CVE-2026-32128: FastGPT Python Sandbox file write restriction bypass via stdout remapping

Summary

CVE-2026-32128 affects FastGPT, an AI Agent building platform, in version 4.14.7 and earlier. Its Python Sandbox (fastgpt-sandbox) relies on static detection and seccomp to block file writes, but an attacker can remap stdout (fd 1) to an arbitrary writable file descriptor using fcntl. Because sys.stdout.write() still satisfies the seccomp rule write(fd==1), the attacker can create or overwrite arbitrary files inside the sandbox container.

Mitigation

The source does not state a fix yet. Check the original advisory for updates.

Topics

• AI agents

Related items

• MediumCVE-2026-121936: easyVoice server-side request forgery through openaiBaseUrl in /api/v1/ttsSimilar attack · NVD/CVE Database
• MediumCVE-2026-118628: MLflow get-artifacts server-side request forgery information disclosureSimilar attack · NVD/CVE Database
• HighCVE-2026-118625: MLflow Tracking Server privilege escalation via get_endpointsSimilar attack · NVD/CVE Database
• InfoWhen trust becomes the payload: Inside the threats, etc quarterly reportSimilar attack · Elastic Security Labs
• MediumCritical Langflow flaws enable authentication bypass and RCESimilar attack · CSO Online