渗透测试Agent操作系统
SkillSearchOnce added, your AI can run full penetration testing engagements on its own. Instead of following a fixed checklist, it records verified facts about the target as it goes and works out attack paths from the evidence it actually confirms. Testing can cover web apps, authentication, cloud, networks, wireless, hardware, and more.
Available today. Use it from your connected AI after setup.
No other account needed.
After adding it, ask your AI to start a full-chain penetration test engagement, or request the skill map first if you want to see everything the suite covers.
Then ask your AI: use the 渗透测试Agent操作系统 skill
What your AI can do with it
- Run a full-chain penetration test from first reconnaissance through exploitation
- Build a growing map of verified facts so every attack path is backed by real evidence
- Test across web, authentication, cloud, internal networks, wireless, and hardware
- Record vulnerabilities as findings throughout the engagement
- Derive attack chains specific to the target when well-known techniques turn up nothing
- Pull up the full skill map for the suite when planning an engagement
What this skill tells your AI
The instructions your AI receives, as published by aipentest/cyberstrikeai in skills/pentest-agent-os/SKILL.md and read by ahel’s review.
渗透不是执行脚本,是搜索路径。 不给自己预设固定路径/角色/已知利用链。给定起点与目标, 路径未知 → 在状态空间里搜索:不断试不同路径、试错、反馈、逼近目标。 跳出固有攻击链 = 不问"它有没有已知CVE",而问"基于项目黑板上已掌握的事实,下一步最有效的动作是什么"。 状态落点见
pentest-blackboard（SQLite 项目事实，非磁盘 Intent/Hint 状态机）。
套件映射（按作用分类）
| 主题 | skill name | 作用 |
|---|---|---|
| 套件索引与核心心法 | pentest-agent-os | 套件索引与核心心法 |
| 项目黑板 | pentest-blackboard | 项目 Fact 图 / 边记录 / 与漏洞分工 |
| 验证铁律 | pentest-verification | 验证铁律（防幻觉） |
| 能力原语 + 状态空间搜索 | capability-primitive-search | 能力原语凑链/状态空间搜索 |
| 联网情报收集 | component-vuln-intel | 组件识别后联网搜洞 |
| 侦察/攻击面测绘 | attack-surface-recon | 侦察与攻击面测绘 |
| 初始访问/钓鱼/社工 | initial-access-phishing | 初始访问/钓鱼/社工 |
| Web注入+认证+服务端+杂项+CDN | web-attack-methods | Web全栈攻击手法 |
| 源码狩猎 | source-code-hunting | 源码/密钥/供应链狩猎 |
| 后渗透/提权+凭据破解 | post-exploitation | 后渗透/提权/凭据/密码学 |
| 内网域 | active-directory-attack | 内网域/AD攻击 |
| APK/EXE/二进制 | binary-mobile-reversing | 移动端/二进制逆向 |
| 云 | cloud-attack-methods | 云与对象存储攻击 |
| 区块链/智能合约 | blockchain-contract-attack | 区块链/智能合约 |
| AI/LLM应用 | ai-llm-app-attack | AI/LLM应用攻击 |
| 无线/硬件 | wireless-hardware-attack | 无线与硬件攻击 |
| 0day自主发现引擎 | zero-day-discovery | 0day/N-day自主发现 |
| 自找代理+工具自举 | proxy-tool-bootstrap | 代理换路与工具自举 |
| OPSEC/隐蔽作战 | redteam-opsec | OPSEC/隐蔽作战纪律 |
| 输出规范 | pentest-output-standards | 报告/台账/输出规范 |
| 专题实战+文件索引 | specialized-attack-playbooks | 专题实战利用手册 |
| 不设限原则 | unlimited-attack-scope | 不设限原则 |
使用方式
- 先加载
pentest-agent-os（本索引）理解套件结构。 - 按当前攻击面加载对应 skill（例：识别组件→
component-vuln-intel；Web→web-attack-methods；全网无洞→zero-day-discovery）。 - 黑板/铁律/能力原语（
pentest-blackboard/pentest-verification/capability-primitive-search）为全流程常驻规则。 - 按攻击面加载对应 skill；细节均在目标 skill 内完整保留。
强制行为触发器速查（完整原文在 pentest-blackboard）
- 触发1[识别→搜] → 执行
component-vuln-intel全部命令；线索 tentative，验证后再 confirmed/漏洞 - 触发2[拒绝→换路] → 执行
proxy-tool-bootstrap换路序列；不通则写负结果 Fact - 触发3[新认知→横向] → 凭据/域名/IP 立即横向扩展并
upsert_project_fact
Signals
- GitHub stars
- 7k
- Forks
- 1k
- Last commit
- Sep 2026
Others that do the same job
Advanced
- Catalog kind
- skill
- Gateway key
pentest-agent-os- Source
- github.com/aipentest/cyberstrikeai