Report · Check Point ResearchAbout CVE-2025-61260Published 1 Dec 2025

CVE-2025-61260 — OpenAI Codex CLI: Command Injection via Project-Local Configuration

Source Check Point Research

Compare

Authoritative record:CVE-2025-61260· Critical 9.8

Not known to be exploited · CVSS 9.8 Critical.

Data as of 5 Nov 2026, 00:01 UTC

CVSS 3.1 base score

9.8Critical

How severe the flaw is on the 0–10 CVSS scale.

CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H

Attack Vector: Network
Attack Complexity: Low
Privileges Required: None
User Interaction: None
Scope: Unchanged
Confidentiality: High
Integrity: High
Availability: High

Bold values are the worst case: easiest to exploit or largest impact.

Attack Vector: Network: The vulnerable component is bound to the network stack and the set of possible attackers extends beyond the other options listed, up to and including the entire Internet. Such a vulnerability is often termed 'remotely exploitable' and can be thought of as an attack being exploitable at the protocol level one or more network hops away (e.g., across one or more routers).
Attack Complexity: Low: Specialized access conditions or extenuating circumstances do not exist. An attacker can expect repeatable success against the vulnerable component.
Privileges Required: None: The attacker is unauthorized prior to attack, and therefore does not require any access to settings or files to carry out an attack.
User Interaction: None: The vulnerable system can be exploited without any interaction from any user.
Scope: Unchanged: An exploited vulnerability can only affect resources managed by the same security authority. In this case, the vulnerable component and the impacted component are either the same, or both are managed by the same security authority.
Confidentiality: High: There is total loss of confidentiality, resulting in all resources within the impacted component being divulged to the attacker. Alternatively, access to only some restricted information is obtained, but the disclosed information presents a direct, serious impact.
Integrity: High: There is a total loss of integrity, or a complete loss of protection. For example, the attacker is able to modify any/all files protected by the impacted component. Alternatively, only some files can be modified, but malicious modification would present a direct, serious consequence to the impacted component.
Availability: High: There is total loss of availability, resulting in the attacker being able to fully deny access to resources in the impacted component; this loss is either sustained (while the attacker continues to deliver the attack) or persistent (the condition persists even after the attack has completed). Alternatively, the attacker has the ability to deny some availability, but the loss of availability presents a direct, serious consequence to the impacted component (e.g., the attacker cannot disrupt existing connections, but can prevent new connections; the attacker can repeatedly exploit a vulnerability that, in each instance of a successful attack, leaks a only small amount of memory, but after repeated exploitation causes a service to become completely unavailable).

Definitions: FIRST CVSS v3.1 specification.

Exploitation evidence

Disclosed: evidence found.
CVE published 14 Apr 2026
Detectable: no evidence on record.
No public nuclei template indexed
Weaponized: no evidence on record.
No public exploit code indexed in Metasploit or Exploit-DB
Exploited in the wildExploited: no evidence on record.
Not in CISA KEV and no source reports exploitation
Observed by OffSeq MirageSeen by Mirage: no evidence on record.
No attack traffic recorded by OffSeq Mirage sensors

A missing step is not evidence of safety. Sources: cve.org (opens in a new tab), the Metasploit, Exploit-DB and nuclei-templates index as of 5 Nov 2026.

Disclosed: The CVE record is public on cve.org. Its publication is t = 0.
Detectable: A public nuclei template can find exposed instances, for defenders and attackers alike.
Weaponized: A Metasploit exploit module or Exploit-DB entry exists (Metasploit scanner modules do not count), or CISA SSVC or the CVSS exploit maturity reports a public proof of concept.
Exploited in the wild: CISA KEV lists it, CISA SSVC rates exploitation active, or a source reports attacks.
Observed by OffSeq Mirage: OffSeq Mirage honeypot sensors recorded attack traffic for this CVE, or found exposed hosts without attack traffic.

Moments are spaced evenly, not to scale; a dashed line marks a gap of six months or more. Deltas count from t = 0. Sources that give only a date (CISA KEV, Exploit-DB, or a time of exactly 00:00Z) read “same day” rather than hours. Metasploit modules carry the flaw’s disclosure date, not their release date, and nuclei templates carry no date, so both appear above but not on the timeline.

local

What to do

Affected software

Product	Affected	Fixed in
npmopenai/codex-cli`pkg:npm/openai/codex-cli`Other threats	Affected <=0.23.0	Fixed in–

Check your own hosts for these packages with threat-finder (opens in a new tab), our free open-source scanner.

AI-suggested

Immediately update all OpenAI Codex CLI installations to version 0.23.0 or later, which includes the fix preventing project-local CODEX_HOME redirection and automatic execution of MCP server commands from untrusted locations.
Audit all repositories for the presence of .env files setting CODEX_HOME and ./.codex/config.toml files containing mcp_servers entries; remove or sanitize any suspicious or unnecessary configurations.
Implement strict repository access controls and code review policies to prevent unauthorized commits or merges that could introduce malicious project-local configurations.
Educate developers about the risks of running Codex CLI on untrusted or newly cloned repositories without verification.
Integrate static analysis or automated scanning tools into CI pipelines to detect and block commits containing potentially dangerous MCP configurations.
Monitor developer machines and CI environments for unusual Codex CLI activity or unexpected command executions.
Consider isolating Codex CLI execution environments using containerization or sandboxing to limit the impact of potential command injection.
Maintain an inventory of all development tools and their versions to ensure timely patching of vulnerabilities.
Collaborate with supply-chain security teams to vet third-party and open-source repositories for malicious configurations before adoption.

Description

By: Isabel Mill & Oded Vanunu OpenAI Codex CLI is OpenAI’s command-line tool that brings AI model-backed reasoning into developer workflows. It can read, edit, and run code directly from the terminal, making it possible to interact with projects using natural language commands, automate tasks, and streamline day-to-day development One of its key features is […] The post CVE-2025-61260 — OpenAI Codex CLI: Command Injection via Project-Local Configuration appeared first on Check Point Research .

Impact

For European organizations, this vulnerability poses significant risks especially for software development teams using OpenAI Codex CLI in collaborative environments. Attackers with repository write access can embed malicious commands that execute on developer machines without detection, leading to credential theft (including cloud tokens and SSH keys), persistent remote access, and lateral movement within corporate networks. Supply-chain attacks can propagate through compromised open-source projects or internal templates, affecting multiple organizations downstream. CI/CD pipelines that run Codex CLI on checked-out code are also at risk, potentially contaminating build artifacts and production deployments. The silent and automatic nature of the execution increases the likelihood of unnoticed compromise, threatening confidentiality, integrity, and availability of development environments and associated infrastructure. This can result in intellectual property theft, disruption of software delivery, and exposure of sensitive data. The vulnerability’s exploitation requires no user interaction beyond normal developer workflows, amplifying its impact. European organizations with extensive software development operations, especially those in technology, finance, and critical infrastructure sectors, face elevated risks.

Summary

OpenAI Codex CLI is a command-line tool that integrates AI-assisted code reasoning into developer workflows, allowing natural language interaction with code projects. A critical feature, the Model Context Protocol (MCP), enables integration of external tools via project-local configuration files. The vulnerability arises because Codex CLI automatically loads and executes MCP server commands defined in project-local configuration files (.env and ./.codex/config.toml) without any interactive approval, validation, or re-checks upon changes. Specifically, if a repository contains a .env file setting CODEX_HOME to a local directory and a corresponding config.toml with mcp_servers entries, Codex CLI treats these as trusted execution material and runs the declared commands at startup. An attacker who can commit or merge such files can inject arbitrary commands that execute silently on any developer’s machine when Codex runs, including reverse shells or payloads that exfiltrate credentials. This creates a stealthy supply-chain backdoor that persists and can be swapped post-merge without triggering alerts. The vulnerability breaks the expected security boundary by implicitly trusting project-supplied files, enabling attackers to compromise developer environments, CI pipelines, and downstream build artifacts. Check Point Research disclosed the issue to OpenAI on August 7, 2025, and a fix was released on August 20, 2025, in Codex CLI version 0.23.0. The patch prevents .env files from redirecting CODEX_HOME to project directories, stopping automatic execution of attacker-controlled commands. Users are strongly advised to update to this version or later to mitigate the risk.

References

research.checkpoint.com/2025/openai-codex-cli-command-injection-vulnerability/ (opens in a new tab)

Community reviews

Loading reviews…