Install
$ agentstack add skill-zhaji2333-ckskills-injection-vulns ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ● Network access Used
- ✓ Filesystem access No
- ● Shell / process execution Used
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
injection-vulns — 注入类漏洞专项深度挖掘
何时调用(触发条件)
- 参数直接拼接到 SQL / NoSQL 查询(搜索、筛选、排序、分页、导出条件)
- JSON 查询条件可控(
{"$gt":0}等 MongoDB 风格) - 用户输入进入 shell 命令(exec/system/ProcessBuilder、ping/curl/文件名拼接)
- 用户输入进入模板引擎(Jinja2/Velocity/Freemarker/Thymeleaf)
- 用户输入进入表达式解释器(SpEL/OGNL/EL/MVEL)
- 危险函数出现在代码审计结果中(见
source-code-audit)
一、漏洞类型全景
| 类型 | 危险函数/场景 | 挖掘要点 | |---|---|---| | C1. SQL注入 | 拼接SQL、动态排序 | 响应差异验证(错误/延迟/条数) | | C2. NoSQL注入 | JSON查询条件可控 | 类型替换、全量查询 | | C3. 命令注入 | exec/system/ProcessBuilder | shell拼接、管道符 | | C4. 模板注入(SSTI) | Jinja2/Velocity/Freemarker | 用户输入进入模板渲染 | | C5. 表达式注入 | SpEL/OGNL/EL | 可控片段进入解释器 |
二、常见场景清单(搜索/查询类)
| 场景 | 漏洞类型 | 挖掘要点 | |---|---|---| | 站内搜索 | SQL注入/XSS/信息泄露 | 搜索词拼接、搜索结果反射 | | 高级筛选/排序 | SQL注入/NoSQL注入 | order by注入、筛选条件可控 | | 自动补全/联想 | 信息泄露/用户枚举 | 补全接口泄露敏感数据 | | 数据统计/图表 | SQL注入/越权 | 统计维度可控、跨权限聚合 | | ES/Solr搜索 | 查询注入/未授权 | DSL注入、搜索接口暴露 |
三、接口 Fuzz 技巧(参数污染与类型混淆)
参数污染:
?id=1&id=2 → 看哪个生效
?id[]=1&id[]=2 → 数组化
?id=1,2,3 → 批量查询
类型混淆:
id=1 → id="1" → id={"$gt":0} → id[]=1
编码绕过:
Unicode: \u002e\u002e/ (../)
双重URL编码: %252e%252e%252f
隐藏参数:
debug=1 / test=1 / admin=1
_method=PUT / X-HTTP-Method-Override
批量操作:
ids=1,2,3,4,5 一次查多个
page=-1 / limit=99999 绕过分页
四、SQL 注入绕过技术
大小写:SeLeCt / 双写:selselectect
注释:/**/、/*!*/、--+、#
编码:URL编码、十六进制、Unicode
等价函数:substr→mid、ascii→ord
五、命令注入绕过技术
分隔符:| / || / & / && / ; / %0a
空格:${IFS} / $IFS$9 / %09 / {cat,/etc/passwd}
通配符:/???/??t /etc/passwd
六、验证要点
- 响应差异对比:基线请求 vs 注入请求的状态码/响应大小/错误信息/响应时间
- 时间盲注:
SLEEP(5)/pg_sleep(5)/WAITFOR DELAY '0:0:5'验证 - 报错注入:
updatexml/extractvalue/ 类型转换错误 - NoSQL:
{"$ne":null}、{"$where":"1==1"}、数组/类型替换 - SSTI:
{{7*7}}、${7*7}、`` 探测 - 命令注入:
id、whoami、DNSLog 外带 - 所有注入点至少尝试到 Level 4 绕过(见 CLAUDE.md 1.1 第五节)
七、修复建议
- SQL:预编译/参数化查询,动态排序用白名单映射
- NoSQL:禁止
$运算符进入查询条件,类型强校验 - 命令:白名单命令 + 参数数组传递,禁止 shell 拼接
- SSTI:模板引擎关闭危险特性(Jinja2 禁用
__class__链) - 表达式:禁止用户可控片段进入 SpEL/OGNL/EL 解释器
八、输出与报告要点
- 完整 PoC(curl 命令),标注注入点参数与 payload
- 差异对比证据(基线响应 vs 注入响应)
- 影响面:可读取的库表、可执行的命令、可拿到的数据量
- 根因分析:输入点 → 传播链 → Sink
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: zhaji2333
- Source: zhaji2333/CkSKILLS
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.