CVE-2026-59415

Resource Injection
Affects
Spring AI
in
Spring
No items found.
Versions
>=1.0.0 <=1.0.10, >=1.1.0 <=1.1.9, >=2.0.0 <=2.0.2

Patch Available.

Exclamation circle icon
Patch Available

This Vulnerability has been fixed in the Never-Ending Support (NES) version offered by HeroDevs.

Overview

Spring AI is an application framework for AI engineering that provides Spring-friendly APIs and abstractions for developing AI applications, including a portable vector store API with implementations for many vector databases such as Typesense.

A resource injection vulnerability (CVE-2026-59415) has been identified in the Spring AI Typesense vector store, which allows attackers who control a document ID passed to a delete operation to delete documents other than the intended ones, or every document in the collection.

Per OWASP: This attack consists of changing resource identifiers used by an application in order to perform a malicious task. When an application defines a resource type or location based on user input, such as a file name or port number, this data can be manipulated to execute or access different resources.

This issue affects multiple versions of Spring AI.

Details

Module Info

Vulnerability Info

This High-severity vulnerability is found in the org.springframework.ai:spring-ai-typesense-store package in multiple versions of Spring AI.

TypesenseVectorStore.doDelete(List<String>), which backs VectorStore.delete(List<String>), builds the Typesense filter_by expression for the delete request by concatenating the caller-supplied document IDs into the form id: [id1,id2,...]:

@Override
public void doDelete(List<String> idList) {
	DeleteDocumentsParameters deleteDocumentsParameters = new DeleteDocumentsParameters();
	deleteDocumentsParameters.filterBy(DOC_ID_FIELD_NAME + ": [" + String.join(",", idList) + "]");
	...
}

The Typesense id field takes bare, unquoted values in this list form, and the IDs are neither validated nor escaped. An ID that contains characters with meaning in the Typesense filter grammar, such as a comma (,), a closing bracket (]) or operator tokens (||, &&), breaks out of the list literal and changes the structure of the filter. A comma injects additional IDs, so documents other than the intended ones are deleted, while an operator breakout can produce a filter that matches every document in the collection, resulting in unintended bulk deletion. Applications that pass externally supplied or user-controlled document IDs to VectorStore.delete(List<String>) backed by TypesenseVectorStore are affected.

Mitigation

Only recent versions of Spring AI receive community support. Older lines are End-of-Life and will not receive public updates to address this issue.

Users of the affected components should apply one of the following mitigations:

  • Upgrade to a currently supported version of Spring AI.
  • Leverage a commercial support partner like HeroDevs for post-EOL security support.

Credits

  • Ishaq Mohammed (finder)
  • Daehyun Kang (finder)
  • LiZhiyao from BUPT (finder)
Vulnerability Details
Severity
Level
CVSS Assessment
Low
>=0 <4
Medium
>=4 <6
High
>=6 <8
Critical
>=8 <10
High
ID
CVE-2026-59415
PROJECT Affected
Spring AI
Versions Affected
>=1.0.0 <=1.0.10, >=1.1.0 <=1.1.9, >=2.0.0 <=2.0.2
NES Versions Affected
Published date
November 19, 2026
≈ Fix date
Category
Resource Injection
Vex Document
Download VEXHow do I use it?
Sign up for the latest vulnerability alerts fixed in
NES for Spring
Rss feed icon
Subscribe via RSS
or

By submitting the form I acknowledge receipt of our Privacy Policy.

Thanks for signing up for our Newsletter! We look forward to connecting with you.
Oops! Something went wrong while submitting the form.