CVE-2026-57125

9.8 Critique

Publiée le 14/09/2026 à 17:17 · Modifiée le 15/09/2026 à 16:45 · Analyse NVD reportée · Source : security-advisories@github.com

Description

PraisonAI is a multi-agent teams system. Prior to praisonai 4.6.59 and praisonaiagents 1.6.59, the unauthenticated POST /api/v1/runs Jobs API accepts attacker-controlled agent_yaml, and the approve field can mark execute_command as YAML-approved before @require_approval checks critical tools. This chain allows a remote caller to cause a configured language model agent to invoke arbitrary operating-system commands without credentials or operator interaction. This vulnerability is fixed in praisonai 4.6.59 and praisonaiagents 1.6.59 as fixed versions.

Traduire en français

Produits concernés

MervinPraison PraisonAI
MervinPraison praisonaiagents

Scores CVSS

Version	Score	Source	Vecteur
3.1	9.8 Critique	security-advisories@github.com	`CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H`

Faiblesses (CWE)

CWE-306
CWE-863

Références

https://github.com/MervinPraison/PraisonAI/commit/2adfe7e8323f6deec66925cf15a885b6238895e9
https://github.com/MervinPraison/PraisonAI/releases/tag/v4.6.59
https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-4869-x4pr-q22x
https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-4869-x4pr-q22x

Ailleurs

NVD
CVE.org
CIRCL Vulnerability-Lookup
CERT-FR
GitHub Advisories