webshell-deploy

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This skill is an explicit malicious playbook for deploying and maintaining webshells and in-memory backdoors—providing ready-to-use RCE payloads (JSP/PHP/ASPX), upload/PUT bypasses, LFI/log injection, reverse-shells, Java memory shell/agent injection, obfuscation/“免杀” techniques and persistence/cleanup instructions—clearly intended to enable unauthorized remote command execution, data access/exfiltration, credential theft and persistent compromise.

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md explicitly instructs making curl requests to arbitrary TARGET URLs (Phase 3 upload examples and Phase 4 verification) and to inspect HTTP status codes and command output from those remote hosts, meaning the agent will fetch and act on untrusted third‑party content that can influence subsequent actions.