CVE-2026-18733 - Prompt injection bypasses shell tool consent gate in Strands Agents Tools
Bulletin ID: 2026-072-AWS Scope: AWS Content Type: Important (requires attention) Publication Date: 08/03/2026 13:30 PM PDT Description: Strands Agents is an open-source SDK for building AI agents. The strands-agents-tools package provides pre-built tools for use with the SDK, including the shell tool for executing operating system commands on the agent's host. We identified CVE-2026-18733. The shell tool includes a human consent gate that prompts the operator to approve commands before they run. The tool also exposed a non_interactive parameter in the input schema that the large language model (LLM) could control. A crafted prompt, for example one delivered through untrusted content the agent reads (indirect prompt injection), could set non_interactive to true, which bypasses the consent gate and allows arbitrary operating system commands to execute on the agent's host without operator approval. Impacted versions: < 0.8.0 Please refer to the article below for the most up-to-date and complete information related to this AWS Security Bulletin.
AI Analysis
Technical Summary
The strands-agents-tools package provides a shell tool that executes OS commands with a human consent gate. The vulnerability arises because the non_interactive parameter, which disables the consent gate, can be manipulated by the LLM via crafted prompts (prompt injection). This allows an attacker to bypass the consent mechanism and execute arbitrary commands on the host system without operator approval. The vulnerability affects all versions before 0.8.0 and is resolved by upgrading to version 0.8.0.
Potential Impact
An attacker can bypass the human consent gate designed to approve shell commands before execution, leading to arbitrary command execution on the host system. This could result in unauthorized system control or compromise. The vulnerability is exploitable via indirect prompt injection through untrusted content the agent processes.
Mitigation Recommendations
A fix is available in strands-agents-tools version 0.8.0. Users should upgrade to this version or later. Until upgrading, do not expose the shell tool to agents processing untrusted content, and run agents using the shell tool in isolated, least-privilege environments to contain potential command execution.
CVE-2026-18733 - Prompt injection bypasses shell tool consent gate in Strands Agents Tools
Description
Bulletin ID: 2026-072-AWS Scope: AWS Content Type: Important (requires attention) Publication Date: 08/03/2026 13:30 PM PDT Description: Strands Agents is an open-source SDK for building AI agents. The strands-agents-tools package provides pre-built tools for use with the SDK, including the shell tool for executing operating system commands on the agent's host. We identified CVE-2026-18733. The shell tool includes a human consent gate that prompts the operator to approve commands before they run. The tool also exposed a non_interactive parameter in the input schema that the large language model (LLM) could control. A crafted prompt, for example one delivered through untrusted content the agent reads (indirect prompt injection), could set non_interactive to true, which bypasses the consent gate and allows arbitrary operating system commands to execute on the agent's host without operator approval. Impacted versions: < 0.8.0 Please refer to the article below for the most up-to-date and complete information related to this AWS Security Bulletin.
Affected software
Run on your own infrastructure? Check whether these packages are installed with threat-finder — our free open-source scanner.
AI-Powered Analysis
Machine-generated threat intelligence
Technical Analysis
The strands-agents-tools package provides a shell tool that executes OS commands with a human consent gate. The vulnerability arises because the non_interactive parameter, which disables the consent gate, can be manipulated by the LLM via crafted prompts (prompt injection). This allows an attacker to bypass the consent mechanism and execute arbitrary commands on the host system without operator approval. The vulnerability affects all versions before 0.8.0 and is resolved by upgrading to version 0.8.0.
Potential Impact
An attacker can bypass the human consent gate designed to approve shell commands before execution, leading to arbitrary command execution on the host system. This could result in unauthorized system control or compromise. The vulnerability is exploitable via indirect prompt injection through untrusted content the agent processes.
Mitigation Recommendations
A fix is available in strands-agents-tools version 0.8.0. Users should upgrade to this version or later. Until upgrading, do not expose the shell tool to agents processing untrusted content, and run agents using the shell tool in isolated, least-privilege environments to contain potential command execution.
Technical Details
- Article Source
- {"url":"https://aws.amazon.com/security/security-bulletins/rss/2026-072-aws/","fetched":true,"fetchedAt":"2026-08-03T20:45:32.702Z","wordCount":235}
- Classification
- {"confidence":0.88,"severitySource":"default","classifier":"rss-v2"}
Threat ID: 6a70fdecbf32cb7a342cfb8e
Added to database: 08/03/2026, 20:45:32 UTC
Last enriched: 08/17/2026, 22:31:03 UTC
Last updated: 09/17/2026, 22:01:33 UTC
Views: 132
Community Reviews
0 reviewsCrowdsource mitigation strategies, share intel context, and vote on the most helpful responses. Sign in to add your voice and help keep defenders ahead.
Want to contribute mitigation steps or threat intel context? Sign in or create an account to join the community discussion.
Actions
Updates to AI analysis require Pro Console access. Upgrade inside Console → Billing.
External Links
Need more coverage?
Upgrade to Pro Console for AI refresh and higher limits.
For incident response and remediation, OffSeq services can help resolve threats faster.
Latest Threats
Check if your credentials are on the dark web
Instant breach scanning across billions of leaked records. Free tier available.