# CVE-2026-82533 — DeepSeek Harness

> As of 2026-09-08, CVE-2026-82533 is a CRITICAL-severity vulnerability in DeepSeek Harness, CVSS v3.1 9.6, EPSS 0.4% (35.0th percentile). Threadlinqs Intelligence links 1 tracked threat campaign to CVE-2026-82533, most recently “DeepSeek Harness Authentication Bypass Lets Sandboxed AI Agents Escape via Single Command (CVE-2026-82533)”.

**Last updated:** 2026-09-08

## What is CVE-2026-82533?

DeepSeek Harness before 0.1.2-alpha.1 contains an authentication bypass vulnerability in its local HTTP control-plane API that allows attackers to gain full agent control by supplying a spoofed Host header, as the server validates only the client-supplied Host header value rather than the actual TCP connection origin. Attackers can exploit this flaw to invoke privileged commands such as commands/execute with danger-full-access permissions, escalate session approval policies to unconfined execution, and retrieve all stored conversations without any credential or API key.

The record classifies CVE-2026-82533 under weakness class [CWE-807](https://cwe.mitre.org/data/definitions/807.html). Its CVSS v3 base vector states that the flaw is reachable remotely over the network, needs no prior authentication, needs a user to take an action first, and has high impact on confidentiality, integrity, availability. 1 affected-product entry is recorded, across 1 vendor, listed below. The identifier was first published 4 days ago.

## Severity and exploitation probability

- **CVSS v3.1 base score:** 9.6 — CRITICAL (`CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H`)
- **CVSS v4.0 base score:** 9.4 (`CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:P/VC:H/VI:H/VA:H/SC:H/SI:H/SA:H`)
- **EPSS (FIRST):** 0.4% probability of exploitation in the next 30 days, higher than 35.0% of all scored CVEs
- **CISA KEV:** Not listed in the CISA Known Exploited Vulnerabilities catalog
- **Threadlinqs priority:** 6.2/10 — a Threadlinqs composite of the CVSS base score, the EPSS percentile and public exploit availability
- **Published:** 2026-09-08

## Is CVE-2026-82533 being exploited?

No public exploitation evidence for CVE-2026-82533 is recorded in the sources Threadlinqs tracks — CISA KEV, Exploit-DB, public proof-of-concept repositories and ProjectDiscovery Nuclei — as of 2026-09-08. That is an absence of evidence in those feeds, not a guarantee that the vulnerability is unexploited.

## Affected products and versions

- **DeepSeek**: Harness

## How to fix CVE-2026-82533

No vendor patch reference has been recorded for CVE-2026-82533 in the tracked sources. Follow the references below for a fix, and treat the products listed above as exposed until the vendor states otherwise.

## Threat activity tracking CVE-2026-82533

1 tracked threat in the Threadlinqs corpus references CVE-2026-82533, either in the campaign’s CVE list or as an indicator on the campaign record.

- [DeepSeek Harness Authentication Bypass Lets Sandboxed AI Agents Escape via Single Command (CVE-2026-82533)](https://intel.threadlinqs.com/threat/TL-2026-2412) — CRITICAL · 2026-09-09

## Sources

Enriched from CVE.org, NVD, FIRST EPSS, GitHub Security Advisories. Last verified by Threadlinqs on 2026-09-09. This product uses the NVD API but is not endorsed or certified by the NVD.

**Vendor advisory and patch**

- [github.com — release notes](https://github.com/deepseek-ai/deepseek-harness/releases/tag/dsh-v0.1.2-alpha.1)
- [github.com — patch](https://github.com/deepseek-ai/deepseek-harness/commit/3e24087bfaeabe40b58ba2f7b936895b8f93fe27)

**Third-party advisory**

- [vulncheck.com — third party advisory](https://www.vulncheck.com/advisories/deepseek-harness-alpha-1-authentication-bypass-via-host-header-spoofing)

Canonical: https://intel.threadlinqs.com/cve/CVE-2026-82533
Full detection coverage and IOCs for threats exploiting CVE-2026-82533 via the Threadlinqs MCP server (Purple tier): https://intel.threadlinqs.com/mcp