Improper Neutralization of Input Used for LLM Prompting Affecting strands-agents-tools package, versions [,0.8.4)
Threat Intelligence
The probability is the direct output of the EPSS model, and conveys an overall sense of the threat of exploitation in the wild. The percentile measures the EPSS probability relative to all known EPSS scores. Note: This data is updated daily, relying on the latest available EPSS model version. Check out the EPSS documentation for more details.
Do your applications use this vulnerable package?
In a few clicks we can analyze your entire application and see what components are vulnerable in your application, and suggest you quick fixes.
Test your applications- Snyk IDSNYK-PYTHON-STRANDSAGENTSTOOLS-19329790
- published27 Aug 2026
- disclosed25 Aug 2026
- creditUnknown
Introduced: 25 Aug 2026
NewCVE-2026-78379 (opens in a new tab)How to fix?
Upgrade strands-agents-tools to version 0.8.4 or higher.
Overview
strands-agents-tools is an A collection of specialized tools for Strands Agents
Affected versions of this package are vulnerable to Improper Neutralization of Input Used for LLM Prompting via the python_repl function in src/strands_tools/python_repl.py. An attacker can execute arbitrary Python code on the agent’s host by supplying a crafted prompt that forwards non_interactive_mode through the batch tool, bypassing the human consent prompt. The vulnerable code reads non_interactive_mode from kwargs, so attacker-controlled tool arguments can suppress confirmation and let python_repl run code without the user’s approval. This breaks the intended consent gate for interactive code execution and can expose the host to arbitrary command execution.