| Titre | deepwisdom MetaGPT v0.8.1 Remote command execution |
|---|
| Description | A Remote Code Execution (RCE) vulnerability exists in the MetaGPT framework's DataInterpreter component due to insufficient input validation on user-provided prompts. An attacker can use prompt injection to manipulate the LLM into generating malicious Python code, which is then automatically executed in a Jupyter Notebook environment without any sandbox restrictions or security checks.
|
|---|
| La source | ⚠️ https://github.com/Ka7arotto/cve/blob/main/MetaGPT-rce2.md |
|---|
| Utilisateur | Goku (UID 80486) |
|---|
| Soumission | 06/03/2026 12:58 (il y a 7 mois) |
|---|
| Modérer | 20/03/2026 15:40 (14 days later) |
|---|
| Statut | Accepté |
|---|
| Entrée VulDB | 352081 [Foundation Agents MetaGPT jusqu’à 0.8.1 DataInterpreter write_analysis_code.py élévation de privilèges] |
|---|
| Points | 19 |
|---|