CVE-2025-59532
Published: 22 September 2025
Summary
CVE-2025-59532 is a high-severity Improper Input Validation (CWE-20) vulnerability. Its CVSS base score is 8.6 (High).
Operationally, exploitation aligns with the MITRE ATT&CK technique Exploitation for Client Execution (T1203); ranked at the 28.0th percentile by exploit likelihood (below the median); it is not currently listed in the CISA KEV catalog.
This vulnerability is AI-related — categorised as Enterprise AI Assistants; in the LLM/Generative AI Risks risk domain.
EU & UK References
- 🇪🇺 ENISA EUVD: EUVD-2025-30278
Vulnerability details
Codex CLI is a coding agent from OpenAI that runs locally. In versions 0.2.0 to 0.38.0, due to a bug in the sandbox configuration logic, Codex CLI could treat a model-generated cwd as the sandbox’s writable root, including paths outside…
more
of the folder where the user started their session. This logic bypassed the intended workspace boundary and enables arbitrary file writes and command execution where the Codex process has permissions - this did not impact the network-disabled sandbox restriction. This issue has been patched in Codex CLI 0.39.0 that canonicalizes and validates that the boundary used for sandbox policy is based on where the user started the session, and not the one generated by the model. Users running 0.38.0 or earlier should update immediately via their package manager or by reinstalling the latest Codex CLI to ensure sandbox boundaries are enforced. If using the Codex IDE extension, users should immediately update to 0.4.12 for a fix of the sandbox issue.
- CWE(s)
AI Security AnalysisAI
- AI Category
- Enterprise AI Assistants
- Risk Domain
- LLM/Generative AI Risks
- OWASP Top 10 for LLMs 2025
- None mapped
- Classification Reason
- Matched keywords: openai
Related Threats
MITRE ATT&CK Enterprise TechniquesAI
Why these techniques?
The sandbox bypass vulnerability enables arbitrary file writes and command execution outside the intended workspace boundary, facilitating exploitation of the client application for code execution (T1203) and evasion of sandbox defenses (T1211).
Affected Assets
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.
Security testing and developer training directly verify and enforce proper input validation, reducing exploitability of injection and malformed-data weaknesses.
Security testing and evaluation at multiple SDLC stages directly detects missing or flawed input validation, with the required remediation process ensuring fixes are applied.
Directly implements checks on information inputs to reject invalid data before processing.
Spam protection mechanisms perform filtering and detection on inbound/outbound messages, directly compensating for missing or weak input validation of unsolicited content.