Description
Improper neutralization of input used for LLM prompting in the python_repl tool in Amazon Strands Agents Tools before 0.8.5 might allow remote actors to execute arbitrary Python code on the agent-s host by bypassing the human consent gate, via a crafted prompt that forwards non_interactive_mode as a keyword argument through the batch tool. To remediate this issue, users should upgrade to version 0.8.5 or later.
CVSS 8.1EPSS 0.32%Risque 0.83
Voir la source- Publication
- 2026-08-25 19:16:54
- Versions concernées
- <0.8.5
- Type
- Bibliothèque
- Dernière modification
- 2026-08-26 16:45:12
- Vecteur
- CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:H