AIKIDO-2026-302437
ash_ai is vulnerable to Remote Code Execution (RCE)
89
High Risk
This Affects:
ash_aiTL;DR
AshAi.Actions.Prompt compiles prompt content as an EEx template and evaluates it with EEx.eval_string. When a prompt action builds its messages from action arguments or other request data, that text is executed as Elixir source on the server. Expressions embedded in the input run in the BEAM process before any model request is made, allowing arbitrary code execution. The fix stops treating runtime prompt content as an EEx template so input is no longer evaluated as code.
Who does this affect?
You are affected if you are using a version that falls within the vulnerable range and you define a prompt action whose prompt content is built from user input.
Background info
ash_ai is vulnerable to Remote Code Execution (RCE) in versions 0.1.0 - 0.8.2.
How to fix this
Upgrade the ash_ai library to the patch version.
Links
GitHub Release
Are You Affected?
Connect your repositories to instantly see whether vulnerable or malicious packages exist in your codebase.
Free. No credit card required.
I consent to receiving marketing communications based on Aikido’s Privacy Policy.
SOC 2Compliant
ISO 27001CompliantISO 42001Compliant
FedRAMPImplementing© All Intel data is openly available and commercially licensed.