PT-2026-24852 · Labring+2 · Fastgpt

Failbutwin + 1 · Publicado 2026-03-11 · Atualizado 2026-03-11 · CVE-2026-32128

Positive Technologies (AI Summary)

CVSS v3.1

6.3

Média

Vetor base

AV:N/AC:L/PR:L/UI:N/S:U/C:L/I:L/A:L

Name of the Vulnerable Software and Affected Versions FastGPT versions 4.14.7 and earlier

Description FastGPT is an AI Agent building platform. The Python Sandbox (fastgpt-sandbox) includes security measures to prevent file writes using static detection and seccomp. These measures can be bypassed by remapping standard output (stdout), file descriptor 1, to an arbitrary writable file descriptor using the fcntl function. Following this remapping, writing through sys.stdout.write() still complies with the seccomp rule write(fd==1), allowing for arbitrary file creation and overwriting within the sandbox container, despite the intended restriction against file writes. The fcntl function is used to manipulate file descriptors.

Recommendations Versions prior to 4.14.7 should be updated.

Exploit

Correção

Incomplete List of Disallowed Inputs

Enumeração de Fraquezas

CWE-184

Identificadores relacionados

CVE-2026-32128

GHSA-6HW6-MXRM-V6WJ

Produtos afetados

Fastgpt

Referências · 6

• 🔥 https://github.com/labring/FastGPT/security/advisories/GHSA-6hw6-mxrm-v6wj · Exploit
• 🔥 https://github.com/CVEProject/cvelistV5/tree/main/cves/2026/32xxx/CVE-2026-32128.json · Exploit
• https://t.me/CVEtracker/47185 · Publicação no Telegram
• https://osv.dev/vulnerability/CVE-2026-32128 · Aviso do Fabricante
• https://nvd.nist.gov/vuln/detail/CVE-2026-32128 · Nota de Segurança
• https://twitter.com/CVEnew/status/2031867623378489632 · Publicação no Twitter