npm detections from LPM Firewall.

The import-time remote payload execution chain is explicit in source and requires no user setup action. The unused-looking telemetry module does not mitigate this behavior.

This is concrete import-time remote payload execution, unrelated to the declared native-bridge API. Lack of an npm lifecycle hook does not mitigate execution on ordinary package import.

Source inspection confirms an import-time remote payload download-and-execute chain unrelated to the declared mobile SDK function. The lack of npm lifecycle hooks does not mitigate the re...

This is unconsented postinstall mutation of a broad, foreign AI-agent control surface. The benign-looking statusline purpose does not remove the persistent global command injection.

This is concrete, unconsented postinstall mutation of a broad foreign AI-agent control surface. The observed lifecycle behavior meets the firewall block boundary even without confirmed da...

The install hook implements concrete, unconsented mutation of broad foreign AI-agent skill locations. Absence of exfiltration does not remove that install-time control-surface risk.

Concrete default telemetry exfiltration to a package-controlled external store is established by source, and is compounded by unconsented postinstall host mutation. Scanner similarity lab...

The install-time mutation of Claude Desktop configuration is concrete and unconsented. This meets the blocking policy even though the runtime editor functionality appears package-aligned.

The confirmed install-time Claude Desktop configuration mutation meets the block policy for a foreign/broad AI-agent control surface. Other shell and network features are runtime video-ed...

Source inspection confirms a postinstall hook that mutates a broad set of foreign AI-agent configuration files. This meets the install-control-surface block criterion regardless of the op...

Direct source inspection confirms a concrete, automatic postinstall AI-agent control-surface takeover. The absence of exfiltration does not mitigate the unconsented cross-tool persistence.

The lifecycle source establishes an unconsented postinstall mutation of foreign, broad AI-agent control surfaces. This meets the blocking policy despite no observed network exfiltration.

The postinstall hook performs unconsented mutation of a broad, foreign AI-agent control surface. This meets the blocking policy regardless of the package's stated browser-automation purpose.

The concrete postinstall fan-out writes behavior-bearing assets into foreign/broad AI-agent control surfaces without explicit user invocation. This meets the install-control-surface block...

Source confirms unconsented postinstall mutation of foreign/broad AI-agent configuration surfaces. This meets the install-control-surface blocking policy despite no confirmed network exfi...

Direct source inspection confirms unconsented postinstall mutation of broad, foreign AI-agent control surfaces. This meets the install-control-surface blocking policy regardless of the ab...

This is concrete, import-triggered remote code execution rather than declared telemetry functionality. The lack of an install hook does not mitigate execution during ordinary use.

This is concrete import-time remote binary retrieval and execution, unrelated to the stated token-provider function. No lifecycle hook is needed because the entrypoint triggers the payloa...

Direct source inspection confirms a reachable import-time downloader and executor, not merely telemetry behavior. The lack of install hooks does not mitigate runtime execution on package...

This is an unconsented import-time staged payload chain culminating in arbitrary native code execution. The harmless-looking public API and unused telemetry source do not mitigate that co...

Source confirms default runtime transmission of rich host application records to hard-coded external webhook destinations. The benign SDK structure and lack of install-time execution do n...