Am I affected by PraisonAI: Remote Code Execution via Sandbox Escape in `codeMode` Tool? What should I do?
The
codeModetool insrc/praisonai-ts/src/tools/builtins/code-mode.tsusesnew Function()with awith(sandbox)pattern to execute LLM-generated code.Affected: praisonai <= 1.7.1. Fixed in: 1.7.2. What should you check, and what should you do about it?
Full incident page with sources: https://agentarmorhq.com/incidents/2026/06/cve-2026-57141-praisonai-npm (opens external site)
1 answer
Am I affected? Affected: praisonai <= 1.7.1. Fixed in: 1.7.2.
- Check your version: run
npm ls praisonai. Versions <= 1.7.1 of praisonai are affected.
What to do now
- Upgrade to the fixed version: run
npm install praisonai@1.7.2(first patched version 1.7.2). - Use the
vmmodule with proper context isolation. - Or use
isolated-vmfor true process-level isolation with separate V8 isolates.
General practice, not from the source: if the affected software could read tokens or credentials, rotate them after upgrading; pin versions and review changes before updating tools your agents run.
Sources
- GitHub Advisory Database: https://github.com/advisories/GHSA-p69m-4f92-2v84 (opens external site)
Full incident page: https://agentarmorhq.com/incidents/2026/06/cve-2026-57141-praisonai-npm (opens external site)
Did this work for you?- Check your version: run