Spring Vulnerabilities: 91 CVEs Expose Supply Chain Risk

Introduction: Spring Vulnerabilities — Why It Matters

Spring Vulnerabilities have emerged as a major open-source security concern after Broadcom published a large batch of Spring security advisories on August 20, 2026. Sonatype tracked 91 CVEs across Spring Framework and related projects, with 209,569 software components identified as affected at the time of its analysis.

The Spring Vulnerabilities disclosure affects widely used technologies including Spring Security, Spring Cloud Config, Spring AI, Spring Data REST, Spring Integration, Reactor Core, Reactor Netty, Spring AMQP and Spring Batch. The vulnerabilities cover several classes, including insecure deserialization, remote code execution under specific conditions, SSRF, path traversal, information exposure, denial of service and improper authorization.

For organizations, the main challenge is not simply counting CVEs. Vulnerable Spring libraries can be introduced directly or through transitive dependencies, meaning teams may be exposed without deliberately adding the affected component.

What Caused the August Spring Security Disclosure?

The August Spring Vulnerabilities release represents a large coordinated security update across the Spring ecosystem. Sonatype said the event reflects a broader increase in vulnerability discovery, including the growing use of AI-assisted security analysis.

Spring had already reported a sharp increase in security reports earlier in 2026. Sonatype noted that Spring received 482 reports across 65 projects in April, compared with 55 in March. The figures illustrate how vulnerability discovery can generate a remediation workload faster than organizations can absorb it.

For broader cybersecurity reporting and incident coverage, organizations can also follow Cyber Incidents.

Spring Vulnerabilities: Technical Breakdown

CVE-2026-59285: Spring GraphQL RCE Risk

One of the most serious flaws is CVE-2026-59285, an unsafe deserialization vulnerability affecting Spring for GraphQL. Spring’s advisory says exploitation can lead to remote code execution when several conditions are present, including Jackson 2.x JSON deserialization, a paginated GraphQL field and exploitable classes available on the classpath.

Sonatype rates the vulnerability CVSS 9.2 Critical, while Spring’s own advisory categorizes it as High severity. Affected Spring for GraphQL versions include 2.0.0 through 2.0.4, with version 2.0.5 listed as the open-source fix.

CVE-2026-59318: Spring AI Tool-Calling Flaw

CVE-2026-59318 affects Spring AI’s tool-calling functionality. Spring says the per-request list of permitted tools is not fully enforced during dispatch, meaning an attacker could potentially manipulate a model into invoking a tool that was not made available to that request. This can result in privilege escalation under certain conditions.

Affected versions include Spring AI 2.0.0, 1.1.0–1.1.8 and 1.0.0–1.0.9. Spring lists 2.0.1 as the open-source fixed release.

Potential Risks and Impact

Remote Code Execution

The GraphQL deserialization flaw can potentially allow remote code execution when its specific prerequisites are satisfied. Internet-facing GraphQL applications should therefore receive priority during vulnerability assessment.

AI Privilege Escalation

The Spring AI issue demonstrates a different risk: prompt injection can become more serious when application-level authorization does not independently enforce which tools an AI system can call.

Software Supply Chain Exposure

Sonatype’s 209,569 affected-component figure includes software that may bundle vulnerable Spring code rather than only applications directly using Spring. This makes software composition analysis and SBOM visibility essential for accurate exposure assessment.

Official Response and Recommended Action

Spring has published individual security advisories and fixed versions for affected projects. The official advisory for CVE-2026-59285 recommends upgrading to a fixed release, while CVE-2026-59318 also requires affected Spring AI deployments to move to supported fixed versions.

Security teams should:

1. Inventory Spring Framework and related dependencies across production and development environments.
2. Review SBOMs and software composition analysis results.
3. Check both direct and transitive Spring dependencies.
4. Prioritize internet-facing GraphQL applications and systems processing untrusted input.
5. Review Spring AI applications that use tool calling or model-driven actions.
6. Upgrade affected components to the appropriate fixed versions and test deployments.
7. Rebuild and redeploy applications after dependency updates rather than assuming the source repository is patched.
8. Monitor vendor software that embeds Spring components and request updated builds where necessary.

For practical security guidance, organizations can use CyberNexora’s Learn & Protect resources.

Industry Context: Why the Disclosure Matters

The August Spring Vulnerabilities disclosure highlights a growing tension in open-source security: vulnerability discovery is becoming faster, while remediation still depends on inventories, testing, maintainers and deployment cycles.

Sonatype said the Spring team had processed findings, produced fixes and published patched versions, but downstream organizations still need to determine where vulnerable code exists and safely deploy those fixes.

The event also reinforces the importance of software supply-chain visibility. A vulnerability can remain present in an organization’s environment even when developers never intentionally selected the vulnerable library.

Key Takeaways

• Broadcom published a large Spring security advisory batch on August 20, 2026.
• Sonatype tracked 91 CVEs and 209,569 affected software components.
• 91 Critical Spring CVEs expose serious security risks across vulnerable Spring components.
• CVE-2026-59318 can potentially enable unauthorized Spring AI tool invocation and privilege escalation.
• Organizations should prioritize dependency inventory, SBOM review and fixed-version upgrades.

Conclusion: Spring Vulnerabilities and What Happens Next

Spring Vulnerabilities demonstrate how a single upstream security disclosure can create a much larger downstream software supply-chain problem. The immediate priority is identifying affected versions, determining actual exposure and deploying appropriate fixes.

Security teams should continue monitoring Spring Vulnerabilities advisories and vendor updates, particularly where applications use GraphQL, AI tool calling, untrusted input or deeply nested third-party dependencies. Additional vulnerability disclosures may continue to increase the pressure on organizations to combine automated discovery with accurate asset and dependency inventories.

For continuing cybersecurity updates and security resources, readers can explore CyberNexora Resources.