CVE-2026-78379
Received Received - Intake

Code Injection in Amazon Strands Agents Tools

Publication date: 2026-08-25

Last updated on: 2026-08-25

Assigner: AMZN

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 Scores

EPSS Scores

Probability:
Percentile:

Meta Information

Published
2026-08-25
Last Modified
2026-08-25
Generated
2026-08-25
AI Q&A
2026-08-25
EPSS Evaluated
N/A
NVD
CVE-2026-78379
EUVD
EUVD-2026-65735

Affected Vendors & Products

Showing 2 associated CPEs
Vendor Product Version / Range
amazon strands_agents_tools to 0.8.5 (exc)
amazon strands_agents_tools 0.8.5

Helpful Resources

Exploitability

CWE
CWE Icon
KEV
KEV Icon
CWE ID Description
CWE-UNKNOWN

Attack-Flow Graph

AI Quick Actions

Instant insights powered by AI
Executive Summary

This is a consent bypass vulnerability in Amazon Strands Agents Tools. The python_repl tool is meant to execute Python code only after getting operator approval. However, a flaw allows remote actors to bypass this consent and run arbitrary Python code on the host system by sending a crafted prompt that sets non_interactive_mode through the batch tool.

Detection Guidance

To detect this vulnerability, check the installed version of strands-agents-tools using pip show strands-agents-tools. If the version is below 0.8.5, the system is vulnerable. Additionally, review agent configurations for the presence of the python_repl or batch tools and ensure they are not exposed to untrusted content.

Impact Analysis

An attacker could execute malicious Python code on your system without your consent. This could lead to unauthorized access, data theft, system compromise, or further attacks within your network. The impact depends on the permissions of the affected application.

Compliance Impact

This vulnerability could lead to unauthorized data access or exfiltration, violating GDPR's data protection principles and HIPAA's security requirements. Organizations may face compliance violations, fines, or legal consequences if exploited.

Mitigation Strategies

Upgrade to strands-agents-tools version 0.8.5 or later to resolve the vulnerability. If an immediate upgrade is not possible, remove either the batch or python_repl tool from the agent's configuration, avoid making python_repl available to agents processing untrusted content, and run such agents in isolated, least-privilege environments.

Chat Assistant

Ask questions about this CVE
Hi! I’m here to help you understand CVE-2026-78379. Ask me anything about the vulnerability, its impact, or mitigation strategies.
0/70

EPSS Chart