claudleak

claudleak 扫描公共GitHub仓库中AI编码工具配置文件（.claude/、.cursor/、.continue/、.codex/、CLAUDE.md、AGENTS.md等）中的泄露凭证。 使用TruffleHog进行秘密检测。 示例.claude/目录： root@kitploit: "permissions": "allow": "Bashgo build:", "BashRAPID7ENCRYPTEDCREDS='eP3g488hQ23QiU4kJahhwPKs3PyE' go test ./pkg/agents/ -run TestRules -v -count=1...

CVE-2021-21975

CVE-2021-21975 SSRF-POC - SSRF 到凭据泄露 首先配置 pd 客户端。 现在运行 ./pd deploy http-new-requests 现在替换脚本中的 pipedream http 请求示例：Randomstring.m.pipedream.net 在一个单独的终端中运行 - ./pd logs new-requests | tee -a req.txt 然后在另一个终端中运行 - python3 CVE-2021-21975.py iplist cat req.txt | grep Basic basic.txt; cat basic.txt |...

CVE-2020-5260

CVE-2020-5260 CVE-2020-5260演示记录 git在执行类似"git clone https://example.com" 这样的命令时,会请求使用协议"https"存储主机"example.com"的凭据，并在远程端请求身份验证时将返回的凭据附加到发出的请求中 root@kitploit: git clone 'http://localhost:8088/%0ahost=github.com%0aprotocol=https 参考链接: https://bugs.chromium.org/p/project-zero/issues/detail?id=2021...

CVE-2020-9758

CVE-2020-9758 描述 LiveZilla Live Chat 8.0.1.3（Helpdesk）的 chat.php 中存在一个问题。在 name 参数中发现了一个盲注 JavaScript 注入。触发此问题可以获取 URI 中客服员工的用户名和密码。这导致了权限提升，从未经身份验证到用户级访问，从而完全接管账户。该攻击会获取多个凭据，因为它们存储在数据库中（存储型 XSS）。这通过 lgn 和 psswrd 参数影响了 mobile/chat URI。 附加信息 通过 URI 泄露凭据可能是自动登录功能的结果。此外，chat.php 表单中的更多参数也可能存在漏洞。 漏洞类...

CVE-2026-77254

The mcp-atlassian Model Context Protocol server (versions prior to 0.22.0 ) contains a vulnerability where unauthenticated HTTP MCP requests can bypass per-user identity checks. The flaw allows a network caller to reach tool handlers that utilize globally configured Jira and Confluence credential...

nuclei-CiscoRV320Dump-CVE-2019-1653

Cisco Router Configuration Exposure Vulnerability Descripción de la Vulnerabilidad Este repositorio detalla el proceso de explotación de una vulnerabilidad en ciertos routers Cisco que permite la exposición del archivo de configuración config.exp mediante una URL pública accesible. Este archivo...

Security Bulletin: IBM InfoSphere Optim Archive Viewer is affected by multiple vulnerabilities in axios (CVE-2026-44486, CVE-2026-44487)

Summary Multiple proxy credential leakage vulnerabilities in the axios HTTP client library CVE-2026-44486, CVE-2026-44487 used by IBM InfoSphere Optim Archive Viewer have been addressed by upgrading the component to version 1.18.0. Vulnerability Details CVEID:CVE-2026-44486 DESCRIPTION: Axios is ...

mongobleed-exploit-CVE-2025-14847

🩸 MongoBleed - CVE-2025-14847 Laboratorio de Investigación de Seguridad CVE-2025-14847 | CVSS 8.7 Alto | Divulgación de Memoria No Autenticada CVE-2025-14847 , apodado MongoBleed , permite a atacantes remotos no autenticados extraer memoria heap no inicializada explotando la lógica de descompresi...

Multiple vulnerabilities impacts AIX due to cURL libcurl

IBM SECURITY ADVISORY First Issued: Mon Sep 21 04:38:42 CDT 2026 The most recent version of this document is available here: https://aix.software.ibm.com/aix/efixes/security/curladvisory12.asc Security Bulletin: Multiple vulnerabilities impacts AIX due to cURL libcurl...

anticipator

Anticipator 多智能体AI系统的运行时安全。 Anticipator 可检测提示注入、凭证泄露、编码攻击、同形文字欺骗、路径遍历以及 LangGraph 管道中的异常智能体行为——在它们成为安全事件之前。 无需大语言模型、无需嵌入、无需外部 API。完全本地化、完全确定性，每条消息检测耗时低于 5 毫秒。 ⚡ 在管道中发现异常？提交 Issue——我们期待看到真实世界的检测案例。 为什么选择 Anticipator 多智能体系统引入了一种新型安全问题。当智能体之间传递消息时，其中任何一条消息都可能携带注入攻击、泄露的凭证、编码载荷或角色操纵——而现有工具并未监控此类流量。...

CVE-2026-21002-Serverless-Cold-Start-Credential-Leakage-via-Reused-tmp

CVE-2026-21002 – Fuga de credenciales en arranque en frío de funciones serverless mediante reutilización de /tmp Código del programa simulación de AWS Lambda en Python root@kitploit: lambdafuncsim.py - Simulated serverless function import os, tempfile, time SECRETFILE = '/tmp/credentials' reused...

CVE-2026-11106-DNS-Zone-Transfer-AXFR-Information-Disclosure

CVE-2026-11106 – Divulgación de información de transferencia de zona DNS AXFR Código del programa servidor DNS en Python + ataque root@kitploit: dnsserversim.py - Simulated DNS server allowing AXFR from dnslib import DNSRecord, QTYPE, RR, TXT, A import socketserver, socket class...

CVE-2025-30374

CVE-2025-30374 Vulnerabilidad de contaminación de clases en Taipy que conduce a RCE, XSS, DoS y fuga de credenciales Resumen Se ha identificado una vulnerabilidad de contaminación de clases en Taipy v4.0.3 la última versión en el momento del descubrimiento. Esta vulnerabilidad permite a atacantes...

EulerOS Virtualization 2.11.0 : python-pip (EulerOS-SA-2026-3551)

According to the versions of the python-pip packages installed, the EulerOS Virtualization installation on the remote host is affected by the following vulnerabilities : Requests is a HTTP library. Due to a URL parsing issue, Requests releases prior to 2.32.4 may leak .netrc credentials to third...

EulerOS Virtualization 2.12.1 : perl-HTTP-Tiny (EulerOS-SA-2026-3606)

According to the versions of the perl-HTTP-Tiny package installed, the EulerOS Virtualization installation on the remote host is affected by the following vulnerabilities : HTTP::Tiny versions before 0.095 for Perl forward credential headers to cross-origin redirect targets.When the server return...

HTB-Mailing-A-Complete-Walkthrough

HTB: Mailing — Una guía completa Por Mursalin Si has estado machacando máquinas de HackTheBox, Mailing es una de esas cajas que realmente te enseña algo. Está clasificada como Fácil, corre en Windows, y encadena varias vulnerabilidades del mundo real: un directory traversal, una filtración de...

USN-8780-1: libsoup vulnerabilities

It was discovered that libsoup incorrectly handled certain URLs when using an HTTP proxy. A remote attacker could possibly use this issue to inject arbitrary HTTP headers. CVE-2026-1467 It was discovered that libsoup did not remove proxy authentication credentials when following HTTP redirects. A...

CVE-2026-86864

pgAdmin 4 (Backup tool) is vulnerable to argument and connection-string injection via the client-supplied database field. Because the value was appended as a bare trailing positional argument to pg_dump, a dash-prefixed value (e.g. --file=/abs/path) was interpreted as an option due to getopt_long...

CVE-2026-92916

Grav is a flat-file CMS. In Grav 1.7.0 through 1.7.53.2 and 2.0.0 through 2.0.21, when the debugger is enabled system.debugger.enabled: true, which is not the default, the Clockwork profiler endpoint is exposed without authentication: InitializeProcessor::handleDebuggerRequest intercepts any path...

keyguard

keyguard 扫描代码库、GCP 项目和 CI 流水线中暴露的 Google API 凭据——在它们被利用之前。 问题： Google 在已有的 API 密钥（Maps、Places 等）上回溯启用 了 Gemini API 访问，这些密钥本来设计为公开并嵌入客户端代码。当团队中有人在同一个 GCP 项目中启用 Gemini 时，那些本已公开的密钥会悄悄变成 Gemini 凭据——导致项目面临未经授权的 AI 使用风险，并可能产生超过 8 万美元的账单事故（背景）。 keyguard 的功能： keyguard scan —— 使用正则表达式 + 熵检测，在源文件和 Git...