Cryptocurrency Theft Detailed Report: Sold for Only $105 on the Dark Web

marsbitPublicado em 2025-12-29Última atualização em 2025-12-29

Resumo

Phishing attacks go beyond stealing credentials through fake links. Stolen data is quickly commodified on the dark web. This report traces how data is collected via email, Telegram bots, and administration panels (like BulletProofLink), then sold and reused in future attacks. Data types range from instantly monetizable information (bank cards, e-wallet logins) to data used for follow-up attacks (account credentials, phone numbers) or targeted schemes (biometric data, ID scans). Analysis shows 88.5% of attacks in early 2025 aimed to steal online account credentials. On dark web markets, data is packaged, validated, and sold—often via Telegram—with prices varying based on account age, balance, and attached services. Old leaked data remains dangerous, as criminals compile comprehensive digital profiles for highly targeted attacks like whaling. Once stolen, it doesn’t disappear. Users must use unique passwords, enable multi-factor authentication, and monitor their digital footprint to reduce risk.

Author: Olga Altukhova Editor: far@Centreless

Compiled by: Centreless X(Twitter)@Tocentreless

Typical phishing attacks often involve users clicking on a fraudulent link and entering their credentials on a fake website. However, the attack is far from over at this point. Once sensitive information falls into the hands of cybercriminals, it immediately becomes a commodity, entering the "pipeline" of the dark web market.

In this article, we will trace the flow path of stolen data: from data collection through various tools (such as Telegram bots and advanced admin panels), to the sale of the data and its subsequent use in new attacks. We will explore how once-leaked usernames and passwords are integrated into vast digital profiles, and why data leaked years ago can still be exploited by criminals to carry out targeted attacks.

Data Collection Mechanisms in Phishing Attacks Before tracking the subsequent whereabouts of stolen data, we first need to understand how this data leaves the phishing page and reaches the cybercriminals.

Through the analysis of real phishing pages, we have identified the following most common data transmission methods:

  • Sent to an email address
  • Sent to a Telegram bot
  • Uploaded to an admin panel

It is worth mentioning that attackers sometimes use legitimate services for data collection to make their servers harder to detect. For example, they may use online form services like Google Forms, Microsoft Forms, etc. Stolen data may also be stored on GitHub, Discord servers, or other websites. However, for the convenience of this analysis, we will focus on the main data collection methods mentioned above.

Email

The data entered by the victim into the HTML form on the phishing page is sent to the attacker's server via a PHP script, which then forwards it to an email address controlled by the attacker. However, due to the many limitations of email services—such as delivery delays, the possibility of the hosting provider banning the sending server, and operational inconvenience when handling large amounts of data—this method is gradually decreasing.

Phishing kit contents

For example, we once analyzed a phishing kit targeting DHL users. The index.php file contained a phishing form for stealing user data (here, email address and password).

Phishing form imitating the DHL website

The information entered by the victim is then sent to the email address specified in the mail.php file via a script in the next.php file.

Contents of the PHP scripts

Telegram Bot

Unlike the method above, scripts using a Telegram bot specify a Telegram API URL containing a bot token and corresponding Chat ID, rather than an email address. In some cases, this link is even hardcoded into the phishing HTML form. Attackers design detailed message templates to be automatically sent to the bot upon successful data theft. A code example is as follows:

Code snippet for data submission

Compared to sending data via email, using a Telegram bot provides phishers with stronger functionality, which is why this method is becoming increasingly popular. Data is transmitted to the bot in real-time, and the operator is notified immediately. Attackers often use disposable bots, which are harder to track and ban. Furthermore, its performance does not depend on the quality of the phishing page hosting service.

Automated Admin Panels

More sophisticated cybercriminals use specialized software, including commercial frameworks like BulletProofLink and Caffeine, often provided as "Platform as a Service" (PaaS). These frameworks provide a web interface (dashboard) for phishing campaigns, facilitating centralized management.

All data collected by the phishing pages controlled by the attacker is aggregated into a unified database and can be viewed and managed through their account interface.

Sending data to the administration panel

These admin panels are used to analyze and process victim data. Specific functions vary depending on the panel's customization options, but most dashboards typically have the following capabilities:

  • Real-time statistics classification: View the number of successful attacks by time, country, and support data filtering
  • Automatic verification: Some systems can automatically verify the validity of stolen data, such as credit card information or login credentials
  • Data export: Support downloading data in various formats for subsequent use or sale

Example of an administration panel

Admin panels are a key tool for organized cybercrime groups.

It is worth noting that a single phishing campaign often employs multiple data collection methods simultaneously.

Data Types Coveted by Cybercriminals

The data stolen in phishing attacks varies in value and purpose. In the hands of criminals, this data is both a means of profit and a tool for carrying out complex multi-stage attacks.

Based on their use, stolen data can be divided into the following categories:

  • Immediate Monetization: Directly selling raw data in bulk, or immediately stealing funds from the victim's bank account or e-wallet
  1. Bank card information: Card number, expiration date, cardholder name, CVV/CVC code
  2. Online banking and e-wallet accounts: Login name, password, and one-time two-factor authentication (2FA) verification codes
  3. Accounts linked to bank cards: Login credentials for online stores, subscription services, or payment systems like Apple Pay/Google Pay
  • Used for subsequent attacks for further monetization: Using stolen data to launch new attacks for more gains
  1. Credentials for various online accounts: Usernames and passwords. It is worth noting that even without a password, just the email or phone number used as a login name has value to attackers
  2. Phone numbers: Used for phone scams (such as tricking users into giving 2FA codes) or phishing via instant messaging apps
  3. Personal Identifiable Information (PII): Full name, date of birth, address, etc., often used for social engineering attacks
  • Used for targeted attacks, extortion, identity theft, and deepfakes
  1. Biometric data: Voice, facial images
  2. Scanned copies and numbers of personal documents: Passport, driver's license, social security card, taxpayer identification number, etc.
  3. Selfies with documents: Used for online loan applications and identity verification
  4. Corporate accounts: Used for targeted attacks against businesses

We analyzed phishing and scam attacks that occurred between January and September 2025 to determine the data types most frequently targeted by criminals. The results showed: 88.5% of attacks aimed to steal various online account credentials, 9.5% targeted personal identity information (name, address, date of birth), and only 2% focused on stealing bank card information.

Selling Data on the Dark Web Market

Apart from being used for real-time attacks or immediate monetization, most stolen data is not used immediately. Let's take a deeper look at its flow path:

1. Data Packaged for Sale

After being consolidated, data is sold on dark web markets in the form of "data dumps"—compressed packages often containing millions of records from various phishing attacks and data breaches. A data dump may sell for as low as $50. The main buyers are often not active scammers, but dark web data analysts, the next link in the supply chain.

2. Classification and Verification

Dark web data analysts filter the data by type (email accounts, phone numbers, bank card information, etc.) and run automated scripts for verification. This includes checking the validity of the data and its potential—for example, whether a set of Facebook account passwords can also log into Steam or Gmail. Since users tend to use the same password on multiple websites, data stolen from a service years ago may still be applicable to other services today. Verified accounts that can still log in normally are sold at a higher price.

Analysts also correlate and integrate user data from different attack incidents. For example, an old social media leaked password, login credentials obtained from a phishing form impersonating a government portal, and a phone number left on a scam website may all be compiled into a complete digital profile of a specific user.

3. Sale on Specialized Markets

Stolen data is usually sold through dark web forums and Telegram. The latter is often used as an "online store," displaying prices, buyer reviews, and other information.

Offers of social media data, as displayed in Telegram

Account prices vary greatly, depending on many factors: account age, balance, linked payment methods (bank card, e-wallet), whether two-factor authentication (2FA) is enabled, and the popularity of the service platform. For example, an e-commerce account linked to an email, with 2FA enabled, a long usage history, and a large number of order records, will be sold at a higher price; for game accounts like Steam, expensive game purchase records increase their value; and online banking data involving high-balance accounts from reputable banks commands a significant premium.

The table below shows examples of prices for various types of accounts found on dark web forums as of 2025*.

4. High-Value Target Screening and Targeted Attacks

Criminals pay particular attention to high-value targets—users who hold important information, such as corporate executives, accountants, or IT system administrators.

Here is a possible scenario for a "whaling" attack: Company A has a data breach containing information on an employee who previously worked there and is now an executive at Company B. The attackers use Open Source Intelligence (OSINT) analysis to confirm that the user is currently employed at Company B. They then carefully forge a phishing email that appears to be from the CEO of Company B and send it to the executive. To enhance credibility, the email even cites some facts about the user from the previous company (of course, the attack methods are not limited to this). By lowering the victim's vigilance, criminals have the opportunity to further infiltrate Company B.

It is worth noting that such targeted attacks are not limited to the corporate sphere. Attackers may also target individuals with high bank account balances, or users holding important personal documents (such as those required for micro-loan applications).

Key Takeaways

The flow of stolen data is like an efficiently operating pipeline, with each piece of information becoming a commodity with a clear price tag. Today's phishing attacks widely use diverse systems to collect and analyze sensitive information. Once data is stolen, it quickly flows into Telegram bots or the attacker's admin panels, where it is then classified, verified, and monetized.

We must be清醒地认识到清醒地认识到 (clearly aware): Once data is leaked, it does not disappear into thin air. On the contrary, it is constantly accumulated, integrated, and may be used months or even years later to carry out targeted attacks, extortion, or identity theft against the victims. In today's online environment, staying vigilant, setting unique passwords for each account, enabling multi-factor authentication, and regularly monitoring one's digital footprint are no longer suggestions, but necessities for survival.

If you unfortunately become a victim of a phishing attack, please take the following measures:

  1. If bank card information is leaked, immediately call the bank to report the loss and freeze the card.
  2. If account credentials are stolen, immediately change the password for that account, and also change the passwords for all other online services that use the same or similar passwords. Be sure to set a unique password for each account.
  3. Enable multi-factor authentication (MFA/2FA) on all supported services.
  4. Check the account's login history and terminate any suspicious sessions.
  5. If your instant messaging or social media account is stolen, immediately notify friends and relatives, reminding them to be wary of fraudulent messages sent in your name.
  6. Use professional services (such as Have I Been Pwned, etc.) to check if your data has appeared in known data breach incidents.
  7. Be highly vigilant of any unexpected emails, phone calls, or promotional information you receive—they may seem credible precisely because attackers are using your leaked data.

Perguntas relacionadas

QWhat are the three most common methods for transmitting stolen data from phishing pages to cybercriminals?

AThe three most common methods are: sending to an email address, sending to a Telegram bot, and uploading to an administration panel.

QWhy are cybercriminals increasingly using Telegram bots over email for data collection?

ATelegram bots provide real-time data transmission, immediate notifications to the operator, are harder to track and block, and their performance is not dependent on the quality of the phishing page hosting service.

QWhat percentage of phishing and scam attacks from January to September 2025 aimed to steal online account credentials?

A88.5% of the attacks aimed to steal various online account credentials.

QWhat is the typical first step in the 'pipeline' of stolen data after it is collected and before it is used in new attacks?

AThe data is packaged and sold as 'dumps' on dark web marketplaces, often for as little as $50.

QAccording to the article, what is one crucial step a victim should take if their online account credentials are stolen?

AThey should immediately change the password for that account and also change the passwords for all other online services where the same or a similar password was used, ensuring a unique password for every account.

Leituras Relacionadas

Claude's Major New Feature: Screen Recording + Voice, Distilling Your Skills into AI Tasks in One Click

Claude has introduced a major new feature called "Record a Skill," available for Pro, Max, and Team users. This function, found in the Claude desktop app's CoWork menu, allows users to create reusable AI skills simply by recording their screen and providing voice narration while performing a task. Claude then automatically analyzes the recording and generates a functional Skill. A hands-on test confirmed the feature works seamlessly. Users start recording via the Skills manager, perform their workflow while verbally explaining the steps and logic, and avoid including sensitive information. After recording, Claude processes the content and creates the Skill, which can be saved and later invoked with a slash command (/). This eliminates the need for manual adjustments or writing complex instruction files. The innovation goes beyond mere efficiency. Previously, creating a Skill required writing a detailed SKILL.md file in Markdown—a significant barrier for non-technical users. "Record a Skill" bypasses this by directly capturing both actions and the implicit reasoning shared in the narration. This lowers the barrier to knowledge transfer and automation, addressing a core challenge in corporate knowledge management: the difficulty of getting experts to write and maintain documentation. However, the feature also highlights a shift in the nature of work. A case study from March 2026 showed a freelancer whose five-year client relationship was effectively replaced by a hand-coded Claude Skill automating their content workflow. With the even lower barrier of screen recording, the ability to distill personal expertise into automatable skills accelerates this trend. The "moat" for work is moving from simply knowing how to do a task to mastering tasks that are difficult or impossible to automate.

marsbitHá 6m

Claude's Major New Feature: Screen Recording + Voice, Distilling Your Skills into AI Tasks in One Click

marsbitHá 6m

Feeding AI "Noise" Can Also Boost Scores, This Work Enables Positive Transfer with Noise

Feeding "Noise" to AI Can Improve Performance: A Method Enables Positive Transfer from Noise This work, Semi-Supervised Noise Adaptation (SSNA), introduces a Noise Adaptation Framework (NAF) that challenges traditional transfer learning. Instead of requiring a labeled source domain of real data (e.g., images, text), NAF uses randomly generated Gaussian noise as the source. For a target task with C classes, it constructs C noise clusters by sampling from Gaussian distributions. Although this synthetic noise contains no semantic meaning, NAF trains it to form a discriminative class structure in a shared representation space—clustering same-class noise and separating different classes. The key is aligning this learned structure from the noise domain to the real, sparsely labeled target domain. A small number of target labels are still essential to establish the correspondence between noise clusters and actual classes. The training objective combines: 1) supervised loss on the few labeled target samples, 2) classification loss for the noise to build its structure, and 3) a distribution alignment loss (using Negative Domain Similarity) to minimize the gap between the noise and target domains in the shared space. Experiments show significant gains in few-label settings. With just 4 labels per class, NAF with a ResNet-18 backbone improves accuracy over a standard supervised baseline (ERM) by +12.35% on CIFAR-10, +7.61% on CIFAR-100, +4.38% on DTD-47, and +2.74% on Caltech-101. It also benefits fine-grained datasets and scales to ImageNet-1K (with 100 labels/class) and text classification (AG News). NAF can be integrated into existing semi-supervised methods like FixMatch for further gains. Ablation studies confirm the transferred benefit comes from the discriminative structure of the noise, not randomness itself. Collapsing all noise into a single point causes negative transfer, while increasing separation between noise cluster centers improves performance. The amount of noise per class is less critical once a basic structure forms. In conclusion, this work demonstrates that for positive transfer, the semantic content of source data may not be necessary. What can be effectively transferred is the *organizational structure* of categories within a representation space. This offers a promising alternative for scenarios where real source data is unavailable due to privacy, copyright, or procurement constraints.

marsbitHá 7m

Feeding AI "Noise" Can Also Boost Scores, This Work Enables Positive Transfer with Noise

marsbitHá 7m

Trading

Spot
活动图片