0sec Labs Open-Sources 0, an AI Security Harness With Web UI, CLI and MCP

What changed

0sec Labs open-sourced a new AI security harness called 0. It supports tasks like code review, pentesting, reverse engineering, and creating fixes. Users can run 0 locally through multiple interfaces: a browser-based web UI, a headless command-line interface, and an MCP server. Importantly, it allows integration with any model provider, so the AI component doesn’t lock users into a specific API or vendor.

Why builders should care

Security workflows often rely on specialized tools that require cloud access or costly proprietary models. With 0 running locally, operators retain control over sensitive data and reduce reliance on external services. The multi-interface approach means 0 can fit into diverse development environments and automation pipelines. Builders gain a flexible AI assistant that blends manual and automated security tasks without exposing their code or vulnerabilities to third parties.

The practical takeaway

For developers and security teams, 0 offers a way to accelerate security assessments while maintaining operational sovereignty. Running AI locally lets users choose models that fit their risk profile and cost structure. The inclusion of reverse engineering and fix generation helps close the loop between identifying and addressing flaws, potentially speeding up remediation cycles. The MCP server adds a way to scale usage across teams or integrate with existing management tools.

What to watch next

Focus will shift to community adoption and contributions now that the product is open source. Watch how many active model providers integrate smoothly and how operators handle the balance between convenience and local control. It will also be interesting to see if 0sec Labs expands support for more complex workflows or integrates with SaaS vulnerability databases to make the toolset more comprehensive.

AI Quick Briefs Editorial Desk