attack-surface-recon

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt explicitly instructs running commands that search for and print likely secrets (grep for 'secret|key|token|password|aws') and includes an example cookie hash, so the agent would capture and output secret values verbatim.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This document is an explicit offensive reconnaissance and exploitation playbook (scanning, credential harvesting, evasion, brute‑forcing and backend exploitation), showing clear malicious intent and actionable steps for abuse.

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

• Third-party content exposure detected (medium risk: 0.30). 该流程在“侦察/攻击面测绘”中通过浏览器导航与多处curl/httpx/nuclei等步骤直接抓取并读取外部可被第三方投毒的文本与响应（例如 crt.sh、web.archive.org、fofa、shodan 以及对目标域名的HTTP响应体），因此存在把外人可控免费文本喂给运行时LLM/Agent的风险。