Categories: Kali Linux

Headless Burp – Automate security tests using Burp Suite

Headless Burp provides an extension to Burp that allows you to run Burp Suite’s Spider and Scanner tools in headless mode via command-line.

However, it can do more! It can produce a JUnit like report which in turn could instruct the CI server to mark the build as “failed” whenever any vulnerabilities are found. You can also mark some issues as false positives and those will not be reported anymore on the next scan reports.

Headless Burp Build

./mvnw

The extension is packaged as a fat jar at target/headless-burp-scanner-master-SNAPSHOT-jar-with-dependencies.jar

Also ReadPeda – Python Exploit Development Assistance for GDB

Usage

Bu8ld the extension as shown above or install it from the BApp Store

Using a loclally built extension jar

On *nix:

java -Xmx1G -Djava.awt.headless=true \
-classpath headless-burp-scanner-master-SNAPSHOT-jar-with-dependencies.jar:burpsuite_pro_v1.7.31.jar burp.StartBurp \
--unpause-spider-and-scanner \
--project-file=project.burp -c config.xml

On Cygwin:

java -Xmx1G -Djava.awt.headless=true \
-classpath "headless-burp-scanner-master-SNAPSHOT-jar-with-dependencies.jar;burpsuite_pro_v1.7.31.jar" burp.StartBurp \
--unpause-spider-and-scanner \
--project-file=project.burp -c config.xml

Using the extension from BApp Store

java -Xmx1G -Djava.awt.headless=true \
-classpath burpsuite_pro_v1.7.31.jar burp.StartBurp \
--unpause-spider-and-scanner \
--project-file=project.burp -c config.xml

On Cygwin:

java -Xmx1G -Djava.awt.headless=true \
-classpath "burpsuite_pro_v1.7.31.jar" burp.StartBurp \
--unpause-spider-and-scanner \
--project-file=project.burp -c config.xml

Configuration

<?xml version="1.0" encoding="UTF-8"?>
<config xmlns="http://nets.eu/burp/config">
    <reportType>HTML</reportType> <!-- JUNIT|HTML|XML -->
    <targetSitemap><![CDATA[http://localhost:5432]]></targetSitemap> <!-- atleast one of targetSitemap or scope must be specified -->
    <scope> <!-- atleast one of targetSitemap or scope must be specified -->
        <url><![CDATA[http://localhost:5432/]]></url> <!-- multiple allowed -->
        <exclusions> <!-- optional -->
            <exclusion><![CDATA[localhost:5432/#/logout]]></exclusion>
        </exclusions>
    </scope>
    <false-positives> <!-- optional -->
        <issue>
            <type>5244416</type>
            <path>.*</path>
        </issue>
        <issue>
            <type>5247488</type>
            <path>.*bower_components.*</path>
        </issue>
    </false-positives>
</config>

For an example configuration file, see config.xml and headless-burp-scanner-config.xsd for the xsd

Command-line options

--project-file=VAL          Open the specified project file; this will be created as a new project if the file does not exist (mandatory)
-c (--config) <file>        Configuration file (mandatory)
-p (--prompt)               Indicates whether to prompt the user to confirm the shutdown (useful for debugging)
-v (--verbose)              Enable verbose output

--diagnostics               Print diagnostic information
--use-defaults              Start with default settings
--collaborator-server       Run in Collaborator server mode
--collaborator-config=VAL   Specify Collaborator server configuration file; defaults to collaborator.config
--config-file=VAL           Load the specified project configuration file(s); this option may be repeated to load multiple files
--user-config-file=VAL      Load the specified user configuration file(s); this option may be repeated to load multiple files
--auto-repair               Automatically repair a corrupted project file specified by the --project-file option

Scenarios

The extension has been designed to be versatile and support several scenarios

Scenario A: Scan URL(s) for security issues using Burp

Create a file – config.xml like below and add the URL(s) to be scanned to the scope.

<?xml version="1.0" encoding="UTF-8"?>
<config xmlns="http://nets.eu/burp/config">
    <reportType>HTML</reportType>
    <targetSitemap><![CDATA[http://localhost:5432]]></targetSitemap>
    <scope>
        <url><![CDATA[http://localhost:5432/auth]]></url>
        <url><![CDATA[http://localhost:5432/users]]></url>
        <url><![CDATA[http://localhost:5432/users/1]]></url>
        <url><![CDATA[http://localhost:5432/users?search=asd]]></url>
        <url><![CDATA[http://localhost:5432/bar/foo]]></url>
    </scope>
</config>

Run as shown in the usage section

Scenario B: Scan URL(s) for security issues using Burp but exclude scanning of certain paths

Add an exclusions block to the configuration file.

<?xml version="1.0" encoding="UTF-8"?>
<config xmlns="http://nets.eu/burp/config">
    <reportType>HTML</reportType>
    <targetSitemap><![CDATA[http://localhost:5432]]></targetSitemap>
    <scope>
        <url><![CDATA[http://localhost:5432/auth]]></url>
        <url><![CDATA[http://localhost:5432/users]]></url>
        <url><![CDATA[http://localhost:5432/users/1]]></url>
        <url><![CDATA[http://localhost:5432/users?search=asd]]></url>
        <url><![CDATA[http://localhost:5432/bar/foo]]></url>
        <exclusions>
            <exclusion><![CDATA[localhost:5432/#/logout]]></exclusion>
            <exclusion><![CDATA[localhost:5432/#/users/delete]]></exclusion>
            <exclusion><![CDATA[localhost:5432/#/creepy/crawly]]></exclusion>
        </exclusions>
    </scope>
</config>

Run as shown in the usage section

Scenario C: Scan URL(s) for security issues using Burp but suppress false positives from the scan report

Add a false-positives block with the issue type and path (these can be retrieved from a burp scan report) to the configuration file. You can find more details about Issue Definitions here

<?xml version="1.0" encoding="UTF-8"?>
<config xmlns="http://nets.eu/burp/config">
    <reportType>HTML</reportType>
    <targetSitemap><![CDATA[http://localhost:5432]]></targetSitemap>
    <scope>
        <url><![CDATA[http://localhost:5432/auth]]></url>
        <url><![CDATA[http://localhost:5432/users]]></url>
        <url><![CDATA[http://localhost:5432/users/1]]></url>
        <url><![CDATA[http://localhost:5432/users?search=asd]]></url>
        <url><![CDATA[http://localhost:5432/bar/foo]]></url>
        <exclusions>
            <exclusion><![CDATA[localhost:5432/#/logout]]></exclusion>
            <exclusion><![CDATA[localhost:5432/#/users/delete]]></exclusion>
            <exclusion><![CDATA[localhost:5432/#/creepy/crawly]]></exclusion>
        </exclusions>
        <false-positives>
            <issue>
                <type>5244416</type>
                <path>.*</path>
            </issue>
            <issue>
                <type>5247488</type>
                <path>.*bower_components.*</path>
            </issue>
        </false-positives>
    </scope>
</config>

Run as shown in the usage section

Scenario D: Scan more than just GET requests. Use data derived from running functional tests as input to the scan

Sometimes, just spidering a target scope and and performing on a scope of URLs doesnt give much value. For e.g. when scanning a web application where routing is handled using JavaScript. Burp scans can discover more if it can scan more “real-world” requests and responses. This way, it can attack the target URLs more effectively and potentially discover more than a shot in the dark spider + scan approach.

To handle such cases, it would be best to let the burp proxy intercept some real traffic to the target and build up a sitemap for itself. The Headless Burp Proxy extension provides an simple way to achieve this.

Follow instructions at Headless Burp Proxy and start up burp proxy and remember to set the --project-file option. This is where the “seed” data for scanning is going to be stored.
Configure your functional/integration tests to go through the burp proxy (defaults to 4646 if you use the extension) by setting HTTP_PROXY or similar.
Run the functional/integration tests against the target.
Create a config.xml with the targetSitemap (typically, the base URL of the application), scope, exclusions, false-positives etc.

<?xml version="1.0" encoding="UTF-8"?>
<config xmlns="http://nets.eu/burp/config">
    <reportType>HTML</reportType>
    <targetSitemap><![CDATA[http://localhost:5432]]></targetSitemap>
    <scope>
        <url><![CDATA[http://localhost:5432]]></url>
        <exclusions>
            <exclusion><![CDATA[localhost:5432/#/logout]]></exclusion>
        </exclusions>
        <false-positives>
            <issue>
                <type>5244416</type>
                <path>.*</path>
            </issue>
        </false-positives>
    </scope>
</config>

Run as shown in the usage section and remember to set the --project-file option

tl;dr;

The headless burp scanner plugin can do these

Run burp scan in headless or GUI mode
Specify target sitemap and add URL(s) to Burp’s target scope
Use the “seed” request/response data generated by any integration/functional tests you might have
Mark issues as false positives, these will not be reported in the scan report anymore.
Spider the target scope.
Actively scan the target scope.
Generate a scan report in JUnit/HTML/XML format.
Shut down Burp

Burp Maven Plugin

Maven plugin that allows you to run Burp Suite’s Proxy and Scanner tools in headless mode.

The plugin is essentially a wrapper around the Headless Burp Proxy and Headless Burp Scanner extensions. It offers easy way to integrate security testing using Burp Suite into the project build lifecycle.

Full example

<build>
    ...
    <plugins>
        ...
        <plugin>
            <groupId>eu.nets.burp</groupId>
            <artifactId>burp-maven-plugin</artifactId>
            <version>master-SNAPSHOT</version>
            <configuration>
                <burpSuite>burp/burpsuite_pro_v1.7.31.jar</burpSuite>
                <burpProjectFile>target/headless-burp-project.burp</burpProjectFile>
                <burpConfig>burp/config.xml</burpConfig>
                <headless>true</headless>
                <promptOnExit>false</promptOnExit>
                <verbose>true</verbose>
                <skip>false</skip>
            </configuration>
            <executions>
                <execution>
                    <id>start-burp-proxy</id>
                    <phase>pre-integration-test</phase>
                    <goals>
                        <goal>start-proxy</goal>
                    </goals>
                </execution>
                <execution>
                    <id>stop-burp-proxy</id>
                    <phase>post-integration-test</phase>
                    <goals>
                        <goal>stop-proxy</goal>
                    </goals>
                </execution>
                <execution>
                    <id>start-burp-scan</id>
                    <phase>verify</phase>
                    <goals>
                        <goal>start-scan</goal>
                    </goals>
                </execution>
            </executions>
        </plugin>
        ...
    </plugins>
    ...
</build>

Headless Burp Proxy

Provides an extension to Burp that allows you to run, stop and capture results from the Burp proxy tool in headless mode.

Features

Starts the burp proxy on a provided port (default 4646)
Register a shutdown listener and wait for a shutdown request (default "SHUTDOWN") on port (default 4444).
On receiving a shutdown request, saves the burp project file along with all the information regarding the proxied requests and responses, and finally shuts down Burp

Usage

Start Burp Proxy

On *nix:

java -Xmx1G -Djava.awt.headless=true \
-classpath headless-burp-proxy-master-SNAPSHOT-jar-with-dependencies.jar:burpsuite_pro_v1.7.31.jar burp.StartBurp \
--project-file=project.burp

On Cygwin:

java -Xmx1G -Djava.awt.headless=true \
-classpath "headless-burp-proxy-master-SNAPSHOT-jar-with-dependencies.jar;burpsuite_pro_v1.7.31.jar" burp.StartBurp \
--project-file=project.burp

Command-line Options

--project-file=VAL          Open the specified project file; this will be created as a new project if the file does not exist (mandatory)
--proxyPort VAL             Proxy port
--shutdownPort VAL          Shutdown port
--shutdownKey VAL           Shutdown key
-p (--prompt)               Indicates whether to prompt the user to confirm the shutdown (useful for debugging)
-v (--verbose)              Enable verbose output

--diagnostics               Print diagnostic information
--use-defaults              Start with default settings
--collaborator-server       Run in Collaborator server mode
--collaborator-config=VAL   Specify Collaborator server configuration file; defaults to collaborator.config
--config-file=VAL           Load the specified project configuration file(s); this option may be repeated to load multiple files
--user-config-file=VAL      Load the specified user configuration file(s); this option may be repeated to load multiple files
--auto-repair               Automatically repair a corrupted project file specified by the --project-file option

Stop Burp Proxy

echo SHUTDOWN >> /dev/tcp/127.0.0.1/4444
or
echo SHUTDOWN | netcat 127.0.0.1 4444
or
echo SHUTDOWN | ncat 127.0.0.1 4444

R K

Next PwnBack - Burp Extender Plugin That Generates A Sitemap Of A Website Using Wayback Machine »

Previous « Firework - Tool To Interact With Microsoft Workplaces Creating Valid Files

Published by

R K

Tags: burpsuiteHackingHUNTWebapp

6 years ago

Bomber : Navigating Security Vulnerabilities In SBOMs

bomber is an application that scans SBOMs for security vulnerabilities. So you've asked a vendor…

13 hours ago

Malware

EmbedPayloadInPng : A Guide To Embedding And Extracting Encrypted Payloads In PNG Files

Embed a payload within a PNG file by splitting the payload across multiple IDAT sections.…

13 hours ago

Cyber security

Exploit Street – Navigating The New Terrain Of Windows LPEs

Exploit-Street, where we dive into the ever-evolving world of cybersecurity with a focus on Local…

3 days ago

Hacking Tools

ShadowDumper – Advanced Techniques For LSASS Memory Extraction

Shadow Dumper is a powerful tool used to dump LSASS (Local Security Authority Subsystem Service)…

3 days ago

Hacking Tools

Shadow-rs : Harnessing Rust’s Power For Kernel-Level Security Research

shadow-rs is a Windows kernel rootkit written in Rust, demonstrating advanced techniques for kernel manipulation…

2 weeks ago

Cyber security

ExecutePeFromPngViaLNK – Advanced Execution Of Embedded PE Files via PNG And LNK

Extract and execute a PE embedded within a PNG file using an LNK file. The…

3 weeks ago

Headless Burp – Automate security tests using Burp Suite

Headless Burp Build

Usage

Using a loclally built extension jar

Using the extension from BApp Store

Configuration

Command-line options

Scenarios

Scenario A: Scan URL(s) for security issues using Burp

Scenario B: Scan URL(s) for security issues using Burp but exclude scanning of certain paths

Scenario C: Scan URL(s) for security issues using Burp but suppress false positives from the scan report

Scenario D: Scan more than just GET requests. Use data derived from running functional tests as input to the scan

tl;dr;

Burp Maven Plugin

Full example

Headless Burp Proxy

Features

Usage

Start Burp Proxy

Command-line Options

Stop Burp Proxy

Related Post

Recent Posts

Bomber : Navigating Security Vulnerabilities In SBOMs

EmbedPayloadInPng : A Guide To Embedding And Extracting Encrypted Payloads In PNG Files

Exploit Street – Navigating The New Terrain Of Windows LPEs

ShadowDumper – Advanced Techniques For LSASS Memory Extraction

Shadow-rs : Harnessing Rust’s Power For Kernel-Level Security Research

ExecutePeFromPngViaLNK – Advanced Execution Of Embedded PE Files via PNG And LNK