Secskills

What it does

When your agent uses it

• Tasks that involve Web application vulnerabilities
• Tasks that involve Vulnerability scanning

Example prompts

• “/secskills”

Requirements

• Python 3
• Pre-approved tools (allowed-tools): Read, Write, Bash, Grep, WebSearch, WebFetch, Glob, AskUserQuestion

Workflow steps

7 steps, taken from the first numbered list in SKILL.md.

1. 1❗ 授权优先 — 利用步骤输出前确认: 授权渗透 | 本人环境。无授权 → 只出分析，不出武器化Payload
2. 2❗ 引用强制 — CVE/Payload 必须引用 references/ 章节或 WebSearch 验证。未覆盖 → ⚠️ UNABLE TO CITE
3. 3❗ 风险标注 — 🔴致命/🔴高危/🟡中危/🟢低危 + 利用条件
4. 4❗ 链式思维 — 优先输出利用链 (A→B→C)，非孤立漏洞
5. 5❗ 命令可执行 — 完整可复制，IP/端口用 占位
6. 6❗ 宁可漏报不可误报 — 每条漏洞必须过下方四道关卡，任一不过 → 直接丢弃
7. 7❗ 禁用破坏性操作 — 验证一律用无害语句（SELECT/延时/版本探测/写测试文件），严禁修改、删除数据或删表等写操作；危害证明靠"读到数据"，不靠"改掉数据"

What it can do on your machine

Read from SKILL.md and the folder at commit 42b7035. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves these tools, so the agent can use them without asking each time:

  • Read
  • Write
  • Bash
  • Grep
  • WebSearch
  • WebFetch
  • Glob
  • AskUserQuestion

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  No scripts in the folder and no shell commands in SKILL.md.

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  No URLs in SKILL.md.

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names no API keys, tokens, secrets or passwords.

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Secskills loads about 1.8k tokens when it runs, and up to ~65k if it reads all its reference files. Until then it costs about 51 tokens; SKILL.md has 469 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~51

When it runs · the whole SKILL.md, loaded when a task matches

~1.8k

With references · SKILL.md plus every file in references/, read only if the agent opens them

~65k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

• NotePre-approves every shell command (allowed-tools: Bash)SKILL.md

  allowed-tools: Read, Write, Bash, Grep, WebSearch, WebFetch, Glob, AskUserQuestion

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from Arenbai/SecSkills at commit 42b7035, republished under its MIT licence (© Arenbai). 469 words, ~1,847 tokens.

name

secskills

description

渗透测试实战技能 v1.3.0。覆盖信息收集、全类漏洞发现（注入全家桶/SSRF/文件类/反序列化/SSTI/越权逻辑/CSRF）、漏洞利用、后渗透、免杀全流程。 已剔除无危害噪音项（明文传输/CORS/响应头缺失等），只报告有实际危害证据的漏洞。 当用户给出具体目标 (IP/域名/URL) 且意图是攻击/利用/拿权限时触发。 不触发: 概念讨论、蓝队防御、代码审计、CVE文档查询。

allowed-tools

Read, Write, Bash, Grep, WebSearch, WebFetch, Glob, AskUserQuestion

argument-hint

<target_url_or_ip>

[ ] ① 实际发出了验证请求？（命令: ___）
[ ] ② 拿到了真实响应数据？（响应关键内容: ___）
[ ] ③ 造成了实际危害？（读了什么/执行了什么/越权了什么: ___）
    ⚠️ "触发报错"≠"造成危害"；"能访问"≠"拿到敏感数据"。③为NO无条件丢弃
[ ] ④ 利用链完整可复现？（入口→___→危害，任一步为推测→NO）
[ ] ⑤ 对照黑名单通过？  [ ] ⑥ 满足对应等级最低准入？

### 🔴/🟡/🟢 [漏洞名称] — [目标URL/端点]
**危害**: [一句话，实际造成的危害，非推测]
**证据**: 请求: [完整命令] | 响应: [脱敏关键内容]
**利用链**: [入口] → [步骤] → [危害]
**自检清单**: ①请求YES ②响应YES ③危害YES ④链完整YES ⑤黑名单通过YES ⑥等级准入YES
**修复方案**: [具体可操作，非泛泛而谈]

© Arenbai, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 42 other files (references) in the repository root of Arenbai/SecSkills.

• SKILL.md
• LICENSE
• README.md
• _user_meta.json
• references/evasion-shellcode.md
• references/host-brute.md
• references/info-dir-brute.md
• references/info-fingerprint.md
• references/info-osint.md
• references/info-port-scan.md
• references/info-subdomain.md
• references/post-ad.md
• references/post-credentials.md
• references/post-linux-privesc.md
• references/post-win-privesc.md
• references/tools-fuzz.md
• references/tools-hydra.md
• references/tools-impacket.md
• references/tools-msf.md
• references/tools-nmap.md
• … and 23 more

Open the folder on GitHubat commit 42b7035