Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →给中文用户和新手用的 Strix Lite 统一入口：先判断该用哪一个 strix-* 工具或漏洞测试 skill，再给最小化起手步骤；适合在 Web 安全测试、工具链使用、漏洞验证时不知道先用哪个 Strix skill 的场景;触发名:strix-beginner-hub
.claude/skills/asdfgh1445-strix/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | 20% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 85% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 42% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 61% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 18% | 0% |
这是给 不会选 Strix skill 的人 用的轻量入口。
它不是为了替代 ctf-super-hub，而是作为你现有 CTF 主包的 增强层：
ctf-super-hub出现下面情况时优先考虑这个入口：
httpx / ffuf / katana / nuclei / sqlmap 哪个strix-httpx：先探测站点、标题、状态码、技术指纹strix-katana：爬路径、JS、路由strix-ffuf：目录、文件、参数模糊测试strix-nuclei：快速模板扫描strix-sqlmap：SQL 注入自动验证与枚举strix-sql-injectionstrix-xssstrix-ssrfstrix-rcestrix-authentication-jwtstrix-idorstrix-information-disclosurestrix-insecure-file-uploadsstrix-open-redirectstrix-csrfstrix-business-logicstrix-broken-function-level-authorizationstrix-path-traversal-lfi-rfistrix-quick：先打高价值、快确认strix-standard：更平衡地覆盖测试面如果用户没有指定：
strix-httpx / strix-katana / strix-ffufstrix-quickstrix-standardreferences/router-cheatsheet.mdreferences/examples.md| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 16,557 | 42,973 | +160% | 1 | 1 | 0% | 2,385 | 2,806 | +18% | 0 | 0 | — |
case-02 | pass→pass | 17,836 | 12,312 | -31% | 1 | 1 | 0% | 2,531 | 2,599 | +3% | 0 | 0 | — |
case-03 | fail→fail | 17,090 | 14,693 | -14% | 1 | 1 | 0% | 2,423 | 2,869 | +18% | 0 | 0 | — |
case-04 | fail→pass | 12,902 | 8,748 | -32% | 1 | 1 | 0% | 1,790 | 2,152 | +20% | 0 | 0 | — |
case-05 | fail→pass | 7,729 | 9,195 | +19% | 1 | 1 | 0% | 1,173 | 2,170 | +85% | 0 | 0 | — |
case-06 | fail→pass | 11,005 | 9,055 | -18% | 1 | 1 | 0% | 1,526 | 2,168 | +42% | 0 | 0 | — |
case-07 | fail→pass | 8,427 | 8,879 | +5% | 1 | 1 | 0% | 1,288 | 2,074 | +61% | 0 | 0 | — |
case-08 | fail→pass | 12,575 | 9,277 | -26% | 1 | 1 | 0% | 1,734 | 2,038 | +18% | 0 | 0 | — |
case-09 | fail→pass | 15,176 | 12,381 | -18% | 1 | 1 | 0% | 2,177 | 2,609 | +20% | 0 | 0 | — |
case-10 | fail→pass | 14,269 | 12,127 | -15% | 1 | 1 | 0% | 1,950 | 2,469 | +27% | 0 | 0 | — |
case-11 | pass→pass | 6,385 | 8,010 | +25% | 1 | 1 | 0% | 906 | 2,003 | +121% | 0 | 0 | — |
case-12 | fail→pass | 12,144 | 6,537 | -46% | 1 | 1 | 0% | 1,809 | 1,639 | -9% | 0 | 0 | — |
case-13 | fail→pass | 9,386 | 9,145 | -3% | 1 | 1 | 0% | 1,443 | 2,047 | +42% | 0 | 0 | — |
case-14 | fail→pass | 10,298 | 8,521 | -17% | 1 | 1 | 0% | 1,373 | 2,085 | +52% | 0 | 0 | — |
case-15 | fail→pass | 10,122 | 9,426 | -7% | 1 | 1 | 0% | 1,412 | 2,218 | +57% | 0 | 0 | — |
case-16 | fail→pass | 20,501 | 8,776 | -57% | 1 | 1 | 0% | 2,702 | 1,963 | -27% | 0 | 0 | — |
case-17 | fail→pass | 16,829 | 9,195 | -45% | 1 | 1 | 0% | 2,261 | 2,049 | -9% | 0 | 0 | — |
case-18 | pass→pass | 12,488 | 9,019 | -28% | 1 | 1 | 0% | 1,728 | 2,007 | +16% | 0 | 0 | — |
case-19 | fail→pass | 13,725 | 4,081 | -70% | 1 | 1 | 0% | 2,084 | 1,318 | -37% | 0 | 0 | — |
case-20 | fail→pass | 15,203 | 7,821 | -49% | 1 | 1 | 0% | 1,997 | 1,858 | -7% | 0 | 0 | — |
case-21 | fail→pass | 15,655 | 11,685 | -25% | 1 | 1 | 0% | 2,208 | 2,602 | +18% | 0 | 0 | — |
case-22 | fail→pass | 11,485 | 8,978 | -22% | 1 | 1 | 0% | 1,593 | 2,038 | +28% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +77 percentage points is the difference between those two pass rates over the 22 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.