CVE-2026-2275

The CrewAI CodeInterpreter tool falls back to SandboxPython when it cannot reach Docker, which can enable RCE through arbitrary C function calling...

CVE-2026-2275 CVE-2026-2275

The CrewAI CodeInterpreter tool falls back to SandboxPython when it cannot reach Docker, which can enable RCE through arbitrary C function calling...

CVE-2026-2275 CVE-2026-2275

The CrewAI CodeInterpreter tool falls back to SandboxPython when it cannot reach Docker, which can enable RCE through arbitrary C function calling...

CVE-2026-2275: Exposed Dangerous Method or Function

The CrewAI CodeInterpreter tool falls back to SandboxPython when it cannot reach Docker, which can enable RCE through arbitrary C function calling...

CVE-2024-10950 Code Injection in binary-husky/gpt_academic

In binary-husky/gptacademic version = 3.83, the plugin CodeInterpreter is vulnerable to code injection caused by prompt injection. The root cause is the execution of user-provided prompts that generate untrusted code without a sandbox, allowing the execution of parts of the LLM-generated code. Th...

CVE-2024-10950 Code Injection in binary-husky/gpt_academic

In binary-husky/gptacademic version = 3.83, the plugin CodeInterpreter is vulnerable to code injection caused by prompt injection. The root cause is the execution of user-provided prompts that generate untrusted code without a sandbox, allowing the execution of parts of the LLM-generated code. Th...

CVE-2024-10950

CVE-2024-10950 affects binary-husky/gpt_academic ≤ 3.83, via the CodeInterpreter plugin. The root cause is prompt injection that causes untrusted prompts to generate code executed without a sandbox, enabling remote code execution (RCE) on the application backend server. The described impact is fu...

CVE-2024-10950: Improper Control of Generation of Code

In binary-husky/gptacademic version = 3.83, the plugin CodeInterpreter is vulnerable to code injection caused by prompt injection. The root cause is the execution of user-provided prompts that generate untrusted code without a sandbox, allowing the execution of parts of the LLM-generated code. Th...