by Kastro
Paper/Archive/CVE-2026-40088

CVE-2026-40088

MervinPraison PraisonAI — Improper Neutralization of Special Elements used in an OS Command ('OS Command Injection') in praisonai

EPSS 0.4%public exploitCVSS 9.7
critical.is priorityof 100
46
CVSS 9.7 (CNA)9.7 / 10 × 3029.1Not in CISA KEV+300.0EPSS percentile 36.1.361 × 207.2Exploit: SSVC poc+1010.0SSVC exploitation: poc+100.0
Description · from the CVE record

PraisonAI is a multi-agent teams system. Prior to 4.5.121, the execute_command function and workflow shell execution are exposed to user-controlled input via agent workflows, YAML definitions, and LLM-generated tool calls, allowing attackers to inject arbitrary shell commands through shell metacharacters. This vulnerability is fixed in 4.5.121.

Affected

ProductAffectedFixed inSource
PraisonAI< 4.5.121–CNA

Scores by source

GitHub_M (CNA)9.7CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H
NVD9.6CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H
CISA ADPSSVCExploitation: poc · Automatable: no · Technical impact: total

References

refgithub.com/MervinPraison/PraisonAI/security/advisories/GHSA-2763-cj5r-c79mrefgithub.com/MervinPraison/PraisonAI/releases/tag/v4.5.121