Vulnerability · Published Oct 12, 2026 · Assigned by VulDB
CVE-2026-108789deepseek-ai deepseek-harness dsh-workflow-worker-thread agent behavioral workflow
Medium 5.5
In plain English
A medium-severity security flaw in deepseek-harness from deepseek-ai. Proof-of-concept exploit code is public. The chance it is exploited in the next 30 days is estimated at 0.2% (higher than 7% of all known vulnerabilities).
Official description
A vulnerability was identified in deepseek-ai deepseek-harness up to 0.1.1-rc.2. This affects the function agent of the component dsh-workflow-worker-thread. The manipulation leads to enforcement of behavioral workflow. The attack is possible to be carried out remotely. The exploit is publicly available and might be used. The vendor was contacted early about this disclosure but did not respond in any way.
Affected products (as reported)
| Vendor | Product | Versions |
|---|---|---|
| deepseek-ai | deepseek-harness | 0.1.1-rc.00.1.1-rc.10.1.1-rc.2 |
Weakness: CWE-841, CWE-840 · Vector: CVSS:3.1/AV:N/AC:L/PR:L/UI:R/S:U/C:L/I:L/A:L/E:P/RL:X/RC:R
References
- VDB-416568 | deepseek-ai deepseek-harness dsh-workflow-worker-thread agent behavioral workflow vdb-entry, technical-description
- VDB-416568 | CTI Indicators (IOB, IOC, IOA) signature, permissions-required
- CVE-2026-108789 | CVE Analysis and Report third-party-advisory
- Submit #959699 | 杭州深度求索人工智能基础技术研究有限公司 DeepSeek Harness 0.1.1-rc.2 逻辑缺陷 third-party-advisory
- https://github.com/deepseek-ai/deepseek-harness/discussions/243 related
- https://mcniyfj5bw9b.feishu.cn/docx/M5H6d9AbioNftYxwk9CchJgGnHc exploit
- https://github.com/deepseek-ai/deepseek-harness/ product