CVE-2026-32128 (GCVE-0-2026-32128)
Vulnerability from cvelistv5 – Published: 2026-03-11 21:30 – Updated: 2026-03-12 16:18
VLAI?
Title
FastGPT Python Sandbox Bypass of File-Write Restriction
Summary
FastGPT is an AI Agent building platform. In 4.14.7 and earlier, FastGPT's Python Sandbox (fastgpt-sandbox) includes guardrails intended to prevent file writes (static detection + seccomp). These guardrails are bypassable by remapping stdout (fd 1) to an arbitrary writable file descriptor using fcntl. After remapping, writing via sys.stdout.write() still satisfies the seccomp rule write(fd==1), enabling arbitrary file creation/overwrite inside the sandbox container despite the intended no file writes restriction.
Severity ?
6.3 (Medium)
CWE
- CWE-184 - Incomplete List of Disallowed Inputs
Assigner
References
| URL | Tags | ||||
|---|---|---|---|---|---|
|
|||||
{
"containers": {
"adp": [
{
"metrics": [
{
"other": {
"content": {
"id": "CVE-2026-32128",
"options": [
{
"Exploitation": "poc"
},
{
"Automatable": "no"
},
{
"Technical Impact": "partial"
}
],
"role": "CISA Coordinator",
"timestamp": "2026-03-12T15:31:49.185736Z",
"version": "2.0.3"
},
"type": "ssvc"
}
}
],
"providerMetadata": {
"dateUpdated": "2026-03-12T16:18:13.936Z",
"orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
"shortName": "CISA-ADP"
},
"title": "CISA ADP Vulnrichment"
}
],
"cna": {
"affected": [
{
"product": "FastGPT",
"vendor": "labring",
"versions": [
{
"status": "affected",
"version": "\u003c= 4.14.7"
}
]
}
],
"descriptions": [
{
"lang": "en",
"value": "FastGPT is an AI Agent building platform. In 4.14.7 and earlier, FastGPT\u0027s Python Sandbox (fastgpt-sandbox) includes guardrails intended to prevent file writes (static detection + seccomp). These guardrails are bypassable by remapping stdout (fd 1) to an arbitrary writable file descriptor using fcntl. After remapping, writing via sys.stdout.write() still satisfies the seccomp rule write(fd==1), enabling arbitrary file creation/overwrite inside the sandbox container despite the intended no file writes restriction."
}
],
"metrics": [
{
"cvssV3_1": {
"attackComplexity": "LOW",
"attackVector": "NETWORK",
"availabilityImpact": "LOW",
"baseScore": 6.3,
"baseSeverity": "MEDIUM",
"confidentialityImpact": "LOW",
"integrityImpact": "LOW",
"privilegesRequired": "LOW",
"scope": "UNCHANGED",
"userInteraction": "NONE",
"vectorString": "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:L/I:L/A:L",
"version": "3.1"
}
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-184",
"description": "CWE-184: Incomplete List of Disallowed Inputs",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-03-11T21:30:26.336Z",
"orgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
"shortName": "GitHub_M"
},
"references": [
{
"name": "https://github.com/labring/FastGPT/security/advisories/GHSA-6hw6-mxrm-v6wj",
"tags": [
"x_refsource_CONFIRM"
],
"url": "https://github.com/labring/FastGPT/security/advisories/GHSA-6hw6-mxrm-v6wj"
}
],
"source": {
"advisory": "GHSA-6hw6-mxrm-v6wj",
"discovery": "UNKNOWN"
},
"title": "FastGPT Python Sandbox Bypass of File-Write Restriction"
}
},
"cveMetadata": {
"assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
"assignerShortName": "GitHub_M",
"cveId": "CVE-2026-32128",
"datePublished": "2026-03-11T21:30:26.336Z",
"dateReserved": "2026-03-10T22:19:36.545Z",
"dateUpdated": "2026-03-12T16:18:13.936Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2",
"vulnerability-lookup:meta": {
"nvd": "{\"cve\":{\"id\":\"CVE-2026-32128\",\"sourceIdentifier\":\"security-advisories@github.com\",\"published\":\"2026-03-11T22:16:32.633\",\"lastModified\":\"2026-03-19T17:38:40.920\",\"vulnStatus\":\"Analyzed\",\"cveTags\":[],\"descriptions\":[{\"lang\":\"en\",\"value\":\"FastGPT is an AI Agent building platform. In 4.14.7 and earlier, FastGPT\u0027s Python Sandbox (fastgpt-sandbox) includes guardrails intended to prevent file writes (static detection + seccomp). These guardrails are bypassable by remapping stdout (fd 1) to an arbitrary writable file descriptor using fcntl. After remapping, writing via sys.stdout.write() still satisfies the seccomp rule write(fd==1), enabling arbitrary file creation/overwrite inside the sandbox container despite the intended no file writes restriction.\"},{\"lang\":\"es\",\"value\":\"FastGPT es una plataforma de creaci\u00f3n de agentes de IA. En 4.14.7 y versiones anteriores, el sandbox de Python de FastGPT (fastgpt-sandbox) incluye barreras de seguridad destinadas a prevenir escrituras de archivos (detecci\u00f3n est\u00e1tica + seccomp). Estas barreras de seguridad son eludibles reasignando stdout (fd 1) a un descriptor de archivo arbitrario con permisos de escritura usando fcntl. Despu\u00e9s de la reasignaci\u00f3n, la escritura a trav\u00e9s de sys.stdout.write() a\u00fan satisface la regla de seccomp write(fd==1), lo que permite la creaci\u00f3n/sobrescritura arbitraria de archivos dentro del contenedor del sandbox a pesar de la restricci\u00f3n intencionada de no escritura de archivos.\"}],\"metrics\":{\"cvssMetricV31\":[{\"source\":\"security-advisories@github.com\",\"type\":\"Secondary\",\"cvssData\":{\"version\":\"3.1\",\"vectorString\":\"CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:L/I:L/A:L\",\"baseScore\":6.3,\"baseSeverity\":\"MEDIUM\",\"attackVector\":\"NETWORK\",\"attackComplexity\":\"LOW\",\"privilegesRequired\":\"LOW\",\"userInteraction\":\"NONE\",\"scope\":\"UNCHANGED\",\"confidentialityImpact\":\"LOW\",\"integrityImpact\":\"LOW\",\"availabilityImpact\":\"LOW\"},\"exploitabilityScore\":2.8,\"impactScore\":3.4}]},\"weaknesses\":[{\"source\":\"security-advisories@github.com\",\"type\":\"Primary\",\"description\":[{\"lang\":\"en\",\"value\":\"CWE-184\"}]}],\"configurations\":[{\"nodes\":[{\"operator\":\"OR\",\"negate\":false,\"cpeMatch\":[{\"vulnerable\":true,\"criteria\":\"cpe:2.3:a:fastgpt:fastgpt:*:*:*:*:*:*:*:*\",\"versionEndIncluding\":\"4.14.7\",\"matchCriteriaId\":\"EE37BFC6-6051-4C66-AED8-9F7F80DC3EE6\"}]}]}],\"references\":[{\"url\":\"https://github.com/labring/FastGPT/security/advisories/GHSA-6hw6-mxrm-v6wj\",\"source\":\"security-advisories@github.com\",\"tags\":[\"Vendor Advisory\",\"Exploit\",\"Mitigation\"]}]}}",
"vulnrichment": {
"containers": "{\"adp\": [{\"title\": \"CISA ADP Vulnrichment\", \"metrics\": [{\"other\": {\"type\": \"ssvc\", \"content\": {\"id\": \"CVE-2026-32128\", \"role\": \"CISA Coordinator\", \"options\": [{\"Exploitation\": \"poc\"}, {\"Automatable\": \"no\"}, {\"Technical Impact\": \"partial\"}], \"version\": \"2.0.3\", \"timestamp\": \"2026-03-12T15:31:49.185736Z\"}}}], \"providerMetadata\": {\"orgId\": \"134c704f-9b21-4f2e-91b3-4a467353bcc0\", \"shortName\": \"CISA-ADP\", \"dateUpdated\": \"2026-03-12T15:31:52.975Z\"}}], \"cna\": {\"title\": \"FastGPT Python Sandbox Bypass of File-Write Restriction\", \"source\": {\"advisory\": \"GHSA-6hw6-mxrm-v6wj\", \"discovery\": \"UNKNOWN\"}, \"metrics\": [{\"cvssV3_1\": {\"scope\": \"UNCHANGED\", \"version\": \"3.1\", \"baseScore\": 6.3, \"attackVector\": \"NETWORK\", \"baseSeverity\": \"MEDIUM\", \"vectorString\": \"CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:L/I:L/A:L\", \"integrityImpact\": \"LOW\", \"userInteraction\": \"NONE\", \"attackComplexity\": \"LOW\", \"availabilityImpact\": \"LOW\", \"privilegesRequired\": \"LOW\", \"confidentialityImpact\": \"LOW\"}}], \"affected\": [{\"vendor\": \"labring\", \"product\": \"FastGPT\", \"versions\": [{\"status\": \"affected\", \"version\": \"\u003c= 4.14.7\"}]}], \"references\": [{\"url\": \"https://github.com/labring/FastGPT/security/advisories/GHSA-6hw6-mxrm-v6wj\", \"name\": \"https://github.com/labring/FastGPT/security/advisories/GHSA-6hw6-mxrm-v6wj\", \"tags\": [\"x_refsource_CONFIRM\"]}], \"descriptions\": [{\"lang\": \"en\", \"value\": \"FastGPT is an AI Agent building platform. In 4.14.7 and earlier, FastGPT\u0027s Python Sandbox (fastgpt-sandbox) includes guardrails intended to prevent file writes (static detection + seccomp). These guardrails are bypassable by remapping stdout (fd 1) to an arbitrary writable file descriptor using fcntl. After remapping, writing via sys.stdout.write() still satisfies the seccomp rule write(fd==1), enabling arbitrary file creation/overwrite inside the sandbox container despite the intended no file writes restriction.\"}], \"problemTypes\": [{\"descriptions\": [{\"lang\": \"en\", \"type\": \"CWE\", \"cweId\": \"CWE-184\", \"description\": \"CWE-184: Incomplete List of Disallowed Inputs\"}]}], \"providerMetadata\": {\"orgId\": \"a0819718-46f1-4df5-94e2-005712e83aaa\", \"shortName\": \"GitHub_M\", \"dateUpdated\": \"2026-03-11T21:30:26.336Z\"}}}",
"cveMetadata": "{\"cveId\": \"CVE-2026-32128\", \"state\": \"PUBLISHED\", \"dateUpdated\": \"2026-03-12T16:18:13.936Z\", \"dateReserved\": \"2026-03-10T22:19:36.545Z\", \"assignerOrgId\": \"a0819718-46f1-4df5-94e2-005712e83aaa\", \"datePublished\": \"2026-03-11T21:30:26.336Z\", \"assignerShortName\": \"GitHub_M\"}",
"dataType": "CVE_RECORD",
"dataVersion": "5.2"
}
}
}
Loading…
Loading…
Sightings
| Author | Source | Type | Date |
|---|
Nomenclature
- Seen: The vulnerability was mentioned, discussed, or observed by the user.
- Confirmed: The vulnerability has been validated from an analyst's perspective.
- Published Proof of Concept: A public proof of concept is available for this vulnerability.
- Exploited: The vulnerability was observed as exploited by the user who reported the sighting.
- Patched: The vulnerability was observed as successfully patched by the user who reported the sighting.
- Not exploited: The vulnerability was not observed as exploited by the user who reported the sighting.
- Not confirmed: The user expressed doubt about the validity of the vulnerability.
- Not patched: The vulnerability was not observed as successfully patched by the user who reported the sighting.
Loading…
Loading…