Claude Security

What it does

Claude Security analyzes source repositories to find vulnerabilities that depend on how an application actually works. It traces behavior across files, validates potential findings, and proposes patches for a team to review. The managed service starts with a baseline scan and continues examining changes, so findings can be tracked through to remediation.

Teams can set repository scope, edit the threat model, examine the reasoning behind a finding, and bring results into existing processes through the API, webhooks, and an MCP connector. A typical workflow is to review a prioritized finding, open the affected code in Claude Code, assess the suggested fix, and mark the issue resolved after the change ships.

The hosted product is in public beta for Claude Enterprise, with administrator setup, repository access, eligible user seats, and usage billing required. A related Claude Security plugin runs inside Claude Code and uses the models available to that account; it is not the same service or the same access to Mythos-powered scans. Both require human review of security findings and proposed code changes.

What it can do

• Scan repositories for context-dependent vulnerabilities
• Validate and prioritize findings
• Propose reviewable security patches
• Track findings as code changes
• Export findings through APIs and webhooks

Works with

• GitHub
• GitHub Enterprise Server
• Git repository credentials
• Claude Code
• MCP
• Webhooks

Example workflows

Scan to remediation

Connect a repository, review validated findings, inspect a proposed patch, and track the accepted fix through deployment.

Feed the existing security queue

Use the API or webhooks to bring findings into your team’s tracking and reporting tools.

Before you choose

• Hosted Security and the Claude Code plugin have different eligibility, runtime, and model access.
• Scans consume tokens; suggested fixes are not automatically safe or complete.

Listed under

• Security
• App and code security
• Appsec
• Code review
• Code security
• Enterprise
• Patch suggestions
• Security teams

Get started

Enable Security and connect repositories

1. Have an Enterprise admin enable Security, grant repository access, and configure spend limits.
2. Ensure the user has an eligible seat and permission to create a Security project.
3. Create a project, choose repository scope, and review the initial scan.

The setup guide states direct token cost with no additional scan platform fee. The related Code plugin follows its own Claude Code access and usage.

Source: academy.claude.com/tutorials/getting-started-with-claude-security, checked 28 Oct 2026. Prices can change; check before you buy.

Sources for this overview

Product pages, documentation, and pricing checked 28 Oct 2026. Each source supports part of the overview; availability and prices can change.

• Claude Security | Claude by Anthropic ↗claude.com
• Plans & pricing | Claude by Anthropic ↗claude.com
• Getting started with Claude Security · Claude Academy ↗academy.claude.com
• Scan your codebase for vulnerabilities - Claude Code Docs ↗code.claude.com

Company website

claude.com: product, pricing and policy pages

YC directory mirror

Batch and company name

Web search

Confirms the official site

Social accounts

Accounts the company links, and its posts on X

Human review

Checked by a person

knowledge_base

Last checked 28 Oct 2026. Every field keeps its source and date; unknown means we couldn’t confirm it.