Parallel code-review lanes converge on a government security checkpoint
Parallel code-review lanes converge on a government security checkpoint
+ Anthropic News

Alberta used Claude Code to scan 466 million lines of government code

Anthropic says Alberta used Claude Code agents to review legacy government systems, find vulnerabilities, generate fixes, and build continuous security-review agents.

Anthropic says the Government of Alberta used Claude Code with Opus and Sonnet models to scan 466 million lines of government code in 20 hours.

The July 6 case study describes a team inside Alberta’s Ministry of Technology and Innovation using about 50 agents in parallel across roughly 1,280 applications and 3,400 repositories. The work targeted security vulnerabilities, infrastructure weaknesses, deployment gaps, and missing technical documentation.

Anthropic says the scanning routine used a two-stage process: a rules engine first flagged known patterns, then Claude Code reviewed those flags and cited exact files and lines so developers could verify the findings.

The claim is scale plus review

The Alberta case study is not just a coding-agent demo. It is a government modernization story, because the systems involved hold sensitive records across ministries, including tax, procurement, and social-services data.

Anthropic says Claude Code could often generate a fix, test it, and build it after a vulnerability was found. Where automated tests were missing, Claude wrote tests first. Where a system was too old or complex to patch efficiently, the team could rebuild it in a more maintainable language.

The case study still keeps humans in the release path. Anthropic says patches were reviewed and approved by ministry engineers before shipping.

That distinction matters. A scan that produces 20 hours of findings can create its own risk if teams cannot triage, verify, test, and approve the output.

Alberta is turning the scan into process

The more durable part of the story is continuous review. Alberta’s cybersecurity team built specialized Claude review agents on top of the Claude Agent SDK.

Anthropic describes a red-team agent that probes applications from the outside and maps possible exploitation paths, followed by a blue-team agent that checks defenses against an international security standard and writes remediation plans with file-level pointers. Additional agents review code quality and public-facing writing.

The case study says every application is checked against roughly 95 security controls on each pass.

Alberta also plans to expand from review and remediation into broader modernization. Anthropic says one ministry has 185 legacy production applications that the government wants to consolidate into 16 reusable applications.

Sources

The AI Feed Desk

The AI Feed Desk

Editorial desk

The AI Feed Desk tracks AI provider updates, model releases, agent tooling, and enterprise adoption, turning fast-moving announcements into source-linked context for builders and operators.

Noticed a typo, incorrect information, or translation error?

Tell us so we can fix it.

Help Improve This Article

Related Articles

Anthropic's official article illustration showing a secure laptop

Anthropic maps AI-enabled cyber threats to MITRE ATT&CK

Anthropic analyzed 832 banned malicious cyber accounts and found AI use moving from basic access work into lateral movement, account discovery, and chained attack activity.

The AI Feed Desk

By The AI Feed Desk

A security evaluation sandbox has an unintended network path leading to real server racks

Anthropic says Claude cyber evals reached real systems

Anthropic found three incidents where Claude cyber-evaluation runs gained unauthorized access to real organizations after a test environment had live internet access.

The AI Feed Desk

By The AI Feed Desk

Prompt cards pass through a policy checkpoint before entering a model core

Anthropic adds Inference hooks for Claude Enterprise prompt control

Anthropic put Inference hooks into beta for Claude Enterprise, letting governed prompts pass through an organization's security server before Claude processes them.

The AI Feed Desk

By The AI Feed Desk

A governed cloud workspace connects an AI model core to a high-performance compute rack

Claude reaches Microsoft Foundry with Azure governance and GB300 compute

Anthropic made Claude generally available in Microsoft Foundry, while NVIDIA framed the Azure deployment as a GB300 Blackwell Ultra agent platform.

The AI Feed Desk

By The AI Feed Desk

A lab workbench connects a molecule model, protein structure, and code notebook through a visible audit path

Claude Science puts scientific AI inside a traceable workbench

Anthropic's Claude Science beta gives researchers a workbench with curated scientific skills, compute access, auditable artifacts, and reviewer agents.

The AI Feed Desk

By The AI Feed Desk