TL;DR

Every vendor you trust is a door into your business — and attackers are walking through them faster than ever. In 2025, third-party breaches doubled in frequency, extortion attacks surged 63% to 6,800 incidents, and the average supply chain compromise now takes 267 days to detect and costs $4.91 million. This post breaks down three real-world breach patterns, explains what went wrong, and gives you the exact security demands to make of every vendor this week.

Third-party risk isn't a theoretical problem for large enterprises anymore — it's the primary attack surface for businesses of every size. When Black Kite's 2026 Third-Party Breach Report verified that each compromised vendor exposed an average of 5.28 downstream companies, the old "it won't happen to us" argument collapsed. Your data is only as secure as the weakest vendor in your supply chain.

The Software Update That Shut Down the World

In July 2024, CrowdStrike — one of the most trusted names in endpoint security — pushed a faulty configuration update to its Falcon sensor. The result wasn't a hack, but the blast radius was indistinguishable from one: approximately 8.5 million Windows devices crashed into a blue-screen boot loop, grounding airlines, freezing hospital systems, and halting financial transactions globally. Insurer Parametrix estimated insured losses at $540 million to $1.08 billion, while total economic damage ran into the tens of billions.

The root cause was a failure in CrowdStrike's own internal testing and staged deployment pipeline. The update bypassed gradual rollout controls and hit millions of endpoints simultaneously. For the businesses affected, the lesson was brutal: even a security vendor you pay to protect you can become the single largest operational threat to your business.

What should have happened: CrowdStrike should have enforced staged, ring-based deployments — pushing updates to 1%, then 10%, then broader cohorts — with automated rollback triggers. On your side, your business should maintain vendor-outage contingency plans and test them regularly.

The Ransomware Pipeline Through Your ERP Provider

Manufacturing and consulting firms were the most-ransomed sectors in 2025, according to Intel 471's dark-web analysis. Attackers aren't breaching factories directly. They're compromising the managed service providers, ERP vendors, and cloud platforms those manufacturers depend on, then pivoting laterally into production networks. When the ransomware hits, it encrypts not just one company's data but every client connected to that shared platform.

A February 2026 incident tracked by PKWARE involved threat actor "Zestix" selling batches of corporate data stolen from multiple companies through a single compromised SaaS provider. The victims had no indication of the breach until their data appeared for sale — the average breach notification lag from third parties is now 117 days, per Black Kite. That's almost four months during which attackers can exfiltrate data, deploy backdoors, and plan their next move.

What should have happened: The SaaS provider needed network segmentation isolating each tenant, real-time anomalous access detection, and contractual breach notification obligations measured in hours, not months. Your business should demand all three before signing a vendor agreement.

The Open-Source Poison in Your Software Stack

Sonatype's 2026 State of the Software Supply Chain Report documented 454,600+ new malicious packages uploaded to npm, PyPI, Maven, NuGet, and Hugging Face in 2025 — a 75% year-over-year jump. These aren't obscure libraries. Attackers use typosquatting (naming packages to resemble legitimate ones) and dependency confusion to inject malware into the software supply chain that your internal tools — and your vendors' tools — pull in automatically.

Group-IB's 2026 threat report identified six organized attack groups now specializing exclusively in supply chain compromise, targeting SaaS platforms, open-source registries, and MSP networks. The industrialization of supply chain attacks means this is no longer opportunistic; it's a structured, profitable business model.

A single poisoned dependency in a vendor's product can expose every customer using that software. When your accounting platform, CRM, or inventory system auto-updates, it may pull compromised code without anyone noticing until customer data is already exfiltrated.

What should have happened: Vendors should maintain a Software Bill of Materials (SBOM) for every product they ship, run automated dependency scanning in CI/CD pipelines, and pin dependencies to verified, hashed versions. Your business should require an up-to-date SBOM as a condition of procurement.

What to Demand From Every Vendor This Week

You don't need a dedicated security team to reduce your third-party risk. Here are five demands every business owner can make — today — of every vendor that touches your data:

  1. Breach notification within 24 hours, in writing. If your vendor's standard contract says "reasonable timeframe," strike it out. Demand 24-hour notification of any security incident affecting your data, with a named contact and escalation path.
  2. SOC 2 Type II or ISO 27001 certification — current, not expired. If a vendor can't produce a current third-party audit report, they haven't been independently verified. Accept nothing less.
  3. An up-to-date Software Bill of Materials (SBOM). Ask: "Show me every software component and dependency in the product we use." If they can't produce one, they don't know what's in their own code.
  4. Multi-factor authentication enforced for all accounts accessing your data. Not offered, not optional — enforced. This includes their internal admin accounts and any API integrations.
  5. A documented incident response plan that includes downstream notification procedures. Ask to see the specific section that covers how they notify customers like you. If it's vague, walk away.

FAQ

Q: My business is small. Do attackers really target companies like mine through vendors?

Yes — and you're a more attractive target than you think. Intel 471's data shows SMBs are disproportionately victimized through third-party compromise precisely because they lack the resources to vet vendors thoroughly. Attackers know that breaching one payroll provider or accounting platform gives them access to hundreds of small businesses at once. You are not too small to be in the blast radius.

Q: What's the single most effective thing I can do to reduce third-party risk?

Inventory every vendor that stores, processes, or transmits your data — then rank them by the sensitivity of that data. Start demanding certifications and breach notification commitments from the top five. Most businesses discover they have 50+ vendors and haven't reviewed security terms for any of them since signing.

Q: Is a SOC 2 report really necessary for every vendor?

Not every vendor needs a full SOC 2, but any vendor holding personally identifiable information, financial data, or health records of your customers should have independent attestation. For lower-risk vendors, a security questionnaire covering access controls, encryption, and incident response is a reasonable minimum.

Q: How do I know if a vendor has already been breached?

Assume they have. IBM's 2025 Cost of a Data Breach Report found the average supply chain breach takes 267 days to identify and contain. Your vendor may be compromised right now and not know it. This is why contractual notification timelines matter — they're your only guarantee that you'll learn about a breach before your customers do.

Conclusion

The era when you could trust your vendors to handle security is over — not because vendors are malicious, but because they are the path of least resistance for sophisticated attackers. With third-party breach involvement doubling in a single year and average detection windows stretching past eight months, the burden has shifted to you. Demand certifications, demand SBOMs, demand 24-hour notification, and walk away from vendors who won't comply.

Your customers' data depends on choices you make about who you do business with. Make those choices with the same scrutiny you'd apply to locking your own front door.

Is your business protected against supply chain threats? Visit consult.lil.business for a free cybersecurity assessment tailored to Australian businesses.

References

  1. Black Kite Third-Party Breach Report 2026
  2. Intel 471 — Ransomware and Extortion Attacks Surge 63% in 2025
  3. Sonatype 2026 State of the Software Supply Chain Report
  4. IBM Cost of a Data Breach Report 2025 — Supply Chain Findings
  5. Group-IB — Six Supply Chain Attack Groups to Watch in 2026

TL;DR

  • A popular tool for programmers had a bad version that stole secret passwords
  • The bad code was hidden inside a helpful extension called Trivy
  • It specifically targeted AI coding assistants to steal information
  • Any business using this tool needs to change all their passwords right now

What Is Trivy and Why Do Programmers Use It?

Imagine you have a robot helper that checks your homework for mistakes before you turn it in. Trivy is like that robot, but for computer programmers. It looks through their code to find security problems so they can fix them before bad guys find them first [1].

Programmers install Trivy as an extension in their coding software (called VS Code)—kind of like adding a new app to your phone. Extensions are supposed to be safe and helpful. But someone found a way to sneak bad code into version 1.8.12 of the Trivy extension [2].

It's like if someone replaced your homework-checking robot with a fake that secretly copies your work and sends it to strangers.

How Did the Bad Code Steal Secrets?

Here's where it gets tricky. Modern programming often uses AI assistants—like smart helpers that can write code, answer questions, and help programmers work faster [3]. These AI assistants need to see the code you're working on to help you.

The bad code in Trivy was designed to spy on these AI assistants. Every time the AI looked at the code to help, the bad code was also watching—stealing passwords, secret keys, and private information [4].

Think of it like this: imagine you have a study buddy who helps you with homework. But someone sneaks a tape recorder into your study buddy's bag. Now everything you say and write is being secretly recorded and sent to strangers.

Why This Is So Dangerous

This attack is scary for three big reasons:

1. It Came from a Trusted Tool

Programmers trusted Trivy to help them stay safe. They didn't expect it to be the thing attacking them. It's like trusting your locker to keep your backpack safe, but someone secretly cut a hole in the back [5].

2. It Stole Everything Important

The stolen secrets weren't just passwords—they were the keys to everything: cloud accounts, databases, private files. It's like someone stealing not just your house key, but your school locker, your diary, and your phone passcode all at once [6].

3. No One Noticed Right Away

This isn't like someone breaking a window. The bad code was quietly hiding in plain sight, doing its stealing without making any noise. That's why it's called a "supply chain attack"—it attacks the tools you trust, not your computer directly [7].

What Businesses Need to Do Right Now

If your business uses programming tools or has developers who code, here's what needs to happen:

1. Find and Remove the Bad Version

Check if version 1.8.12 of the Trivy extension is installed anywhere. If it is, remove it immediately—like finding a rotten apple and throwing it out before it spoils the whole bunch [8].

2. Change Every Password That Might Have Been Stolen

Every password, every secret key, every login that could have been seen while the bad code was running needs to be changed. It's a lot of work, but it's like changing all the locks on your house after losing your keys [9].

3. Check for Strange Activity

Look at the records of where your computers have been connecting. Are there connections to places you don't recognize? That might be the bad code sending your secrets to strangers [10].

4. Be Careful What You Install

Just because an extension is popular doesn't mean it's always safe. Check who made it, read reviews, and only install from sources you trust. It's like not accepting candy from strangers—you have to be careful about what you let into your computer [11].

The Big Lesson: Even Helpful Tools Can Be Tricky

The scary thing about this attack is that Trivy was supposed to help programmers stay safe. It was a tool designed to find security problems. But someone figured out how to turn it into a weapon [12].

This is happening more and more as we use AI and smart tools in everything we do. Every tool that has access to your information is like a door—and bad guys are always looking for doors they can sneak through [13].

FAQ

Yes, but not version 1.8.12. Other versions are safe. The bad version has been removed from the app store. If you update to a newer version, you're protected [14].

If your company has programmers who use VS Code and they installed the Trivy extension around March 2026, you need to check which version they have. Version 1.8.12 is the dangerous one [15].

Sometimes attackers want to steal secrets to sell them. Other times they want to break into systems later using the stolen passwords. And sometimes they just want to cause trouble. Whatever the reason, they're looking for ways to get into places they shouldn't be [16].

This attack specifically targeted programming tools. But the same idea—tricking people into installing bad software—happens with regular apps and games too. That's why you should only download from official sources and be careful with permissions [17].

References

[1] OffSec Radar, "CVE-2026-28353: CWE-506: Embedded Malicious Code in aquasecurity trivy-vscode-extension," OffSec Threat Radar, March 5, 2026. [Online]. Available: https://radar.offseq.com/threat/cve-2026-28353-cwe-506-embedded-malicious-code-in--facddd89

[2] Ibid.

[3] Flashpoint, "Navigating 2026's Converged Threats: Insights from Flashpoint's Global Threat Intelligence Report," Flashpoint, March 11, 2026. [Online]. Available: https://flashpoint.io/blog/global-threat-intelligence-report-2026/

[4] OffSec Radar, "CVE-2026-28353," 2026.

[5] CISA, "Software Supply Chain Security," Cybersecurity and Infrastructure Security Agency, 2025. [Online]. Available: https://www.cisa.gov/software-supply-chain-security

[6] OffSec Radar, "CVE-2026-28353," 2026.

[7] CISA, "Software Supply Chain Security," 2025.

[8] OffSec Radar, "CVE-2026-28353," 2026.

[9] Ibid.

[10] NIST, "Computer Security Incident Handling Guide (SP 800-61 Rev. 2)," National Institute of Standards and Technology, 2025. [Online]. Available: https://csrc.nist.gov/publications/detail/sp/800-61/rev-2/final

[11] OffSec Radar, "CVE-2026-28353," 2026.

[12] Ibid.

[13] Flashpoint, "Navigating 2026's Converged Threats," 2026.

[14] OffSec Radar, "CVE-2026-28353," 2026.

[15] Ibid.

[16] CISA, "Software Supply Chain Security," 2025.

[17] Flashpoint, "Navigating 2026's Converged Threats," 2026.

Worried about your development tools? Book a free cybersecurity consultation at consult.lil.business—we'll help you build a secure development environment.