Skip to content

Vulnerability detection. With Code Intelligence.

Fuzzing is the most effective approach to automatically find and fix bugs and vulnerabilities in your software. Companies like Google already find 80% of their bugs completely automated.


Efficient testing. Cover the unexpected.

CI Fuzz makes the high end testing technology used in critical systems available to everybody and allows to save hours of creating and maintaining traditional tests.

google-2015-3Deutsche_Telekom_2022 1-3bosch-logo-simple 1-2Secunet_Security_Networks_Logo-2Continental_AG_logo 1-2Cariad_Logo-2ETAS-Logo-2

Vulnerabilities found with Code Intelligence.

A selection of open-source code vulnerabilities we discovered in customer projects

Our impact


Security issues disclosed in Open-source

Our CI Fuzz solution not only detects issues in C/C++ code, but also identifies security vulnerabilities in Java code. This technology has successfully uncovered hundreds of security issues and code vulnerabilities in prominent open-source projects such as log4j and OpenJDK.

Improve your security testing with Code Intelligence.


Securing Telco Software

CI Fuzz enabled Telekom’s security testing experts to complete projects much faster and to increase productivity in an easy way.


Securing Automotive Software

Continental managed to test a large safety module with 18,00 lines of code (LoC) within only one week.


Securing Supply Chains

The sodge IT team has implemented automated security testing in their supply chain and thus detected and fixed 3 critical vulnerabilities (CVEs).


12 Vulnerabilities

OISF / Suricata

Sucicata is an open-source threat detection engine capable of real-time intrusion detection (IDS). Even though Sucicata used fuzzing with AFL extensively, our vulnerability researchers were able to find 12 more critical code vulnerabilities (e.g. data corruptions, remote code executions, and many other unwanted program behaviors).


11 Vulnerabilities

German COVID-19 Tracing App

The Swiss/German COVID-19 tracing app was downloaded more than 25 Million times. Testing it for security vulnerabilities played an important role in ensuring public trust and increasing compliance. With CI Fuzz, our developers found 2 code vulnerabilities in the German version (#163 and #164), and 9 vulnerabilities in the Swiss version of the app.


2 Vulnerabilities

OWASP / JSON Sanitizer

OWASP's JSON Sanitizer transforms JSON-like input into valid and fully embeddable JSON. It is an Open Source sanitizer with over 200 dependencies. With CI Fuzz, our developers found two critical code vulnerabilities in this project, that would have exposed it to XML/HTML injections. (CVE-2021-23899 and CVE-2021-23900).


1 Vulnerability


LibHTP is used to securely parse HTTP protocols and the related bits and pieces. Using CI Fuzz, our developers found a critical code vulnerability in LibHTP, that remote attackers could have exploited for a buffer over-read (CVE-2018-10243).


8 Vulnerabilities

Zint Barcode Generator

Zint is a barcode generator with an encoding library that includes over 50 different symbologies. With CI Fuzz, our security researchers found 2 critical code vulnerabilities (CVE-2020-9385, CVE-2021-27799 CVE-2021-39247) and 5 more vulnerabilities. 


1 Vulnerability

audi / a_util

Audi is one of the leading innovators worldwide when it comes to making automotive software more secure. With CI Fuzz, our developers managed to fix an exploitable heap buffer overflow in Audi's software.


1 Vulnerability


CppCMS is a free, Open Source framework for web application development in C++, especially adept at handling extremely high loads. With CI Fuzz, our security researchers managed to find a critical code vulnerability (CVE-2018-11367) within the project.


1 Vulnerability

DLR COBS Encoder

The DLR is the federal german space research center, which focuses its research on energy, space, transport, and security technology. With CI Fuzz, our developers were able to find a heap-buffer overread in one of the DLR's COBS encoder.


7 Vulnerabilities


GnuPG Made Easy or simply GPGME is a library that makes GnuPG access easier for applications. CI Fuzz enabled our developers to find 7 code vulnerabilities in GPGME that would have led to severe usability issues.


1 Vulnerability


GENIVI DLT provides a log and trace interface, based on the standardised protocol specified in the AUTOSAR standard 4.0 DLT. Here we found heap-based buffer overflow (CVE-2020-36244) that could allow an attacker to remotely execute arbitrary code on the DLT-Daemon.


19 Vulnerability


Jsoup is an open-source Java library used for extracting data from HTML and manipulating HTML outputs.  It can also be used to parse and build XML. With Jazzer, we found over 19 bugs (CVE-2021-37714) in jsoup. Eight of them are potentially critical for the availability of web applications using jsoup.


3 Vulnerabilities

Apache PDFBox

The Apache PDFBox library is an open-source Java tool for working with PDF documents. It allows the creation of new PDF documents, manipulation of existing documents, and the ability to extract content from documents. Our fuzzers were able to find an Infinite Loop (CVE-2021-27807), and an OutOfMemory-Exception (CVE-2021-27906).


5 Vulnerability


Our tests on the Java library Fastjson, which converts Java Objects to JSON, revealed a number of security issues, including stackoverflow errors and undeclared exceptions, which resulted in five approved CVE's (CVE-2022-40173CVE-2022-40174CVE-2022-40175CVE-2022-41855CVE-2022-41856).


6 Vulnerability

Apache Commons Compress

Apache Commons Compress software defines an API for working with compression and archive formats. We extensively tested this application and responsible disclosed four CVEs (CVE-2021-35516CVE-2021-35515CVE-2021-35517CVE-2021-36090) and two more vulnerabilities.


4 Vulnerabilities


Json-smart is a performance focused, JSON processor lib. The application (json-smart-v1, json-smart-v2) threw uncaught exceptions. When uncaught, they may have caused programs using the library to crash or expose sensitive information. (CVE-2021-27568).


2 Vulnerability


In OpenJDK, we found two vulnerabilities (CVE-2022-21360CVE-2022-21366) that load and run untrusted code. These vulnerabilities could have been exploited by using APIs in the specified Component, e.g., through a web service that supplied data to the APIs.


2 Vulnerability


Protocol Buffers are Google's language-neutral and platform-neutral, extensible mechanism for serializing structured data. Two issues in protobuf allowed malicious payload to occupy the parser for several minutes, causing frequent, repeated pauses (CVE-2021-22569, CVE-2022-3171).


2 Vulnerabilities

Faster XML/Jackson

Faster XML Jackson is a high-performance JSON processor for Java. Its developers extol the combination of fast, correct, lightweight, and ergonomic attributes of the library in which we found undeclared exceptions.


5 Vulnerability


Our team discovered 26 bugs while reviewing the open-source, multi-format 1D/2D barcode image processing library, ZXing ("zebra crossing"). Among these were 5 security-relevant uncaught exceptions, which we discovered with our automated security testing platform. 


46 Vulnerability


In Javaparser, a widely-used tool for analyzing Java code, we were able to find 46 security vulnerabilities through our use of the open-source fuzzer Jazzer. These issues primarily pertained to uncaught exceptions and security exceptions.


3 Vulnerabilities


Our team found 3 security issues in ANTLR (ANother Tool for Language Recognition), a powerful parser generator for structured text or binary files, through the use of our open-source fuzzer Jazzer. ANTLR is commonly used to build languages, tools, and frameworks.


1 Vulnerability


This is a multi-module umbrella project for Jackson modules needed to support Java 8 features, especially with Jackson 2.x that only requires Java 7 for running (and until 2.7 only Java 6). Here we found a security exception, which frequently occurs on the Linux platform.


2 Vulnerability


Our open-source security team identified two security issues (security exception, uncaught exception) in the Java modules of Apache Arrow, a development platform for in-memory analytics, which enables big data systems to process and move data quickly.

Guava (3)-1

3 Vulnerabilities


Our team found 3 security exceptions in the widely-used Guava libraries, a set of core Java libraries from Google that includes new collection types, immutable collections, a graph library, and utilities for concurrency, I/O, hashing, caching, primitives, strings, and more.


8 Vulnerability


Code Intelligence found 8 security exceptions in Apache Tomcat, a free and open-source implementation of Jakarta Servlet, Jakarta Expression Language, and WebSocket technologies, which provides a pure Java HTTP web server environment for Java code to run.


1 Vulnerability


Our team identified a security exception in spring-data-redis, a library that simplifies Redis access and config for Spring apps. Spring Data Redis offers both low-level and high-level abstractions for interacting with the store and eliminates infrastructural concerns for the user.


3 Vulnerabilities


Our open-source fuzzer Jazzer helped to identify 3 security exceptions in Spring Data for MongoDB, a library that is part of the Spring Data project which provides a consistent programming model for new data stores while retaining store-specific features and capabilities.


1 Vulnerability


TIFF is a Java library for reading/writing TIFF files developed by the National Geospatial-Intelligence Agency (NGA). With Jazzer, we were able to find and fix a security issue that made the application vulnerable to IndexOutOf BoundsExceptions.


1 Vulnerability


Klaxon is a library to parse JSON in Kotlin. Our open-source security team was able to fix a security issue in klaxon, with the help of our open-source fuzzer, Jazzer. The security issue caused unexpected exceptions when using the default parser. 


5 Vulnerabilities


Code Intelligence found 5 StackOverflowErrors in SnakeYAML, a YAML 1.1 processor used by over 3,646 Maven artifacts while fuzzing the application with Jazzer. (CVE-2022-38749CVE-2022-38750CVE-2022-38751, CVE-2022-38752CVE-2022-38754).


1 Vulnerability


Our open-source security team identified a vulnerability (CVE-2022-40151) in XStream, a library for serializing XML. This vulnerability may leave users open to Denial of Service attacks if the parser is on user-supplied input and can be crashed by stack overflow.


2 Vulnerability


Jettison is a Java library for converting XML to JSON and vice-versa with the help of StAX. With our open-source fuzzer, Jazzer, we could prove that those using Jettison to parse untrusted XML or JSON data may be vulnerable to Denial of Service attacks (DoS) (CVE-2022-40149CVE-2022-50150).


1 Vulnerabilities


Woodstock is a general-purpose Java XML parser that can be used with the most popular Java frameworks. Our open-source security team discovered a vulnerability in woodstock (CVE-2022-40152) that made users vulnerable to Denial of Service attacks (DoS) if DTD support was enabled.


1 Vulnerability


A remote code execution bug was found in apache/JXPath with the support of our open-source fuzzer, Jazzer (CVE-2022-41825). JXPath is used by 364 Maven artifacts, including Maven and Apache Commons configuration (having over 1000 usages with the potential to have transitive risk).

hsqldb (2)

1 Vulnerability


As part of our goal to continuously improve our vulnerability detectors, we continuously test various open-source projects with Jazzer in collaboration with OSS-Fuzz. In this case, a test run yielded a severe finding with a potential remote code execution in HSQLDB (CVE-2022-41853)


1 Vulnerabilities


Our open-source security team found a vulnerability in Apache Commons Imaging, a library for various image formats, previously known as Apache Commons Sanselan. The vulnerability allowed for undeclared exceptions in parsers for multiple image formats.

Meet our researchers.

Join us on our mission to secure the world’s information, critical infrastructure, and favorite devices.
Khaled Yakdan
Khaled Yakdan
Co-Founder & Chief Scientist
Norbert Schneider
Open-Source Security Engineer
Fabian Meumertzheim
Open-Source Security Engineer
“One of the biggest advantages of instrumented fuzz testing is that you can execute your code in a Software-in-the-Loop simulator. My favourite part of instrumented fuzzing is that finding the root cause is so easy, and for a manager, it means I can save budget.”
Michael von Wenckstern 2024
Michael Von WencksternProduct Cybersecurity Governance, Risk and Compliance Specialist, Continental AG
"Thanks to Code Intelligence fuzzing approaches, our security testing became significantly more effective. All our developers are now able to fix business critical bugs early in the development process, without false-positives."


Andreas Weichslgartner
Andreas WeichslgartnerSenior Technical Security Engineer, CARIAD
”Thanks to Code Intelligence we were able to remediate deeply hidden issues, allowing us to ensure our vehicular software’s optimal functionality and safety. Coming up with the right unit tests for these cases would have been super difficult. With Code Intelligence’s AI-powered tests, we had the first finding within hours!”
Saleh HeydariVP of Software Engineering, XOS Trucks
”Code Intelligence helps developers ship secure software by providing the necessary integrations to test their code at each pull request, without ever having to leave their favorite environment. It's like having an automated security expert always by your side.”
Thomas DohmkeCEO, GitHub

Why choose Code Intelligence?

Join Industry Leaders and follow in the footsteps of companies like CARIAD, Bosch, and Continental. Detect critical bugs early in the testing stages and achieve compliance with industry standards.

Book your free demo with one of our senior engineers now and take the first step towards robust, secure software development with Code Intelligence.

  • Automate software testing for embedded systems.
  • Detect critical bugs & vulnerabilities early in the development.
  • Uncover only actual issues without false positives.
  • Enable developers to reproduce & fix issues in minutes, not weeks.
  • Ensure compliance with industry standards.

Security resources


SAST, DAST, IAST and Feedback-Based Fuzzing

In today's software testing industry acronyms like SAST, DAST or IAST are omnipresent, with IAST being the most recent trend in 2019.

Bug Detectors for log4j Are Now Available in Google’s OSS-Fuzz

To help contain the damages that arise from the log4j vulnerability, Code Intelligence collaborated with Google. Together, we implemented effective bug detectors.

CI and Google Secure Open-Source JavaScript Landscape

Collaboration will allow open-source developers to continuously test their JavaScript components for bugs and vulnerabilities, and assist them in writing more secure and reliable code.