Secure Code Comments: One Easy Way to Steward Your Application Security Culture
While traditional code comments focus on explaining the code's functionality, security-focused comments are crucial to promoting secure coding practices throughout the development lifecycle (SDLC). By making this simple tactic part of your natural workflow, you can assert your knowledge and become a security champion. Let's explore how integrating security comments into your code can benefit you and the security team. Leading Forward Using Secure Code Comments Integrating security into your daily coding isn't just about ticking requirement boxes; it's about building a security mindset that makes you indispensable. Secure-code comments are low-hanging fruit for sharing knowledge, learning from others, and making security a seamless part of your day. Senior developers and application security champions can quickly and effectively educate other developers about best practices without leaving the comfort of their Integrated Development Environment (IDE). Best practice for code comments suggests emphasizing the why, not the what. Security-focused comments are no different. Meanwhile, they play a crucial role in promoting secure coding practices, enabling teams to: Explain Key Security Moves: Share the rationale behind specific security measures, such as input validation, encryption, and access control mechanisms. Flag Red Flags: Spot potential weaknesses in your code, like SQL injection, cross-site scripting (XSS), and unprotected data. Share Knowledge: Link to relevant security standards, guidelines, and resources and facilitate efficient code reviews. Enhancing Code with Security Comments–Two Examples Example 1: Preventing SQL Injection with Parameterized Queries (Python) Let’s consider a simplified Python function, which performs a simple insert operation into a database: def insert_user(conn, name, email): """ Inserts a new user into the 'user' table. Args: conn: A sqlite3 connection object. name: The name of the user. email: The email address of the user. Returns: None This function uses a parameterized query to prevent SQL injection vulnerabilities. See: CWE-89 https://cwe.mitre.org/data/definitions/89.html By using placeholders (e.g., `?`) and passing the actual values as separate arguments, we avoid direct string concatenation. This ensures that user-supplied input cannot be manipulated to modify SQL commands. """ sql = """INSERT INTO user (name, email) VALUES (?, ?)""" cur = conn.cursor() cur.execute(sql, (name, email)) conn.commit() cur.close() As you can see, in addition to the regular docstring, we succinctly mention why we’re using parameterized queries over string concatenation. We also reference a CWE and provide a link for anyone who wants to learn more. With just three extra sentences in a function comment, we’ve given less experienced developers who are code spelunking a quick lesson (or reminder) about why and how to prevent SQL injection. Example 2: Mitigating XSS Vulnerabilities with DOMPurify (React) Let’s take a look at another example, this time on a React frontend. Here, we’re knowingly doing something potentially dangerous but effectively communicating to other developers the mitigations applied. /** * Displays user-generated HTML content, sanitizing it with DOMPurify to prevent XSS vulnerabilities. * * This component uses `dangerouslySetInnerHTML` because the content being displayed *must* include HTML markup. * Alternatives like rendering plain text or using a limited subset of HTML tags are not sufficient for this use case. See: https://kanban.system/t/123 * **Security Considerations:** * * **CWE-79 (Improper Neutralization of Special Elements used in an HTML Page): https://cwe.mitre.org/data/definitions/79.html * This code directly addresses CWE-79 by sanitizing the user-provided HTML before rendering it. Without sanitization, malicious * users could inject JavaScript code that would be executed in the context of the website, leading to * XSS attacks. * * **Why not just use textContent?** If we used `textContent` or similar methods, any HTML tags in the user * input would be treated as plain text and displayed as-is. This would prevent XSS, but it would * also defeat the purpose of allowing users to input HTML in the first place. * * **Why DOMPurify?** DOMPurify is a widely used and well-maintained library specifically designed for * sanitizing HTML. It's more robust and secure than attempting to create a custom sanitization * solution. It handles a wide range of potential XSS attack vectors. * * @param {string} htmlContent The user-generated HTML content. This is assumed to be untrusted. * @returns {JSX.Element} The sanitized HTML rendered within a div. */ function SafeHTMLDisplay({ htmlContent }) { const sanitizedHTML = DOMPurify.sanitize(htmlContent); return ( <div dangerouslySetInnerHTML={{ __html: sanitizedHTML }} /> ); } This time we go into more detail about the why. Let’s break it down: First, it’s important to acknowledge that the original approach isn’t best practice. Second, you can level up developer awareness about alternative options.Then, connect the dots for maximum impact, sharing why this approach is required to satisfy product requirements. Finally, it’s important to detail security considerations with CWE IDs, codifying the weakness you’re proactively mitigating; yes, you can even justify the introduction of another dependency compared to a custom implementation. Any future developer tasked with modifying the comment feature will quickly understand the importance of keeping this mitigation rather than “cleaning up the code” because it still functions “the same” without it. As an AppSec developer or security champion, you’ve just avoided another security report being raised because of a regression introduced by an over-eager junior developer. It's well worth the 15 lines of extra code. Even better, any security engineer performing a secure code review will be much more confident that their developer understands why they wrote the code the way they did. This knowledge, in turn, expedites that coveted ‘approve’ on their pull request, reducing the time to get the code safely into production. Identifying and codifying vulnerabilities with Find the Flaw Setting the tone with security-focused comments largely falls to the lead developer or security champion, presenting an easy opportunity for aspiring champions to stand out. Remember, the goal is to identify and codify weaknesses in code before or as it is being written; this approach ensures others can craft easy-to-understand security comments too. Whether you’re just starting out or looking to grow your skillset, Immersive AppSec’s Find the Flaw collections provide ample opportunities to build critical DevSecOps muscle memory. You’ll learn to identify various common vulnerabilities in code and recognize what CWE IDs they correspond to. Writing security-focused comments will feel like second nature when you're coding up a storm! Beyond Code Comments: Empowering Your Manager to Recognize the Power of AppSec Training Code comments are a valuable AppSec tool, albeit only one piece of the puzzle. To cultivate a developer-led security culture, organizations need managers who recognize the power of comprehensive training programs for their elite developers. These programs support proactive developers with the knowledge and skills to build secure applications from the ground up. As a result, the organization achieves development velocity SLAs and application security simultaneously. Remember to share your experience learning by doing, gaining the attackers’ perspective, which Immersive Labs AppSec offers. Your manager and teammates should recognize the value of using safe, real-world scenarios and interactive exercises, such that the training you do (now) targets the problems you have (now). Share your thoughts Check out this Find the Flaw collection and then share your thoughts with The Human Connection community: For developers: Does adding security rationale to your comments feel like an ‘easy enough' lift? For security champions: Are you already using this technique or something similar? How have you convinced other developers to adopt this style of commenting?Human Connection Challenge: Season 1 – Scanning Walkthrough Guide (Official Version)
Time’s Up! Congratulations to everyone who completed Lab 2: Scanning from the Human Connection Challenge: Season 1. In this walkthrough, I'll share some strategies for efficiently completing the lab, based on my perspective as the author. Remember, there are often multiple ways to approach a challenge, so if you used a different method and succeeded, that's perfectly fine! The goal is to learn, and I hope these notes help clarify any steps and reinforce key concepts for the next challenge. This challenge has now ended, but the lab remains available for practice. While prizes are no longer up for grabs, you can still complete the lab and use this walkthrough guide for support if needed. I’ve also used placeholders in some of the commands that would give away an answer directly, so if you see anything enclosed in angle brackets, such as <name server>, please make sure you replace it with the actual value, such as nameserver. With all that considered, let's get started. Overview Task: Identify the name server records of tinytown.bitnet. 1. What is the IP of the first name server for tinytown.bitnet? You’ll first need to open a Terminal on the Kali desktop. Next, you’ll need to query the DNS Server IP (found in the Machines panel) about the tinytown.bitnet domain using the nslookup (Name Server Lookup) tool. You’re specifically looking for NS (Name Server) records, so you can use the -type=ns parameter with nslookup to specify this: nslookup -type=ns tinytown.bitnet [DNS Server IP] The output of this command will return two name servers for the domain labelled with 1 and 2. Your next step is to identify what IP address is associated with the first name server (1). To do this, you can use nslookup along with the name server, domain, and DNS Server IP: nslookup <name server>1.tinytown.bitnet [DNS Server IP] This command will then return an IP address for the name server. 2. What is the IP of the second name server for tinytown.bitnet? As you’ve already identified both name servers, you’ll just need to run the previous command, except with the second (2) name server: nslookup <name server>2.tinytown.bitnet [DNS Server IP] You’ll then find the IP address associated with it. Task: Identify port service information for Target 1. 3. What service version is running on port 53? A network scanning tool like Nmap can help you identify the service version running on a specific port. To do this with Nmap, you can use the -sV option for service detection: nmap -sV [Target 1 IP Address] The output will show what service version is running on port 53. 4. What is the full service banner of port 22? There are a couple of ways to find the full service banner of port 22 – such as with Nmap or Netcat. If you’re using Nmap, you can modify the previous command to include the “banner” script along with the port number: nmap -sV -script=banner [Target 1 IP Address] -p22 The command line will then display the service banner from port 22. You can alternatively use netcat to manually connect to the SSH server. When a client connects, Netcat may present a banner that contains version information. To use Netcat, you’ll need the nc command along with the Target 1 IP address and specify you want to connect to port 22: nc [Target 1 IP Address] 22 When you run this command, the banner appears before the terminal hangs. Task: Identify a token on one of the ports. 5. What is the token? With the previous Nmap command, you initially found that three ports were open on Target 1. However, you’ll need to do a more thorough network scan to find another open port, one not initially found with the previous scans. To do this, you can expand your port scan to cover a much wider range by using Netcat to scan for open ports from 1 through 9000: nc -zvn <Target 1 IP Address> 1-9000 Here, -z will scan for listening services but won’t send any data, -v is verbose mode, which provides more detailed information, and -n tells Netcat not to resolve hostnames via DNS. This command will reveal a fourth open port. Now, you can use Netcat to connect to this port: nc <Target 1 IP Address> <open port> The token will then be displayed in the terminal. Task: Scan the TLS configuration on Target 2. 6. How many protocols are enabled? To scan for SSL/TLS configurations, you can use the sslscan tool. By default, sslscan scans port 443 and will return supported server ciphers, certificate details, and more. You can use sslscan like this: sslscan <Target 2 IP Address> The returned output will be verbose, but you can find and count the number of enabled protocols under the SSL/TLS Protocols subheading. 7. Name an enabled protocol. Using the previous output, name one of the enabled protocols. 8. What exploit are the protocols NOT vulnerable to? Using the same output, scroll down through the results until you find a subheading that’s named after a vulnerability and contains a similar string to: <Protocol> not vulnerable to <vulnerability name> The vulnerability has the same name as the subheading. Task: Identify and extract information from an SMB share on Target 3. 9. What Disk shared directory can you access? To extract information from an SMB (Server Message Block) share, you can use the smbclient tool. First, you’ll need to list the SMB shares on the target using the -L flag (the list/lookup option) with: smbclient -L //<Target 3 IP> You’ll then be prompted for a password, but you can press Enter to skip this. A list of SMB shares will then be displayed, three of which are shown to be a Disk type, so you know the answer will be one of these. You can now begin to go through the list and try to connect to the shares with: smbclient //<Target 3 IP>/<Sharename> However, this time when you’re prompted for a password and you press Enter, you might encounter a message when you try and connect to a share: NT_STATUS_ACCESS_DENIED If you attempt to connect to all shares, you’ll find you can connect to one share without a password. You’ll then be greeted with the following prompt to show the successful connection: smb: \> 10. What is the token stored in the directory? Now that you’re connected, you can execute commands to interact with the SMB share. If you run ls, you’ll find a token.txt file in the current directory. You can then download the file from the share onto your local machine with: get token.txt On the Kali desktop, open the Home folder and the token.txt will be inside. Open this file and find the token. 11. What is the username stored in the directory? After you’ve run ls in the SMB share, you’ll find not only token.txt, but also a file named creds.txt. Use the same command as you just did previously to download the file onto your machine: get creds.txt This file will also be downloaded to the Home folder, where you can find a username and password. Task: Identify open services on Target 3. Task: Connect to Target 3 with the previously found credentials. 12. What is the token stored in the user's /Documents directory? For this final task, you first need to scan the target using Nmap. You’ll find that if you attempt to scan the target without using the -Pn flag, you’ll get a response saying that the host seems down. However, if you run Nmap with -Pn, you’ll find some ports are open: nmap -Pn <Target 3 IP Address> However, the ports returned from this command don’t offer a way to connect to the target. You’ll also need to scan the 6000 most popular ports: nmap -Pn --top-ports 6000 <Target 3 IP Address> These results will now show two additional ports are open regarding the Web Services Management (Wsman) protocol, which is used to communicate with remote machines and execute commands. One of the tools that implement this protocol is Windows Remote Management (WinRM) which is Microsoft’s implementation of Wsman. Knowing this, you can now use Metasploit to interact with the target. In your terminal, run: msfconsole Once loaded, you can use the the following auxiliary module to connect to a system with WinRm enabled and execute a command with: set cmd ls You’ll then need to set the following options, using the credentials you found in the creds.txt file: set username <username> set password <password> set rhosts <Target 3 IP Address> Next, you need to set the cmd option with the command you want to run. If you use the ls command, you’ll be able to find what out files are in the directory you connect to: set cmd ls With all the options set, you can now run the module: run The results of the executed command will be printed on the screen and also saved to a directory, but both show the existence of a token.txt file in the current directory. You can now set the cmd option to type token.txt in Metasploit: set cmd type token.txt Once set, use the run command to send the updated command: run The contents of token.txt will then be displayed on the screen and outputted to a file. Tools For this challenge, you’ll use a range of tools including: Nslookup Nmap Netcat Sslscan Smbclient Metasploit Tips You can use different tools and parameters within those tools to scan for and find information, so don’t be afraid to try out a few different things! If you want to learn more about some of the tools within this lab, take a look at the following collections: Reconnaissance Nmap Infrastructure Hacking Introduction to Metasploit Post Exploitation with Metasploit Conclusion The steps I’ve laid out here aren’t the only way to find the answers to the questions, as long as you find the answer, you did it – well done! If you found another way to find some of these answers and think there’s a better way to do it, please post them in the comments below! I hope you enjoyed the challenge and I’ll see you for the next one.
