by Kastro
Paper/Archive/CVE-2026-78379

CVE-2026-78379

Amazon strands-agents-tools — Consent bypass in python_repl tool via batch kwargs forwarding in Amazon Strands Agents Tools

EPSS 0.6%CVSS 9.2
critical.is priorityof 100
37
CVSS 9.2 (CNA)9.2 / 10 × 3027.6Not in CISA KEV+300.0EPSS percentile 47.5.475 × 209.5No public exploit+100.0SSVC exploitation: none+100.0
Description · from the CVE record

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.

Affected

ProductAffectedFixed inSource
strands-agents-tools< 0.8.50.8.5CNA

Scores by source

AMZN (CNA)9.2CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N
AMZN (CNA)8.1CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:H
CISA ADPSSVCExploitation: none · Automatable: no · Technical impact: total

References

patchpypi.org/project/strands-agents-tools/0.8.5advisoryaws.amazon.com/security/security-bulletins/2026-089-aws