Vulnerability · Published Sep 28, 2026 · Assigned by VulDB

CVE-2026-101102deepseek-ai deepseek-harness Code Mode Sandbox run_code sandbox

Medium 6.3

In plain English

A medium-severity security flaw in deepseek-harness from deepseek-ai. There are no confirmed reports of exploitation so far. The chance it is exploited in the next 30 days is estimated at 0.2% (higher than 13% of all known vulnerabilities).

Severity (CVSS 3.1)6.3Medium
Exploit odds (30 days)0.2%top 87.5% of all CVEs
Exploited in the wildNo reportsCISA SSVC: none
Attack toolingNoneno public tooling known

Official description

A vulnerability was found in deepseek-ai deepseek-harness up to 0.1.0-rc.7. Impacted is the function run_code of the component Code Mode Sandbox. The manipulation results in sandbox issue. The attack can be executed remotely. The vendor's own code, SAFETY.md, and design notes all explicitly state the worker is "containment, not a security boundary". The vendor was contacted early about this disclosure but did not respond in any way.

Affected products (as reported)

VendorProductVersions
deepseek-ai deepseek-harness 0.1.0-rc.00.1.0-rc.10.1.0-rc.20.1.0-rc.30.1.0-rc.40.1.0-rc.50.1.0-rc.60.1.0-rc.7

Weakness: CWE-265, CWE-264 · Vector: CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:L/I:L/A:L/E:X/RL:X/RC:C

References