Raw vector
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:HSummary
CVE-2025-5120 is a critical-severity Code Injection (CWE-94) vulnerability in Huggingface Smolagents. Its CVSS base score is 10.0 (Critical).
Operationally, exploitation aligns with the MITRE ATT&CK technique Command and Scripting Interpreter (T1059); ranked in the top 3% of CVEs by exploit likelihood; it is not currently listed in the CISA KEV catalog; a public proof-of-concept is referenced.
This vulnerability is AI-related — categorised as AI Agent Protocols and Integrations; in the LLM/Generative AI Risks risk domain.
The strongest mitigations our analysis identified map to SA-11 (Developer Testing and Evaluation) and SI-10 (Information Input Validation) — see the control section below for these in your framework.
Deeper analysis AI-assisted summary
Synthesised by an AI model from the NVD description and linked references — a reading aid, not an authoritative source.
A sandbox escape vulnerability was identified in huggingface/smolagents version 1.14.0, specifically in the local_python_executor.py module. The component employs static and dynamic checks to restrict Python execution but fails to prevent use of whitelisted modules and functions, enabling attackers to bypass the sandbox and achieve remote code execution. The flaw carries a CVSS score of 10.0 and is tracked under CWE-94; it was fixed in version 1.17.0.
An unauthenticated attacker with network access can supply crafted code that abuses permitted imports and calls to escape the restricted environment, resulting in arbitrary command execution on the host, data exfiltration, or broader compromise of any system integrating the library. The attack requires no user interaction and breaks the core isolation boundary intended for untrusted agent-generated code.
The issue is addressed by the patch at commit 33a942e62b6fbf6a35d41f1c735bda2d64c163d0, with further details available in the associated huntr.com bounty report. The EPSS score has remained flat at 0.0187 since disclosure, indicating limited observed exploitation interest to date. The affected library is an AI agent framework, making the sandbox bypass particularly relevant to deployments that execute model-generated Python in shared or production environments.
OWASP Top 10 for Web (2025)
EU & UK References
- 🇪🇺 ENISA EUVD: EUVD-2025-22815
Vulnerability Data
A sandbox escape vulnerability was identified in huggingface/smolagents version 1.14.0, allowing attackers to bypass the restricted execution environment and achieve remote code execution (RCE). The vulnerability stems from the local_python_executor.py module, which inadequately restricts Python code execution despite employing static…
more
and dynamic checks. Attackers can exploit whitelisted modules and functions to execute arbitrary code, compromising the host system. This flaw undermines the core security boundary intended to isolate untrusted code, posing risks such as unauthorized code execution, data leakage, and potential integration-level compromise. The issue is resolved in version 1.17.0.
- CWE(s)
AI Security AnalysisAI
- AI Category
- AI Agent Protocols and Integrations
- Risk Domain
- LLM/Generative AI Risks
- OWASP Top 10 for LLMs 2025
- None mapped
- Classification Reason
- Matched keywords: huggingface
Related Threats
MITRE ATT&CK Enterprise Techniques
CVEs Like This One
Affected Assets
Mitigating Controls
Control response
—
—
—
V1.3.1
Mitigating Controls (NIST 800-53 r5) AI
Developer testing and evaluation finds code paths that accept and execute externally influenced strings.
Input validation directly stops untrusted data from being used to construct executable code without neutralization.
Least privilege limits the damage an injected code fragment can perform once executed.
Requiring documented secure development standards and tools enforces use of safe code-generation APIs and escaping.
Mitigating Controls (NIST CSF 2.0) AI
Derived directly from the weakness types (CWEs) cited in the NVD entry via our AI-authored CWE→CSF cross-walk (authority under review) — links open the control.
PR.PS-06's SDLC practices directly target injection flaws via secure coding and testing (mostly), yet as a single broad outcome it leaves many code-generation specifics unaddressed (partial).
PR.DS-10 protects runtime data confidentiality/integrity but has no bearing on neutralizing externally influenced input during code generation, so neither direction shows any preventive effect.
Mitigating Controls (ISO/IEC 27001:2022 Annex A) AI
Derived directly from the weakness types (CWEs) cited in the NVD entry via our AI-authored CWE→ISO cross-walk (authority under review) — links open the control.
Banning unapproved code samples and unauthenticated web services, combined with secure-coding standards and SAST, prevents the dynamic generation or inclusion of attacker-supplied code.
Controls that restrict unauthorized or malicious code from being introduced via external networks or removable media limit opportunities for an attacker to inject and execute arbitrary code.