npm security reports, with evidence.
Investigate malicious npm packages and AI-agent policy findings. Each report identifies the package version, supporting evidence, advisory sources, and current Firewall decision. Malware findings and policy violations are labelled separately.
Security reports — page 672
Updated 28 Oct 2026, 01:24 UTC. Ordered by latest scan.
Direct source inspection confirms a concealed, import-time remote-command execution chain. The lack of install hooks does not mitigate arbitrary command execution during ordinary package...
The package contains a concrete, automatically triggered remote execution chain with stealth and system-modification behavior unrelated to its stated WordPress functionality. Empty Telegr...
This is a concrete import-time remote command execution chain, not a package-aligned CSS-processing feature. Lack of an install hook does not mitigate execution when the package is import...
Direct inspection confirms concrete, automatic credential exfiltration in the package manifest. This is not package-aligned functionality and meets the blocking threshold.
The package's advertised transformer functionality is absent, while its automatic lifecycle hook retrieves and executes a remotely controlled payload. This is concrete install-time remote...
Direct inspection of the sole shipped file confirms concrete install-time credential harvesting and external exfiltration. This is malicious behavior, not a package-aligned feature.
Static source inspection confirms unconsented install-time and import-time environment harvesting followed by network transmission. Conditional backend configuration does not remove the c...
Direct source inspection confirms unconsented postinstall credential harvesting and remote exfiltration. This is concrete malicious behavior, not a scanner-only inference.
Direct inspection confirms install-time collection and remote transmission of the installer environment. No legitimate package functionality is present to justify this behavior.
The package contains a concrete, externally directed runtime exfiltration path for account data and potentially IMAP passwords. Benign local lifecycle and explicit MCP behavior do not mit...
Source inspection confirms a runtime network path that sends account configuration, including a field used as the IMAP password, to a third-party logging host. This is concrete unconsente...
The package contains concrete import-time arbitrary payload execution in its published Node entrypoints. The behavior is concealed in build artifacts and is unrelated to the library's blo...
The package's real export path introduces obfuscated system reconnaissance and remote-capable execution primitives that are not required for Chai assertions. This is malicious import-time...
The package contains an undocumented, unconditionally triggered detached subprocess and database-connection payload unrelated to its stated testing utility. This is concrete malicious run...
The package contains a concrete, automatically triggered hidden child-process payload unrelated to its advertised testing functionality. The added dependency and suppressed child output r...
The package contains an unrelated, concealed child-process payload invoked during normal plugin initialization. Its missing `spawn` import makes the attempt fail in standard Node, but doe...
This is concrete import-time arbitrary-code execution from externally staged content, not a package-aligned SDK function. The absence of the loader in source makes the shipped distributio...
The package contains a concrete, stealthy runtime chain that fetches and executes attacker-controlled remote JavaScript. Absence of an install hook does not mitigate arbitrary code execut...
Direct source inspection confirms an automatically triggered credential/configuration exfiltration chain plus attempted SSH persistence. The hard-coded external endpoints and remote-contr...
The package's benign exported helpers do not justify a postinstall flow that retrieves and executes mutable remote code. This is concrete install-time remote code execution.
Direct source inspection confirms an obfuscated install-time reverse-shell payload with no legitimate implementation to justify it. The package is malicious and should be blocked.
Direct source inspection confirms import-time reverse-shell behavior and outbound host callback. This is concrete unauthorized remote command execution, not merely a scanner match.
Direct inspection of the sole package file confirms an install-time credential-harvesting payload with third-party network exfiltration. This is malicious behavior, not a package-aligned...
Direct source inspection confirms import-time reverse-shell behavior in the package main entrypoint. The lack of lifecycle scripts does not mitigate execution when the entrypoint is impor...
Direct inspection of the sole packaged file confirms automatic install-time credential harvesting and external exfiltration. This is concrete malicious behavior, not a scanner-only infere...