Cyber Resilience

CVE-2026-45758

CriticalRCEUpdated

Published: 05 June 2026

Published
05 June 2026
Modified
17 June 2026
KEV Added
Patch
CVSS Score v3.1 9.6 CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H
EPSS Score 0.0028 19.4th percentile
Risk Priority 70 floored blend · peak EPSS

Summary

CVE-2026-45758 is a critical-severity Embedded Malicious Code (CWE-506) vulnerability in Guardrailsai Guardrails Ai. Its CVSS base score is 9.6 (Critical).

Operationally, exploitation aligns with the MITRE ATT&CK technique Compromise Software Supply Chain (T1195.002); ranked at the 19.4th percentile by exploit likelihood (below the median); it is not currently listed in the CISA KEV catalog.

This vulnerability is AI-related — categorised as NLP and Transformers; in the Supply Chain and Deployment risk domain.

OWASP Top 10 for Web (2025)

EU & UK References

No EU or UK CSIRT advisories indexed for this CVE.

Vulnerability details

Guardrails AI is a Python framework that helps build AI applications. On May 11, 2026 at approximately 6:00 PM Pacific, an attacker published a malicious version of `guardrails-ai` (0.10.1) to PyPI. Aany user who installed `guardrails-ai==0.10.1` from PyPI on May…

more

11, 2026 may be affected. Security researchers identified the malicious package within approximately 2 hours of publication, and PyPI quarantined the repository. Based on our telemetry, Guardrails AI maintainers have observed no requests to Guardrails AI infrastructure originating from the malicious 0.10.1 version, and a review of system and access logs has produced no evidence of user data exfiltration through their systems. Users should upgrade to version 0.10.2 or downgrade to version 0.10.0, both of which are unaffected. Those who installed version 0.10.1 should rotate any credentials accessible from their machine (GitHub PATs, cloud provider keys, package registry tokens, API keys) and audit their GitHub account for unauthorized workflows or repositories.

CWE(s)

AI Security AnalysisAI

AI Category
NLP and Transformers
Risk Domain
Supply Chain and Deployment
OWASP Top 10 for LLMs 2025
None mapped
Classification Reason
Matched keywords: ai

Related Threats

MITRE ATT&CK Enterprise TechniquesAI

T1195.002 Compromise Software Supply Chain Initial Access
Adversaries may manipulate application software prior to receipt by a final consumer for the purpose of data or system compromise.
Why these techniques?

Malicious package published to PyPI is a direct software supply chain compromise.

Confidence: HIGH · MITRE ATT&CK Enterprise v19.0

Affected Assets

guardrailsai
guardrails ai
0.10.1

Mitigating Controls

Likely Mitigating Controls AI

Per-CVE control mapping for this CVE has not run yet; the list below is derived from the weakness types (CWEs) cited in the NVD entry.

addresses: CWE-506

Restricting software to licensed versions and controlling P2P prevents introduction of software containing embedded malicious code from unauthorized sources.

addresses: CWE-506

The control prevents users from installing software that contains embedded malicious code.

addresses: CWE-506

Regular inventory reviews and updates make it harder to conceal or exploit embedded malicious code by requiring all components to be documented and accounted for.

addresses: CWE-506

Reverting to a known state removes any malicious code embedded by an attacker.

addresses: CWE-506

The approval and review process for maintenance tools can prevent introduction or continued use of tools containing embedded malicious code.

addresses: CWE-506

Supply chain strategy requires vetting and controls during acquisition to prevent or detect insertion of malicious code by vendors or integrators.

addresses: CWE-506

Background screening for development or deployment roles makes intentional insertion of malicious code by insiders materially harder to accomplish.

addresses: CWE-506

The capability explicitly searches for embedded malicious code and backdoors as indicators of compromise.

References