database-tactics

该技能与其声明用途高度一致：它就是一份数据库攻击/提权方法论，而不是伪装成其他用途的窃密技能。未见隐藏外传、可疑中转域名、预执行指令或恶意下载执行链，因此不像恶意软件；但它显式赋予 AI 代理进行爆破、RCE、文件读写和提权的攻击能力，属于高风险的 offensive security skill。

This fragment is not a software module; it is directly actionable offensive documentation for compromising Redis. It describes multiple high-impact attack paths (unauthorized access and brute forcing, replication-based module RCE, Lua sandbox escape via EVAL, and enumeration/extraction). If included in a dependency package, it is a strong malicious-content supply-chain red flag. No embedded malware behavior can be confirmed from this text alone, but the intent and operational exploit guidance are clear.

This artifact is not a software module; it is an attacker-focused MSSQL exploitation guide describing credential guessing, SQL-to-OS command execution (xp_cmdshell/CLR/Agent/linked servers), file read/write via SQL primitives, privilege escalation, and NTLM hash harvesting through UNC authentication to an external listener. It contains no code obfuscation, but it is highly actionable for unauthorized compromise and credential theft. In a supply-chain context, its presence in a dependency/package repository would be a major security governance red flag even though it is not itself a deployable malware payload.

This fragment is an explicit offensive PostgreSQL compromise guide, describing credential brute force/auth bypass, potential RCE via PostgreSQL command-execution features (COPY FROM PROGRAM / procedural language/UDF paths), filesystem read/write via database primitives, sensitive data extraction (including password hash references), privilege escalation, and Windows NTLM hash capture via UNC-trigger + Responder. If this content appears in a repository or dependency, it is a strong indicator of malicious intent and should be treated as a severe supply-chain security concern.