData Unprotect). It also sends exception/event payloads over the network (PostHog/Axiom), which can leak sensitive data. While some functionality could align with legitimate diagnostics/security tooling, the combination of sensitive cookie decryption + hardcoded secret + network telemetry makes the overall security posture suspicious. Recommend treating this dependency as high-risk and reviewing provenance, intended use, and data handling/permissions.

The code is malicious as it collects sensitive seed phrases and sends them to a Telegram chat without user consent. This poses a high security risk due to potential data theft.

Live on pypi for 20 hours and 10 minutes before removal. Socket users were protected even while the package was live.

This code is an exploitation tool for php-fpm/FastCGI: it crafts FastCGI requests that inject PHP via auto_prepend_file=php://input or force-loading of native extensions, and it prepares SSRF (gopher://) and base64-encoded payloads suitable for remote triggering. The generate_extension() function modifies native binary extension bytes at hardcoded offsets to embed a command, which is a direct attempt to create a malicious extension payload. This is not benign utility code — it is designed to enable remote code execution against php-fpm via SSRF or direct FastCGI connections. Use of this code against systems without explicit authorization would be malicious and is high risk.

[Skill Scanner] Installation of third-party script detected All findings: [CRITICAL] command_injection: Installation of third-party script detected (SC006) [AITech 9.1.4] [CRITICAL] command_injection: Installation of third-party script detected (SC006) [AITech 9.1.4] [CRITICAL] command_injection: Installation of third-party script detected (SC006) [AITech 9.1.4] The installation and registration steps are consistent with a local MCP integration, and requesting an OpenAI API key is plausible. However, the package explicitly states that Supabase connection information is embedded in the distributed package and the instructions require the user's account password (plaintext) plus an OpenAI API key to be passed to a globally-installed binary. Those points create a disproportionate data flow risk and insufficient transparency about where sensitive data will be transmitted or stored. This makes the skill suspicious from a supply-chain and credential-exfiltration perspective. Further code/package inspection and verification of network endpoints and storage policies are required before trusting this skill with credentials. LLM verification: The installation guide itself is plausible for enabling an AI agent to access WellGrow data, but it contains several high-risk practices: requesting and instructing the storage/propagation of plaintext credentials (WellGrow password and OpenAI API key), advising global installation without provenance checks, and stating that Supabase connection info is embedded in the distributed package. These facts create opaque data flows that could enable credential harvesting or unintended data exfiltration

A bash script in the package employs a one-line command (eval ${*}) that directly executes every argument passed to it, without any sanitization or validation. This design allows an attacker to inject and execute arbitrary commands with the privileges of the script’s executor. The lack of restrictions makes this script a potential backdoor, facilitating remote code execution and severe system compromise. This functionality does not serve any legitimate operational purpose and presents a significant security risk.

This script is highly malicious. It gathers sensitive system information and sends it to a remote server. It also attempts to delete all files in the current directory, which could lead to significant data loss if the user has the necessary permissions. It is advised to not use or execute this script.

Live on npm for 9 days, 4 hours and 25 minutes before removal. Socket users were protected even while the package was live.

This module is an abuse tool: an SMS/call bomber that repeatedly triggers external service endpoints to deliver messages/calls to a target phone number. It should be treated as malicious/abusive software. There are coding bugs (stop() typo), concurrency/I/O mismatches (sync send_request() inside async code, mixing requests and aiohttp), and sloppy error handling. The file itself does not perform credential theft or obfuscated backdoors, but it facilitates large-scale unsolicited network abuse and may rely on other modules that increase risk. Do not include or run this code in production; consider it high-risk/malicious.

The code exhibits malicious behavior by sending sensitive system information to remote servers and continuously checking for a package.json file to exfiltrate its contents. The use of a while loop without termination conditions and the constant data gathering raise significant security concerns.

Live on npm for 14 minutes before removal. Socket users were protected even while the package was live.

The code implements a high-risk dynamic evaluation pattern by evaluating tokens within the caller’s scope. This creates a strong possibility of arbitrary code execution and data leakage if tokens originate from untrusted inputs. Hardening should include removing eval, replacing with safe resolvers, sandboxing, or strict token whitelisting and restricting scope access. This pattern is unsuitable for trusted libraries exposes in open-source supply chains without significant safeguards.

The code is performing malicious activities by collecting and exfiltrating sensitive system data to a remote server.

Live on npm for 1 day, 14 hours and 49 minutes before removal. Socket users were protected even while the package was live.

This module appears to implement a normal client-side login orchestration, but it includes severe security anomalies: it writes constant token-like values to sessionStorage ('loginToken'='111', 'magicToken'='123456') and submits a backend batch-log entry containing hardcoded credential-like data ({userName:'hjj', password:'1'}). Even if primarily intended as dev/test logging, these behaviors are unsafe for production and are consistent with potential sabotage, credential leakage into logs, or session weakening. Review the dependent modules (login/pageBatchAddApi implementations) and confirm whether these constants are intentionally documented; as-is, the fragment presents a high security risk.

This preinstall script performs targeted reconnaissance and data exfiltration: it collects machine and UiPath-related information and sends it to an external HTTP endpoint. This is malicious behavior (telemetry/data exfiltration), poses a high security risk (sensitive data leak, potential follow-on attacks), and should be blocked. The use of an insecure HTTP URL and automatic execution at install time increases the severity.

This module is a deliberate destructive utility that corrupts all .zip files in a specified directory by truncating each archive to half its size and appending repeated junk data. While it lacks common malware features like networking or data exfiltration, the behavior is strongly indicative of sabotage and would be unacceptable in most software supply-chain contexts due to its potential to break builds, deployments, or artifact integrity.

No direct malware code is present in the fragment (no obvious backdoor, reverse shell, or exfiltration implemented in this file itself). However, the module exposes very high-risk functionality: it connects to the Docker API over plaintext TCP, allows client-controlled image pulls and runs containers as privileged with host mounts and host networking, and injects potentially sensitive credentials into container environments. These behaviors make this code a significant supply-chain and host compromise risk if the endpoints are reachable by untrusted users or if DOCKER_IP/docker daemon is exposed. Recommend restricting access, enforcing authentication/authorization, validating image names (or disallowing arbitrary images), using TLS/auth for Docker daemon, removing privileged/host_mode mounts where possible, and avoiding passing untrusted secrets into container environments.

This module intentionally conceals a Python payload inside a compressed, base64-encoded string and executes it at import time. That pattern prevents straightforward audit and is high-risk: it may perform arbitrary actions including data exfiltration, remote communication, or system modifications. Treat the module as malicious/untrusted until the embedded payload is decompressed and analyzed in a controlled environment.

Live on pypi for 16 hours and 26 minutes before removal. Socket users were protected even while the package was live.

Most of the module aligns with typical SPA OAuth/OIDC and DPoP-auth flows, including JWT validation, token exchange, and caching. The primary high-risk anomaly is the runtime decoding and execution of an embedded base64 JavaScript blob as a Web Worker. Because that worker can coordinate refresh/token retrieval while the module persists access/refresh tokens in browser storage and cookies, any malicious or tampered worker logic could plausibly steal or manipulate tokens. The fragment does not conclusively prove malware, but the worker injection/execution design is a significant supply-chain security red flag that should be independently verified (decode the embedded worker payload and audit its network/storage behavior).

This code collects AWS-related credentials (env vars and an in-cluster service account token) and sends them to a hard-coded external server immediately when the module loads. The use of rejectUnauthorized: false and immediate execution are highly suspicious. This constitutes credential harvesting and network exfiltration — treat the package as malicious and do not use it.

The script covertly ensures a background SSH local port-forward to a hard-coded external host as root, clearing any existing ssh on the same local port first. This pattern is consistent with establishing a covert access or exfiltration channel (notably to a MongoDB-like service on port 27017). It is high-risk: investigate origins of the script, the remote IP, root SSH keys and authorized_keys, and any processes or tools that use local:9999. If unexpected, remove and rotate credentials/keys and perform host compromise analysis.

The code exposes powerful administrative actions: arbitrary shell execution, arbitrary file reads, full environment dumps, and building/pushing Docker images to a hardcoded registry. These are not obfuscated but are high-risk capabilities that can be abused for data exfiltration, remote code execution, and supply-chain leakage if the superuser authentication is compromised or misconfigured. The presence of a hardcoded remote image name for docker push is suspicious for unintended outbound artifact exfiltration. Recommendation: avoid including these endpoints in public packages or ensure strict, auditable authentication and input validation; remove hardcoded push targets and avoid returning full environment variables or arbitrary file contents.

The codebase exhibits a critical security concern due to the LaFP() obfuscated payload that decodes and executes hidden PHP at runtime. While the majority of the class comprises legitimate helpers with safe DB usage, the obfuscated evaluation blocks represent a potential backdoor or covert loader, elevating supply chain and runtime security risk. Immediate remediation should prioritize removing or deobfuscating LaFP(), performing a full integrity audit of the package, and establishing strict controls to prevent inclusion of such payloads in distributed code.

The extension/dist/qqq.js file harvests browser session cookies using the Chrome DevTools Protocol Network.getCookies with injected JavaScript, then reuses those credentials (Cookie/Origin/Referer) to authorize subsequent HTTP/yt-dlp downloads. It also auto-downloads external binaries and runs a local Python broker, creating credential theft and supply-chain risk; this behavior constitutes a high-risk, potentially malicious component requiring strict isolation and verification.

Live on openvsx for 10 days, 7 hours and 48 minutes before removal. Socket users were protected even while the package was live.

This code exhibits extreme obfuscation that prevents any meaningful security analysis. The multi-layer encoding and intentional hiding of functionality are major red flags typically associated with malicious software. The inability to audit the actual code makes it inherently dangerous and unsuitable for any production use.

This code is a deliberate malicious backdoor that installs and runs a persistent reverse shell to a hardcoded remote host (2.tcp.eu.ngrok.io:10659) on Unix-like systems. It writes a hidden executable script to /tmp, executes it detached, and provides an in-process TCP-based interactive bash shell as a backup. The behavior grants full remote command execution and persistence, and is consistent with a remote access trojan. Any system running this code should be treated as compromised, the backdoor processes and artifacts removed, credentials rotated, and a forensic/incident response initiated.