Anthropic Adds Continuous Scanning to Claude Security

Anthropic has expanded Claude Security with continuous code scanning and automatic fix tracking capabilities that monitor whether identified vulnerabilities remain resolved as codebases evolve. The updated version was released on October 21, 2026 in public beta for Claude Enterprise customers, introducing features designed to help security teams track vulnerabilities through remediation rather than relying on periodic scans.

The service scans codebases for vulnerabilities and suggests patches for review, and now includes the ability to detect when previously fixed issues reappear in changed code.

Continuous Monitoring Replaces Point-in-Time Scans

The core operational change shifts Claude Security from scheduled reviews to real-time monitoring. When a team first connects a repository, Claude Security runs a deep baseline scan, then switches to lighter scans focused on each new commit. The tool now monitors connected repositories continuously and rescans automatically whenever something changes.

Projects can include multiple repositories, allowing the system to catch weaknesses that cross service boundaries, such as a backend trusting input that a client was expected to validate. Organizations connect repositories through the Claude GitHub app, which supports GitHub Enterprise Server, or use access tokens for GitLab and other Git providers.

Claude Security checks open findings against changed code and marks them fixed when the correction lands, including fixes completed outside Claude Code. This fix tracking system automatically reopens issues if the same vulnerability recurs in subsequent code changes.

Context-Aware Threat Models and Integrations

The updated service allows teams to customize which risks get flagged. Teams can provide deployment context and edit generated threat models, shaping severity and helping filter risks already handled by the environment. Each project receives a knowledge base where teams can document their specific security requirements and deployment context.

APIs, webhooks, and an MCP connector bring findings into developer workflows. The new API and Model Context Protocol (MCP) integrations enable security teams to connect Claude Security with existing development tools and processes. This integration is particularly relevant given recent announcements from security vendors like Trust3 AI, which has introduced enterprise agent security for Claude Code.

Model Technology and Access Requirements

The service is powered by the company’s full family of models, including Claude Mythos, without requiring Cyber Verification Program access. This contrasts with earlier restrictions that limited access to certain Anthropic security capabilities.

The service is available exclusively for Claude Enterprise customers. Eligible organizations receive $25,000 in free usage credits per month for two months under a launch offer running through December 31, 2026. Organizations that sign up after November 1, 2026 will not receive the full two months of credits.

Adoption and Vulnerability Count

Since Claude Security entered public beta in April 2026, thousands of organizations have used it to find over a million vulnerabilities. However, this figure represents discoveries rather than confirmed repairs. The new tracking features are designed to provide visibility into which vulnerabilities have been addressed and which remain open.

Jake Meredith of Gusto’s product security team said Claude Security scanned the company’s code once to set a baseline and now keeps up as the code changes, while Nathan Hunstad, director of security at Vanta, said Claude Security found a number of logic flaws that none of their other tools had caught.

Integration with Claude Code and Safety Concerns

The continuous scanning update arrives as Claude Code usage increases within development teams. Anthropic platform engineering head Katelyn Lesse recently stated that Claude writes all the code produced by her team, with engineers focusing on system design, verification, and oversight.

The relationship between Claude Security and Claude Code has raised questions about how safety classifiers handle security audit outputs. A recent GitHub issue reported that Claude Code blocked the final HTML report of a long, authorized defensive security audit, raising concerns about safety-classifier behavior and transparency during extended agent workflows.

Users should also be aware of potential security dependencies. While Claude Code itself does not rely on Node.js, connected development tools may be affected by the 10 Node.js vulnerabilities addressed in recent releases, particularly for users running Node-based tools such as Playwright MCP.

Recent Platform Updates

The security update coincides with other Claude Code enhancements. Claude Code version 2.1.307 adds a /claude-api api-version command to support migration to API version 2026-10-21, along with changes to safety-event recording and HTTP MCP fixes.

Key Facts

Claude Security update released October 21, 2026 for Claude Enterprise customers in public beta
Continuous scanning monitors code changes with automatic rescans on each commit after an initial baseline scan
Fix tracking automatically marks vulnerabilities resolved and reopens them if they recur
Over one million vulnerabilities discovered since April 2026 public beta launch
$25,000 monthly usage credits for two months available through December 31, 2026
Multi-repository projects supported to catch cross-service vulnerabilities
API, webhooks, and MCP connector enable integration with existing developer workflows

Sources

Written by

Mike Gingerich

Author

Author: Mike Gingerich, President of web firm Digital Hill, Co-Founder of TabSite .
Digital and Social Media Marketer, Speaker, and Business Consultant. Part geek, part marketer, total digital junkie! Seeking to add value, make the complex simple, and leave a positive impact.

Follow me on twitter: @mike_gingerich.