supply-chain-attacker

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The prompt explicitly instructs creating malicious install scripts that read environment variables and exfiltrate them (e.g., concatenating process.env into an HTTPS request), which directs an agent to access and embed secret values verbatim in outgoing requests.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This document provides explicit, actionable instructions for performing supply-chain attacks—dependency confusion, typosquatting, malicious install scripts (preinstall/postinstall) that exfiltrate credentials, reverse shells, repository/CI compromise, and build backdoors—constituting deliberate malicious behavior and high-risk abuse patterns.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 1.00). The skill explicitly instructs creating and publishing malicious packages, adding install/preinstall scripts that exfiltrate or open shells, tampering with repositories/CI (force-push, malicious PRs), and other actions that change system or remote state and enable compromise, so it clearly pushes offensive, state-changing behavior and must be flagged.