CVE-2026-40158
Full vulnerability record from the local CVENow database.
Description
PraisonAI is a multi-agent teams system. Prior to 4.5.128, PraisonAI's AST-based Python sandbox can be bypassed using type.__getattribute__ trampoline, allowing arbitrary code execution when running untrusted agent code. The _execute_code_direct function in praisonaiagents/tools/python_tools.py uses AST filtering to block dangerous Python attributes like __subclasses__, __globals__, and __bases__. However, the filter only checks ast.Attribute nodes, allowing a bypass. The sandbox relies on AST-based filtering of attribute access but fails to account for dynamic attribute resolution via built-in methods such as type.getattribute, resulting in incomplete enforcement of security restrictions. The string '__subclasses__' is an ast.Constant, not an ast.Attribute, so it is never checked against the blocked list. This vulnerability is fixed in 4.5.128.
View original CVE JSON
{
"id": "CVE-2026-40158",
"sourceIdentifier": "[email protected]",
"published": "2026-04-10T17:17:13.603",
"lastModified": "2026-10-26T21:10:00.170",
"vulnStatus": "Analyzed",
"cveTags": [],
"descriptions": [
{
"lang": "en",
"value": "PraisonAI is a multi-agent teams system. Prior to 4.5.128, PraisonAI's AST-based Python sandbox can be bypassed using type.__getattribute__ trampoline, allowing arbitrary code execution when running untrusted agent code. The _execute_code_direct function in praisonaiagents/tools/python_tools.py uses AST filtering to block dangerous Python attributes like __subclasses__, __globals__, and __bases__. However, the filter only checks ast.Attribute nodes, allowing a bypass. The sandbox relies on AST-based filtering of attribute access but fails to account for dynamic attribute resolution via built-in methods such as type.getattribute, resulting in incomplete enforcement of security restrictions. The string '__subclasses__' is an ast.Constant, not an ast.Attribute, so it is never checked against the blocked list. This vulnerability is fixed in 4.5.128."
},
{
"lang": "es",
"value": "PraisonAI es un sistema de equipos multiagente. Antes de la 4.5.128, el sandbox de Python basado en AST de PraisonAI puede ser eludido utilizando el trampoline type.__getattribute__, permitiendo la ejecución de código arbitrario al ejecutar código de agente no confiable. La función _execute_code_direct en praisonaiagents/tools/python_tools.py utiliza filtrado AST para bloquear atributos peligrosos de Python como __subclasses__, __globals__ y __bases__. Sin embargo, el filtro solo verifica nodos ast.Attribute, permitiendo una elusión. El sandbox se basa en el filtrado basado en AST del acceso a atributos, pero no tiene en cuenta la resolución dinámica de atributos a través de métodos incorporados como type.getattribute, lo que resulta en una aplicación incompleta de las restricciones de seguridad. La cadena ’__subclasses__’ es un ast.Constant, no un ast.Attribute, por lo que nunca se verifica contra la lista bloqueada. Esta vulnerabilidad está corregida en la 4.5.128."
}
],
"affected": [
{
"source": "[email protected]",
"affectedData": [
{
"vendor": "MervinPraison",
"product": "PraisonAI",
"versions": [
{
"version": "< 4.5.128",
"status": "affected"
}
]
}
]
}
],
"metrics": {
"cvssMetricV31": [
{
"source": "[email protected]",
"type": "Secondary",
"cvssData": {
"version": "3.1",
"vectorString": "CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H",
"baseScore": 8.6,
"baseSeverity": "HIGH",
"attackVector": "LOCAL",
"attackComplexity": "LOW",
"privilegesRequired": "NONE",
"userInteraction": "REQUIRED",
"scope": "CHANGED",
"confidentialityImpact": "HIGH",
"integrityImpact": "HIGH",
"availabilityImpact": "HIGH"
},
"exploitabilityScore": 1.8,
"impactScore": 6.0
},
{
"source": "[email protected]",
"type": "Primary",
"cvssData": {
"version": "3.1",
"vectorString": "CVSS:3.1/AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H",
"baseScore": 7.8,
"baseSeverity": "HIGH",
"attackVector": "LOCAL",
"attackComplexity": "LOW",
"privilegesRequired": "NONE",
"userInteraction": "REQUIRED",
"scope": "UNCHANGED",
"confidentialityImpact": "HIGH",
"integrityImpact": "HIGH",
"availabilityImpact": "HIGH"
},
"exploitabilityScore": 1.8,
"impactScore": 5.9
}
],
"ssvcV203": [
{
"source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
"ssvcData": {
"timestamp": "2026-04-10T18:31:02.529674Z",
"id": "CVE-2026-40158",
"options": [
{
"exploitation": "poc"
},
{
"automatable": "no"
},
{
"technicalImpact": "total"
}
],
"role": "CISA Coordinator",
"version": "2.0.3"
}
}
]
},
"weaknesses": [
{
"source": "[email protected]",
"type": "Secondary",
"description": [
{
"lang": "en",
"value": "CWE-94"
},
{
"lang": "en",
"value": "CWE-693"
}
]
}
],
"configurations": [
{
"nodes": [
{
"operator": "OR",
"negate": false,
"cpeMatch": [
{
"vulnerable": true,
"criteria": "cpe:2.3:a:praison:praisonai:*:*:*:*:*:*:*:*",
"versionEndExcluding": "4.5.128",
"matchCriteriaId": "56CDE5F5-B03C-4C3A-9A92-F61C9DFDA9B1"
}
]
}
]
}
],
"references": [
{
"url": "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-3c4r-6p77-xwr7",
"source": "[email protected]",
"tags": [
"Exploit",
"Vendor Advisory"
]
}
]
}