CVE-2026-78379— Consent bypass in python_repl tool via batch kwargs forwarding in Amazon Strands Agents Tools
Quick assessment
Affected
Amazon strands-agents-tools
Exploitation
No confirmed in-the-wild exploitation; assess based on exposure
Recommended action
Check the vendor advisory and references for a fixed version. If immediate upgrade is impossible, restrict exposure and increase monitoring.
在 Amazon Strands Agents Tools 版本 0.8.5 之前，python_repl 工具中存在对用于大语言模型（LLM）提示的输入项处理不当的问题。攻击者可通过构造恶意提示，利用 batch 工具将 non_interactive_mode 作为关键字参数进行传递，从而绕过人工确认机制，进而实现在智能体宿主机上执行任意 Python 代码。为修复此问题，建议用户将软件升级至版本 0.8.5 或更高版本。
Possible ATT&CK Techniques 3 AI
T1055 · Process Injection T1190 · Exploit Public-Facing Application T1059.004 · Unix ShellAffected Version Matrix 1
| Vendor | Product | Version Range | Status |
|---|---|---|---|
| Amazon | strands-agents-tools | < 0.8.5 |
affected |
Get alerts for future matching vulnerabilities
Log in to subscribe
I. Basic Information for CVE-2026-78379
Vulnerability Information
Have questions about the vulnerability? See if Shenlong's analysis helps!
View Shenlong Deep Dive ↗
Although we use advanced large model technology, its output may still contain inaccurate or outdated information.Shenlong tries to ensure data accuracy, but please verify and judge based on the actual situation.
Vulnerability Title
Consent bypass in python_repl tool via batch kwargs forwarding in Amazon Strands Agents Tools
Source: CVE Program / CVE List V5
Vulnerability 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.
Source: CVE Program / CVE List V5
CVSS Information
CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:H
Source: CVE Program / CVE List V5
Vulnerability Type
CWE-1427
Source: CVE Program / CVE List V5
Affected Products
| Vendor | Product | Affected Versions | CPE | Subscribe |
|---|---|---|---|---|
| Amazon | strands-agents-tools | 0 ~ 0.8.5 | - |
|
II. Public POCs for CVE-2026-78379
| # | POC Description | Source Link | Shenlong Link |
|---|
AI-Generated POC
Premium
No public POC found.
Login to generate AI POCIII. Intelligence Information for CVE-2026-78379
请登录查看更多情报信息。
Vendor Advisories for CVE-2026-78379 (1)
Other References for CVE-2026-78379 (1)
IV. Related Vulnerabilities
Same vendor: Amazon
-
CVE-2026-18657
Executable Resolution from Untrusted Project Directory in Ki -
CVE-2026-18656
Executable Resolution from Untrusted Project Directory in Ki -
CVE-2026-16318
QUIC Transport Parameters Memory Leak During HelloRetryReque -
CVE-2026-15746
Credential disclosure in Strands Agents Tools elasticsearch_ -
CVE-2026-15738
Cross-namespace traffic interception via incorrect route pre
Same weakness: CWE-1427
-
CVE-2026-75130
Context7 2.1.2 Prompt Injection via Custom AI Instructions -
CVE-2026-21832
HCL AION is affected by multiple security vulnerabilities. -
CVE-2026-18733
Prompt injection bypasses shell tool consent gate in Strands -
CVE-2026-15077
Improper Neutralization of Input Used for LLM Prompting in G -
CVE-2026-44688
Eclipse theia 软件供应链问题漏洞
V. Comments for CVE-2026-78379
No comments yet