Claude Security Adds Continuous Scanning for Enterprise Repositories
Anthropic has opened continuous scanning in Claude Security as a public beta for Claude Enterprise. After a deep baseline scan, it re-scans repository changes based on risk and complexity, tracks findings as fixes land, and can surface findings through an API and MCP connector.
Claude Security now re-scans connected repositories as code changes and tracks findings through their resolution. Anthropic opened the capability in public beta for Claude Enterprise customers on October 21, 2026; it starts with a deep baseline scan, then checks changes regularly based on risk and complexity rather than promising an immediate scan for every commit (Anthropic's announcement).
How continuous scanning works
After a team connects a repository, Claude Security runs a deep initial scan. It then revisits changes according to their risk and complexity, keeping findings closer to the current codebase without requiring teams to schedule every scan themselves (Anthropic).
Finding status follows the code. Claude Security checks open findings against changed code and marks them fixed when the issue is resolved, including when a team fixes it outside Claude Security. If the issue appears again later, the finding reopens. Anthropic also says each project gets a threat model generated from its code that the team can review and edit, giving later scans project-specific context for what to flag (Anthropic).
This is more than a scheduled scan with a new label: the useful change is the ongoing connection between repository changes and the finding lifecycle. Teams can still review severity and the explanation for how an attacker could reach vulnerable code, and some findings can include an attack-path diagram (Anthropic).
Repository, API, and MCP access
Teams can connect repositories through Anthropic's GitHub app, including GitHub Enterprise Server, or with an access token from GitLab and other Git providers. Projects can contain multiple repositories, each with a selected branch and scope (Anthropic).
The API can list and read projects and findings, resolve or reopen findings, and create or manage projects and repositories. Anthropic says a personal access token can only see and change what its owner can. Webhooks can notify other systems when a finding appears, and the MCP connector lets engineers work with findings from Claude.ai or Claude Code (Anthropic).
Administrators can enable Claude Security in the admin console and use role-based access controls to decide who can view projects, triage findings, or create projects. This makes access and triage roles part of the rollout checklist, not just repository setup (Anthropic).
Availability and usage credits
The continuous-scanning update is a public beta for Claude Enterprise customers. Anthropic says organizations receive $25,000 in free usage credits each month for two months; the offer applies to usage through December 31, 2026, and the full terms are in the Help Center (Anthropic).
For an Enterprise security team, a cautious first step is to pilot a limited repository set, choose the branch and scope deliberately, and agree on who can triage or resolve findings before enabling broader access. That recommendation follows Anthropic's documented project-scoping and role controls; the announcement does not claim that scanning replaces a team's existing review or response process (Anthropic).
Read next
Related coverage- Claude
Claude Security Brings Mythos Scans to Enterprise Customers, Now on Mythos 5.1
Claude Security scans now run on Claude Mythos 5.1 for all Claude Enterprise customers; they moved there from Mythos 5, which powered them from August 21. The public-beta product scans connected codebases, returns CWE, confidence, severity, and suggested-fix details, and opens fixes in Claude Code on the web without exposing the Mythos model directly.
- Claude
Salesforce in Claude Brings Sales Workflows into Claude
Anthropic and Salesforce released Salesforce in Claude in beta. The plugin brings seller accounts, opportunities, and pipeline data into Claude, includes 37 pre-built sales skills, and can draft or apply CRM updates after seller approval. It is available on paid Claude plans after an admin connects Salesforce.
- OpenAI
OpenAI Shares a Defense Factory Playbook for Agentic Cyber Defense
OpenAI is sharing a Defense Factory playbook that connects security tools, skills, reproducible environments, and agents into a continuous loop for discovering, validating, assigning, remediating, and independently verifying vulnerabilities.
- OpenAI
OpenAI Expands Daybreak With GPT-5.6-Cyber
OpenAI expanded its Daybreak cybersecurity program with Blue and Red access tiers and introduced GPT-5.6-Cyber, a purpose-trained model for authorized vulnerability research, exploit validation, and security testing. Access is limited to approved defenders and organizations.
Sources
- Anthropic, Claude Security now scans continuously and tracks fixes (October 21, 2026)
- Claude, continuous-scanning announcement on X (October 21, 2026)
- Claude, API, MCP, and usage-credit details on X (October 21, 2026)
Frequently Asked Questions
What changed in Claude Security?
Claude Security performs a deep baseline scan after setup, then scans repository changes regularly based on their risk and complexity. It checks open findings against changed code, marks findings fixed when the issue is resolved, and reopens a finding if the issue returns.
Who can use continuous scanning?
Anthropic says the feature is in public beta for Claude Enterprise customers. Administrators can enable it and choose who on their team can access it.
Which repository providers are supported?
Anthropic says teams can connect repositories through its GitHub app, including GitHub Enterprise Server, or use an access token for GitLab and other Git providers.