The patch queue got longer, and the fixes got riskier
In 2026, AI models started finding software flaws at a pace nobody had planned for. Published vulnerabilities are on course for roughly double the 2025 count, and Microsoft's July 2026 Patch Tuesday alone fixed about 570. People have started calling it Patchmageddon.
For a business running a .NET application on Azure, the risk sits in two places. The first is the gap between a fix being published and it reaching your app: a vulnerable NuGet package, a self-contained build that never picks up runtime patches, a container image nobody rebuilds. The second is the fix itself. When the April 2026 ASP.NET Core Data Protection flaw (CVE-2026-40372) shipped, many apps had to be rebuilt and redeployed, and a team without a tested pipeline either waited or deployed straight to production and hoped.
The review answers both: what is exposed today, and what it takes to make the next patch routine. Our guide to the ASP.NET Core and Azure App Service security checklist covers the checks you can run yourself first.
What the security review covers
The application and the Azure resources around it, reviewed together, because most exposure lives where the two meet.
Packages and runtimes
Every NuGet package, direct and transitive, checked against known advisories. The .NET version against its support date, and whether the app actually receives runtime patches the way it is published. The free .NET end-of-support checker dates each runtime.
The ASP.NET code paths that matter
Authentication and authorization on every endpoint, antiforgery, input handling and raw SQL, file uploads, Data Protection key storage, error pages that leak detail, and the security headers the app sends.
Secrets and identity
Connection strings and keys in source, config or pipelines. Managed identities and Key Vault references in their place. Who holds Owner and Contributor in the subscription, and which of them still should.
Azure hosting configuration
App Service, Functions and Azure SQL settings: HTTPS only, minimum TLS, FTP and basic publishing credentials, remote debugging, network access, firewall rules, and what Microsoft Defender for Cloud already flags.
The patch path to production
How a fix gets from a package update to the live site: build, tests, staging slot, swap and rollback. If that path is manual or missing, a security update is an outage waiting to happen.
Logging, backups and alerts
Whether you would know about an incident, and could investigate it: diagnostic logs, sign-in and activity logs, database backups and their retention, and who is told when something fails.
Read-only, from the code and the subscription
The Security Review changes nothing in production. You work directly with the senior engineer doing it, remotely, in US Central time.
Access, kept narrow
Read access to the repository and the Reader role on the Azure subscription or resource group. We don't ask for production passwords. If we find shared secrets along the way, we tell you which to rotate.
Scan, then read
Automated checks come first: package audit, Defender for Cloud recommendations, configuration export. A scanner's list is long and mostly noise, so a senior engineer then reads the code and settings behind each finding and decides whether it is reachable in your app.
A written report in one to two weeks
A plain-language summary, then each finding ranked by real exposure, with the fix, its effort and a fixed price. A one-page patch plan says how the next security update should reach production.
You decide. The report is yours.
Fix it with your own team, hand it to another firm, or have us do it. If we do the work, the review fee comes off the price.
What this is, and what it is not
- It is a configuration and code review. We read what you built and how it is hosted, and say where it is exposed. That finds most of what goes wrong in small and mid-sized applications.
- It is not a penetration test. We do not attack the live system. If a customer contract or insurer requires a pen test, we tell you so and the report gives the tester a head start.
- It is not a certification. We flag exposure; we do not certify compliance with SOC 2, HIPAA or PCI DSS. The findings map to the controls those audits ask about.
- You may not need it. If the app is on a supported .NET version, packages are current, Defender for Cloud is quiet and patches already reach production through a tested pipeline, the free call will say so.
- If nobody understands the app, start elsewhere. Security comes after knowing what you have. See software project rescue, or the legacy .NET modernization assessment for .NET Framework applications.
What a security review costs
Three steps, each with a price you can see before you talk to anyone.
A free call or written review
$0 · 30 minutes
Tell a senior engineer what the app is, where it runs and what worries you, and we say whether a review makes sense. Prefer it in writing? Describe the app and get a reply by email within two business days.
Request a free written reviewSecurity Review
$3,500 fixed · 1–2 weeks
A read-only review of one .NET application and the Azure resources it runs on, with a written report of ranked findings, fixed prices and a patch plan. Several applications or a larger estate get a fixed price on the call. The full $3,500 comes off any fix you commission from us.
Remediation sprint
From $12,000 · 2–4 weeks
The fixes the review ranked highest, plus the pipeline that makes the next patch routine: package and runtime updates, secrets moved to Key Vault, hosting settings tightened, a staging slot and tests around the paths that matter. Fixed after the review, approved before work starts.
In the Security Review
- Package, runtime and support-status audit
- Read-only review of the code paths that handle sign-in, data and uploads
- Azure configuration review of the resources the app uses
- A ranked findings report with fixed prices and a patch plan
- A 60-minute walkthrough of the report with your team
Not included, priced separately
- Penetration testing or a formal security audit (we flag exposure; we do not certify)
- Changes to production before you have approved the report
- Microsoft 365, endpoint and office network security
- Incident response for a breach in progress
- Applications outside .NET on Azure (AWS on request)
Reviewed by the people who build and run .NET on Azure
Developers, not a scanner report
Anyone can export a list of 400 warnings. We build and host .NET applications on Azure, so we can tell you which six matter in your code and what each one takes to fix.
Fixes that don't break production
The patch plan is the part most reviews leave out. A security update that takes the site down is its own incident.
Fixed prices, not hourly
The review is a number. Each fix it recommends is a number. You approve each one before work starts.
20+ years combined in the Microsoft stack
.NET, Azure, SQL Server and Dynamics 365, as a Microsoft Partner. For ongoing hosting and support afterward, see Azure cloud support.
Tell us what you run
A few lines are enough: what the application does, its .NET version, how it is hosted on Azure and what prompted the question. A senior Craftware engineer replies by email within two business days (US Central) with whether a review makes sense. It's free, needs a work email, and no sales call is required.
Request a free written review.NET and Azure security review: FAQs
-
Is this a penetration test?
No. We read the code and the Azure configuration and do not attack the live system. If a contract or insurer requires a penetration test, the report tells the tester where to look first.
-
Do you need access to production?
Not to write to it. We ask for read access to the repository and the Reader role on the Azure subscription or resource group. Nothing in production changes during the review.
-
Isn't Microsoft Defender for Cloud enough?
It is a good start and we use its findings. It sees the Azure configuration, not your code, and it cannot tell you which warnings are reachable in your app or how to ship the fix safely.
-
App Service patches .NET for us. Why review it?
App Service updates its built-in .NET runtime, but not the NuGet packages inside your app, a self-contained build or a container image. Those only change when you rebuild and redeploy, which is what the patch plan covers.
-
How big an application does the $3,500 cover?
One .NET application with its database and the Azure resources it runs on, reviewed in one to two weeks. Several applications or a larger estate get a fixed price on the free call before anything starts.
-
Can you review an app built with AI coding tools?
Yes. Where the code came from doesn't change the method, though generated code often repeats one mistake across many files, so we look for the pattern as well as the instance.
-
Our app is on .NET Framework. Can you still review it?
Yes. The report will say plainly what the framework version means for patching, and the legacy .NET modernization assessment covers the move to .NET 10 if that is the answer.
-
Is the review fee refundable if we don't go ahead?
No, but the report is yours: the ranked findings, fixed prices and patch plan can go to any team. If we do the fixes, the full $3,500 comes off the price.
-
Do you work outside Houston?
Yes. The work is remote. We are based in Sugar Land, Texas, and work with clients across the United States.
-
Can we have you sign an NDA first?
Yes. Send yours, or we will send ours.
Not sure where you stand?
That's what the free call is for: 30 minutes with a senior Craftware engineer, or a free written review if you'd rather not meet.