vmware-vcenter-attack

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The skill shows examples that embed plaintext credentials and session tokens into CLI commands (e.g., curl -u 'user: ' and -H "vmware-api-session-id: "), which would require an LLM to include secret values verbatim if those placeholders are instantiated — an exfiltration risk.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This document is a high-risk offensive playbook: it provides explicit, actionable instructions for unauthorised reconnaissance, pre-auth remote code execution, backdoor/webshell installation, credential abuse, and data exfiltration targeting VMware infrastructure.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). SKILL.md describes an LLM-driven runtime workflow that directly ingests untrusted HTTP responses/bodies from outsider-controlled targets (e.g., curl ... /ui/login, /sdk/vimServiceVersions.xml, /sso-adminserver/..., /websso/.../Metadata/..., and other probe endpoints) and uses those contents for version/CVE decisions.