Your Compliance Policy Ends Where the `TEXT` Field Begins.
Your data policies are perfect. Your database schemas are approved. And none of it matters the moment a well-intentioned user pastes an unredacted document into a generic `Notes` field. This is the gap between intention and reality, and it's where your real risk lives. Annual audits are just photographs of a crime scene months after it happened; Vatra is the live security feed.
Vatra answers your critical questions:
- How can we be sure that no PII exists in our analytics tables, right now?
- How can we detect when users start storing sensitive data in the wrong fields, bypassing our application logic?
- How do we shrink the discovery window for a data spill from months to minutes?
Key Business Values:
- Achieve Continuous Compliance
- Transforms data governance from a periodic, manual audit into an automated, continuous process that provides constant assurance that your data handling policies are being met in your live environment.
- Drastically Reduce Time-to-Detection
- Instead of discovering a data spillage in a quarterly audit, Vatra can detect violations on a continuous basis, massively reducing your window of risk and the potential blast radius of an incident.
- Find Real Violations in Real Data
- Vatra doesn't just check schemas; it performs a "hard look" at your actual production data. This allows it to find not just design flaws, but also user-driven violations where sensitive information is being stored in unexpected places.
- Create a Living, Auditable Exception Record
- The `vatra.exceptions.json` file acts as a version-controlled, auditable record of all accepted risks. When auditors ask a question, the answer is a `git log` away—complete, timestamped, and auditable.
That `varchar(max)` Field Is Not Your Friend.
You built the database correctly. You can't control what users, or a buggy microservice, will dump into it. Every generic text field is a potential data quality time bomb. Vatra is your automated lookout, giving you visibility into the reality of your data, not just its structure.
The "Policy-as-Code" Workflow:
- Codify and Manage Your Data Policies
- Define your data governance rules as code. A library of regex `patterns` (including a rich set of built-in rules for PII/PHI) are grouped into logical `policies`, which are then applied to specific database tables in your version-controlled `audit.json` file.
- Detect Unintended Sensitive Data Exposure
- Vatra connects directly to your database and scans the actual data. If it finds an email address in a `Notes` field where it doesn't belong, it flags a violation and can fail the build, preventing further deployments until the data is remediated.
- Triage Violations and Track Remediation
- Launch the Vatra UI to see all active violations from the latest scan. From the Triage Center, you can add a temporary or permanent exception, or create a work item in JIRA and automatically add a "tracked" exception that is retired when the fix is merged.
The Security Tool That Actually Respects Your Perimeter.
Connecting a tool to your production database is the highest level of trust you can grant. Vatra was engineered for this reality, built on our "Anti-SaaS" philosophy of Zero Trust and Zero External Dependencies. It enhances your security without expanding your attack surface.
Execution Modes:
execute: The primary CI mode. It reads the configuration files from the local repository, scans the live database, generates a report artifact, and returns a pass/fail exit code.interactive: Starts a local web server that provides an interactive UI for building configuration files, managing policies, and triaging violations.serve-docs: Starts a lightweight server to display embedded tool documentation for offline and air-gapped environments.
Secure and Resilient by Design
Vatra is a stateless, ephemeral container that runs as a single step in your existing pipeline. It does its job and disappears, minimizing attack surface. It provides a critical data security gate without requiring you to punch holes in your firewall or grant a third-party SaaS vendor access to your production data.
Licensing & Degraded Mode
If a license cannot be acquired, the tool runs in a Degraded Mode. It logs a warning message and exits with a success code (0), skipping all database scanning and reporting operations. This ensures that a licensing issue does not block your pipeline.