Consistent pseudonymization
An email address gets the same masked value everywhere it appears. Your support team can still follow an issue through a ticket and see what went wrong without seeing who it belongs to.
Ready.Aim.Redactifire!
Redactifire finds and masks PII, CUI, and OUO in log files, HAR captures, and support bundles. It runs inside your own network, and a value gets the same replacement in every file of a job.
The 3 AM scenario
A major defect just took a system offline, and your team can't diagnose it alone. The logs that would explain it are full of usernames, IP addresses, host names, and account data that nobody outside your network may see. You need to get useful information to someone fast, without waiting on legal or exposing customer data.
You search and replace by eye, under pressure, while the system is still down. Miss one IP address and you've leaked it anyway. Redact too much and the file is no use for diagnosing the defect.
It works today and breaks the next time the log format changes or a new field appears. Someone has to maintain it for good, and 3 AM during an outage is the worst time to debug your own redaction tooling.
How it works
01Ready.Drop in log files, HAR captures, or a ZIP support bundle.
Pick a profile, from Light to Max, or tune each entity type: email addresses, host names, IP addresses, SSNs, credit cards, user IDs.
Check every value side by side, then export the clean files and the Decoder Ring.
Works across every file
Upload a whole batch at once. A value redacts the same way in every file of the job, so your vendor can still cross-reference the files.
The Decoder Ring maps every original value to its replacement, so it is as sensitive as the data it protects. Redactifire keeps it on your server and never sends it anywhere. It is also where you decide how to redact: set the mode for a whole entity type, or change one value by itself.
Features
An email address gets the same masked value everywhere it appears. Your support team can still follow an issue through a ticket and see what went wrong without seeing who it belongs to.
Suppress, mask, or pseudonymize, and set the mode for each entity type. For example, keep IP addresses traceable with consistent fake values and fully redact SSNs and email addresses.
Exempt one value from a rule that covers its whole type. For example, redact every email address except the one your vendor needs to see.
Select a host name or API key that detection missed and tag it yourself. Redactifire then redacts it everywhere it appears in the job.
Rules, profiles, and per-value overrides stay in place between sessions. Set them up once, and the next file takes seconds.
Change the text or the rules and the sanitized output updates at once, so you see what a mode change does before you run the job.
Upload a whole batch at once. A value redacts the same way in every file of the job, so your vendor can still cross-reference the files.
Connect your LDAP or Active Directory server, and Redactifire recognizes real usernames and security groups that match no pattern. On an air-gapped network, list known names in a profile instead.
Local accounts and LDAP or Active Directory sign-in are built in. OIDC and SAML single sign-on come with Pro.
Each job keeps its original text, its sanitized output, and its Decoder Ring, so you can prove what was masked and how, long after the job ran.
Your vendor can work with the redacted logs, so they find the problem sooner and your users wait less.
It is one Docker container. Connect your LDAP server and you can sanitize real data in minutes, without a long integration project.
A planned REST API will let your CI/CD pipeline send files for redaction on every build, from GitHub Actions, Jenkins, GitLab CI, or anything that can call an HTTP endpoint. Build artifacts, HAR captures, and test output would be sanitized with no one in the loop.
Deploy on Kubernetes with several replicas behind a highly available endpoint. Scale out across programs or build pipelines, with a shared Oracle database that keeps every node consistent.
Try it
Each cell shows what leaves your network with that starter profile. Hover or tap a cell to see the original value.
hover or tap any cell to see the original
Try the demoWhy Redactifire vs. others
Most redaction tools are made for general documents, or they break the structure of a log. We built Redactifire for one problem: the days you lose waiting for security approval before you can send a log file to a vendor.
You no longer have to redact files by hand, line by line, under pressure, and hope you caught everything before the file left the building. Redactifire does that job for you.
Why self-hosted
Redactifire runs on your own servers: on premises, air-gapped, in your own cloud tenancy, or a mix of these. We never process your logs or the values in them, and neither does anyone else.
There's no multi-tenant SaaS back end, so you have one less system to trust, and a breach somewhere else can't reach your data.
Your logs and the PII in them never leave your network for a third party to process.
Run fully air-gapped for your most sensitive data, or deploy in your own cloud account for a lighter footprint. A mix of both also works.
It runs in Docker or Kubernetes on your own servers, so the only infrastructure you have to trust is yours.
Our pteredactyl. Always on redaction duty, so you don't have to be.