#59 - AI Cloud Analysis, Challenge Pentest Severity and Device Protection for WSL2 and Docker

06 Nov 2026

#Features

• AI Cloud Analysis: Attack Paths and AutoTriage - AI reprioritizes your cloud issues after every scan and shows how an attacker could exploit each one, so you fix real risks first
• Challenge Pentest Severity - Disagree with the severity of a pentest finding? Propose a new one and explain why. The AI weighs your argument and updates the severity only if it holds up
• Device Protection for WSL2 and Docker - Block malicious packages wherever your developers install them

#Aikido Intel & News

Our research team has discovered over 415 new vulnerabilities over the last two weeks!

A busy two weeks for patching: PHP2Shell, Node2Shell and a 9.8-rated Linux kernel TCP flaw all allow remote code execution. On the supply chain side, LiteLLM was compromised again with an infostealer and RAT on PyPI, and the Binance hack shows how much one developer laptop can put at risk.

Not sure where to start? Our checklist for surviving the rest of 2026 shows how to limit what the next compromise can reach.

#Improvements

• Zen - Playbooks
  • Block attacks and attack waves
  • Stop users from abusing endpoints with custom events
  • Learn how to get the most out of Zen and new Playbook feature
• Cloud - Teams can be granularly assigned using Cloud Resource tags
• Deep Review - Chat with agent on findings in your PR/MR
• Code Quality - Autofix now supports your custom rules
• Integrations - ClickUp support for custom fields
• MCP - Codex Plugin 1.0.0 released
• Aikido Terraform - Data source for reading a single repository
• Zen Firewall - Agent releases for Dotnet, PHP, and Golang
• IDE - New releases for VSCode, Jetbrains, and Visual Studio
• API - Additional webhooks for AI Pentesting
• API - Unlink a Jira task
• API - Add domain now supports attack surface management