@kg-ai/kugou-skill@0.1.7

High-risk behavior combination matched malicious policy.

npm installation

Installed agents may automatically load package-controlled instructions, including secret-handling and update directives.

postinstall cross-agent skill injection

Installing the package runs scripts/install.js without a user command. It detects existing foreign agent skill roots, creates a kugou-skill subdirectory, and copies the package-controlled SKILL.md and references into each. The injected instructions tell agents to solicit and persist user-provided authentication secrets, while the opaque bundled binary contains automatic-update functionality. This is an unconsented broad AI-agent control-surface mutation at install time.

Direct source inspection confirms an unconsented postinstall mutation of six foreign AI-agent skill surfaces. This meets the firewall block threshold regardless of the CLI’s stated music purpose.