{"id":"CVE-2026-123675","published":"2026-11-06T16:17:54.737","lastModified":"2026-11-06T18:18:04.420","description":"A flaw has been found in OpenBMB Internet of Agents up to 9d78e88e47160eb42ee1ee8538c05e1c2c47877a. This affects the function execute_code of the file im_client/agents/tools/code_executor.py of the component Code Executor Tool. This manipulation of the argument filename/lang causes code injection. The attack can be initiated remotely. The exploit has been published and may be used. Continious delivery with rolling releases is used by this product. Therefore, no version details of affected nor updated releases are available. The project was informed of the problem early through an issue report but has not responded yet.","cvssScore":8.3,"cvssSeverity":"HIGH","cvssVector":"CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:L/I:L/A:L","cwes":["CWE-74","CWE-94"],"vendors":[],"products":[],"references":[{"url":"https://github.com/OpenBMB/IoA/issues/18","tags":[]},{"url":"https://vuldb.com/cve/CVE-2026-123675","tags":[]},{"url":"https://vuldb.com/submit/1033599","tags":[]},{"url":"https://vuldb.com/vuln/431792","tags":[]},{"url":"https://vuldb.com/vuln/431792/cti","tags":[]}],"exploitRefs":[{"url":"https://github.com/OpenBMB/IoA/issues/18","tags":[]}],"hasPoc":true,"ai":{"summary":"A code injection vulnerability exists in the Code Executor Tool of OpenBMB Internet of Agents, where manipulation of the 'filename' or 'lang' arguments in the execute_code function allows remote attackers to inject and execute arbitrary code. This is a high-severity flaw (CVSS 8.3) due to its network-exploitable nature and lack of authentication or user interaction requirements.","exploitability":"Exploitation is straightforward for remote attackers since it requires no authentication, user interaction, or special conditions—only sending a crafted request to the vulnerable execute_code endpoint.","blast_radius":"Successful exploitation could lead to limited confidentiality, integrity, and availability impacts (C:L/I:L/A:L), potentially allowing attackers to read local files, modify system behavior, or disrupt agent operations within the compromised environment.","remediation":"No fixed version is published; disable the Code Executor Tool feature entirely by commenting out or removing calls to execute_code in im_client/agents/tools/code_executor.py until a patch is released.","detection":"Monitor for unexpected subprocess or shell command execution originating from Python processes running the code_executor module, especially with non-standard or user-supplied 'filename' or 'lang' values; also inspect logs for abnormal file creation or access patterns in temporary directories used by the tool.","tags":["code-injection","rce","web","agent-platform"],"model":"garnet-coder","analyzedAt":"2026-11-10T09:55:25.365Z"}}