by Kastro
Paper/Archive/CVE-2026-61447

CVE-2026-61447

MervinPraison PraisonAI — PraisonAI before 1.6.78 Remote Code Execution via CodeAgent

EPSS 15%public exploitCVSS 10.0
critical.is priorityof 100
59
CVSS 10.0 (CNA)10.0 / 10 × 3030.0Not in CISA KEV+300.0EPSS percentile 96.8.968 × 2019.4Exploit: SSVC poc+1010.0SSVC exploitation: poc+100.0
Description · from the CVE record

PraisonAI before 1.6.78 contains a remote code execution vulnerability in CodeAgent._execute_python() that executes LLM-generated Python code without AST validation, import restrictions, or sandbox enforcement. Attackers can influence LLM output through prompt injection to exfiltrate all environment secrets and execute arbitrary code on the host system.

Affected

ProductAffectedFixed inSource
PraisonAI< 1.6.781.6.78CNA

Scores by source

VulnCheck (CNA)10.0CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:H/VA:H/SC:H/SI:H/SA:H
VulnCheck (CNA)10.0CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H
CISA ADPSSVCExploitation: poc · Automatable: yes · Technical impact: total

References

exploitgithub.com/MervinPraison/PraisonAI/security/advisories/GHSA-2xv2-w8cq-5gxwwriteupvulncheck.com/advisories/praisonai-before-remote-code-execution-via-codeagent