Severity by source
CVSS:4.0/AV:N/AC:L/AT:N/PR:L/UI:A/VC:H/VI:H/VA:N/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X
Authenticated user supplies input (PR:L) with required interaction (UI:R); command execution stays within the sandbox authority (S:U) yielding high confidentiality and integrity but no availability impact.
Primary rating from Vendor (AMZN).
CVSS VectorVendor: AMZN
Lifecycle Timeline
5DescriptionCVE.org
Improper neutralization of argument delimiters in the install_packages() method in AWS Bedrock AgentCore Python SDK before 1.18.1 might allow a remote authenticated user to execute arbitrary commands within the Code Interpreter sandbox via crafted package name arguments.
To mitigate this issue, users should upgrade to the patched version 1.18.1.
Articles & Coverage 1
AnalysisAI
Argument injection in the AWS Bedrock AgentCore Python SDK before 1.18.1 lets a remote authenticated user run arbitrary commands inside the Code Interpreter sandbox by embedding shell/argument delimiters in package names passed to install_packages(). Because the CVSS vector is PR:L, an attacker needs some level of authenticated access and the flaw requires user interaction (UI:A) with a high confidentiality and integrity impact (CVSS 4.0 base 8.4). There is no public exploit identified at time of analysis and the issue is not listed in CISA KEV.
Technical ContextAI
The affected component is the AWS Bedrock AgentCore Python SDK, the client library used to build and orchestrate agents on Amazon Bedrock AgentCore, which includes a managed Code Interpreter sandbox for executing agent-generated code and installing dependencies. The root cause is CWE-88 (Improper Neutralization of Argument Delimiters in a Command, 'Argument Injection'): the install_packages() method passes user-influenced package name strings into an underlying package-installation command without properly neutralizing argument delimiters (such as spaces, dashes introducing new flags, or shell metacharacters). This allows crafted 'package name' input to be reinterpreted as additional command-line options or separate commands, turning a dependency-install feature into an arbitrary command execution primitive within the sandbox. The CPE identifies the affected package as cpe:2.3:a:aws:bedrock-agentcore_1.18.1, i.e. the bedrock-agentcore Python distribution on PyPI prior to 1.18.1.
RemediationAI
Vendor-released patch: upgrade the bedrock-agentcore Python package to version 1.18.1 or later (for example via 'pip install --upgrade bedrock-agentcore==1.18.1'), as directed by AWS security bulletin 2026-065 (https://aws.amazon.com/security/security-bulletins/2026-065-aws/) and advisory GHSA-j6g5-3hh3-pgw8. This is the primary and authoritative fix. Until the upgrade is deployed, reduce exposure by not passing untrusted or agent/LLM-generated package names directly into install_packages(): validate package names against a strict allowlist or a conservative regex (letters, digits, dots, hyphens, underscores only, rejecting whitespace, leading dashes, and shell metacharacters), and pin dependencies to a vetted manifest rather than accepting free-form package arguments - the trade-off is reduced flexibility for agents that dynamically install packages. Because impact is confined to the Code Interpreter sandbox, also review and tighten the sandbox's outbound network and permission scope so that even successful in-sandbox command execution cannot reach sensitive resources.
Wazuh SIEM platform versions 4.4.0 through 4.9.0 contain an unsafe deserialization vulnerability in the DistributedAPI t
BentoML version 1.4.2 and earlier contains an unauthenticated remote code execution vulnerability through insecure deser
pgAdmin 4 contains critical remote code execution vulnerabilities in the Query Tool download and Cloud Deployment endpoi
The renderLocalView function in render/views.py in graphite-web in Graphite 0.9.5 through 0.9.10 uses the pickle Python
BentoML is a Python library for building online serving systems optimized for AI apps and model inference. Rated critica
OpenSSL before 0.9.8za, 1.0.0 before 1.0.0m, and 1.0.1 before 1.0.1h does not properly restrict processing of ChangeCiph
pyLoad download manager version prior to 0.5.0b3.dev77 exposes the Flask SECRET_KEY through an unauthenticated endpoint.
Langflow (a visual LLM pipeline builder) contains a critical unauthenticated code execution vulnerability (CVE-2026-3301
In Mercurial before 4.1.3, "hg serve --stdio" allows remote authenticated users to launch the Python debugger, and conse
Unauthenticated remote code execution affects Kestra OSS (the open-source event-driven orchestration platform) prior to
Unauthenticated remote code execution in Marimo ≤0.20.4 allows attackers to execute arbitrary system commands via the `/
pyLoad is the free and open-source Download Manager written in pure Python. Rated medium severity (CVSS 5.3), this vulne
Share
External POC / Exploit Code
Leaving vuln.today
EUVD-2026-48422
GHSA-j6g5-3hh3-pgw8