Skip to content

Release Notes — March 27, 2026

You can now collaborate on vulnerabilities with inline comments and configure the 100k+ rules we apply when scanning for vulnerabilities.

Vulnerability comments are now available. Your team can discuss findings, add context, and coordinate remediation directly on any vulnerability — no more switching between tools to collaborate on security issues.

You now have full control over which checks run during your scans. Customize rule severity, disable noisy checks, and filter by category — with improved sorting performance and a polished management experience. If you find yourself repeatedly ignoring vulnerabilities based on a specific rule, you’ll now be prompted to consider disabling the rule to reduce noise in the future.

Fencer now automatically detects expiring SSL certificates and creates vulnerabilities so your team is alerted before a certificate lapses — preventing downtime and security warnings for your users.

You can now disable scanning for auto-detected assets individually, giving you granular control over your scan scope without removing the assets from your inventory.

New invertible filters, event detail filter actions, and improved quick search make it faster to find and investigate security events in the SIEM module.

A new timespan selector for storage metrics and an informational notice in the Data Ingestion section give you better visibility into your data ingestion volumes over time.

All integrations now report their health status, making it easy to spot broken or misconfigured connections at a glance. Connections that need attention are clearly flagged so you can fix them before they affect your security posture.

Manual re-sync buttons are now available on Vanta and Drata integrations, giving you direct control over when vulnerability data is synchronized.

You can now enable or disable vulnerability syncing per GRC integration, with automatic asset scope synchronization to keep everything aligned.

The Jira integration now properly handles secrets with multiple associated vulnerabilities, creating accurate linked issues for each.

Better pull request annotations for Bitbucket, fixed redirect handling for Git connections, improved SSH key processing, and general connection reliability improvements across all Git providers.

Added Kinesis ingestion support for API Gateway, improved GCP monitoring notification handling, and fixed authentication in the authorizer Lambda for more reliable data ingestion.

The scan settings page has been redesigned for clarity. Scan throttling is now smarter, scanner page headers stay visible while scrolling, and unnecessary navigation links have been cleaned up.

The fixer panel now links directly to errored fix attempts, the “Has Fix” column is available in the vulnerability grid as a “Has Comments” column, and multi-secret vulnerability views have been improved.

  • Resolved vulnerabilities now correctly sync as resolved in Vanta, and HTTP 400 errors during package vulnerability sync have been fixed.
  • Vulnerability comments now display correctly on the detail page.
  • Network scans now work properly with IPv6 targets.
  • Fixed several scan reliability issues including server errors, scan history not updating after manual triggers, and background task timeouts.
  • Fixed SIEM issues including IP address matching in raw data, GitHub data ingestion showing zero bytes, and Postgres log regex truncation.
  • Fixed dark mode styling in the payment flow, Bitbucket annotation length limits, ARN resource names with spaces, and Google Workspace access errors.
  • Onboarding steps now only show for features available on your plan, and organization subscriptions are properly cleaned up on deletion.
  • Application scan errors are now visible so you can understand and act on scan failures.
  • Fixed default sort order to show latest items first.
  • Fixed scan failure notification email layout alignment.

You can now request penetration tests directly from within Fencer. Assign testers, track request status, and receive email notifications at every step — from initial request through assignment and completion.

Scan your external network services for vulnerabilities across multiple cloud providers, including Cloudflare. Results are automatically deduplicated across providers and presented in a unified view with detailed vulnerability content and configuration controls.

Import identities in bulk via CSV upload. A preview step lets you review exactly what will be created before committing, making it easy to onboard large teams or migrate from other tools.

Add vendor identities manually and assign owners to each vendor. This gives you complete visibility into who has access to what, even for vendors without an automated integration.

A complete redesign of domain scanning is in early access, with structured DNS and SSL record handling, certificate transparency monitoring, and improved domain discovery.

A new code analysis engine is being validated in shadow mode alongside the existing scanner. Once fully verified, it will deliver broader detection coverage with faster scan times.

A modernized vulnerability detail view is in testing, featuring integrated comments, an improved fixer panel, and bulk actions — all in a faster, more responsive layout.

Code review annotations are being refined to only flag issues in the files you actually changed, reducing noise and making scan results more actionable during code review.

A new identity integration with MailerSend enables automatic user discovery and access tracking for your email infrastructure.

Interested in early access to any of these features? Reach out to us and we’ll get you set up.