Building Redactinator: How I Used AI and Python to Solve a Real-World Redaction Problem

As someone with a background in IT, cybersecurity, and compliance, I’ve worked with more infrastructure logs, debug files, and diagnostic reports than I can count. And if there’s one constant across every environment I’ve touched, it’s this: troubleshooting often reveals more than it should.

When preparing files for external vendors, audits, or cross-team collaboration, I constantly found myself scrubbing sensitive information—IP addresses, hostnames, environment-specific metadata—by hand (good old find/replace or some regex magic). Sometimes this meant scanning through gigabytes of (.log) or (.txt) files. Other times, it meant digging into ZIP archives, changing folder names, and ensuring nothing sensitive slipped through. It was time-consuming and tedious.

That’s the problem that sparked the idea for Redactinator.

From Need to Tool

I didn’t set out to build an application—I just wanted a better way to protect sensitive data. But after experimenting with Python scripts to strip out IP addresses using regex, I realized there was potential to create something more robust. With the help of AI, I refined and expanded the tool’s capabilities:

• Recursive redaction of IP addresses from (.txt), (.log), and ZIP files
• Detection and removal of custom hostnames
• Cleaning of file and folder names (yes, even inside archives)
• In-place updates with logging for auditability
• A simple GUI for usability, and eventually, packaging as a (.exe)

I leaned on AI not just for debugging but for brainstorming UI ideas, icon design, and even packaging tips. What started as a quick automation script grew into a lightweight but powerful redaction tool built for people just like me—technical professionals who don’t want to risk exposing their infrastructure details in a rush to share logs when under the stress of timely troubleshooting.

Real-World Use: Cisco Firewalls and ESXi Hosts

One of the first places I put Redactinator to work was on support bundles from Cisco Firepower firewalls and VMware ESXi hosts. These diagnostic packages are often large, deeply nested, and packed with sensitive internal details—IP addresses, device hostnames, configuration paths, and more. Redactinator quickly sanitized these files (over 300 files and 150,000 redactions in less than a minute). The tool allowed me to share these bundles with support teams and vendors without compromising network security and while spitting out a log for our own compliance purposes. This alone saved hours of manual and tedious work.

Why It Matters

Redactinator doesn’t rely on the cloud. Everything runs locally. That was a non-negotiable design choice. Whether you’re working in a NERC CIP environment, dealing with SOC 2 preparation, or simply cleaning logs for a support ticket, your data stays on your device. No uploads. No remote servers. No exposure.

I built Redactinator because I needed it—and I know I’m not the only one.

The Result

Redactinator now gives me peace of mind. It’s a fast, transparent, and repeatable way to clean sensitive files before sharing them. What used to take an hour of meticulous searching and replacing can now be handled with a click. I packaged it as a (.exe), gave it a custom icon, and even designed a simple website to support it.

It’s not a flashy SaaS product. It’s a purpose-built tool for professionals who just need redaction done right—and done fast.

If you’ve ever worried about leaking IPs or hostnames in a file you’ve shared, Redactinator might be for you too.

What's Next: Expanding Redactinator

I’ve not a developer and have spent my career thus far on the operational side of IT. I’ve put scripts together and can CLI pretty well. However, dipping my toes into developing some custom software has been exciting and I look forward to the iterative process of finding and killing bugs, and expanding features. Thanks to AI this is easier than ever and the learning curve has lessened significantly.

I'm already planning the next phase of Redactinator's evolution. One major feature on the roadmap is a dedicated PII module—capable of identifying and redacting personally identifiable information such as Social Security Numbers (SSNs), dates of birth (DOB), phone numbers, and email addresses. This would make the tool even more powerful for industries dealing with customer or patient data.

Additionally, I want to introduce a non-destructive scanning mode. Instead of redacting data, this feature would scan and analyze files to identify patterns, summarize findings, and present the results using clear visualizations. Organizations could then use this insight to inform decisions, enhance compliance audits, or improve internal data handling practices—without having to scrub anything upfront.

Wrapping Up

Please check out the tool and let me know if you have any feedback. (it’s free) Head to https://redactinator.com/