CVE-2026-82533 (CVSS 9.6)

Description

DeepSeek Harness before 0.1.2-alpha.1 contains an authentication bypass vulnerability that grants unauthenticated access to its local HTTP agent-control API by accepting a client-supplied loopback Host header in place of validating the actual TCP connection origin. On the default configuration, a confined tool-executed process can reach the loopback API without any port exposure and use it to escape its own OS sandbox, escalate to unconfined execution, and disable the approval prompt. When the port is externally reachable via tunnel, SSH forward, or reverse proxy, a remote attacker can exploit the same flaw to create sessions, execute arbitrary commands, and exfiltrate stored conversation transcripts without credentials.

Is your site exposed to CVE-2026-82533?

Run a free security scan — no signup, results in seconds.

CVSS v3.1 Score

9.6

CRITICAL

CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H

EPSS — Exploit Prediction

0.0062

Probability of exploitation

0.48%

Percentile rank

EPSS estimates the probability that this vulnerability will be exploited in the wild within the next 30 days. A higher score means more likely to be exploited.

Weakness Type (CWE)

CWE-807 CWE-807

References

Other References

https://github.com/deepseek-ai/deepseek-harness/commit/3e24087bfaeabe40b58ba2f7b936895b8f93fe27 https://github.com/deepseek-ai/deepseek-harness/releases/tag/dsh-v0.1.2-alpha.1 https://www.ox.security/blog/cve-2026-82533-deepseek-harness-ai-agent-sandbox-escape/ https://www.vulncheck.com/advisories/deepseek-harness-alpha-1-authentication-bypass-via-host-header-spoofing

Frequently Asked Questions

What is CVE-2026-82533? +

DeepSeek Harness before 0.1.2-alpha.1 contains an authentication bypass vulnerability that grants unauthenticated access to its local HTTP agent-control API by accepting a client-supplied loopback Host header in place of validating the actual TCP connection origin. On the default configuration, a confined tool-executed process can reach the loopback API without any port exposure and use it to escape its own OS sandbox, escalate to unconfined execution, and disable the approval prompt. When the port is externally reachable via tunnel, SSH forward, or reverse proxy, a remote attacker can exploit the same flaw to create sessions, execute arbitrary commands, and exfiltrate stored conversation transcripts without credentials. It has a CVSS v3.1 base score of 9.6 (CRITICAL).

How severe is CVE-2026-82533? +

CVE-2026-82533 has a CVSS v3.1 score of 9.6 out of 10, rated CRITICAL. This is a critical vulnerability that should be patched immediately. The EPSS score is 0.0062, placing it in the 0th percentile for exploitation probability.

How do I check if I'm vulnerable to CVE-2026-82533? +

You can use Secably's free Website Scanner to check your website for known vulnerabilities. For infrastructure scanning, use the Port Scanner to identify exposed services that may be affected. Check the vendor advisories linked above for specific patch and version information.

Related Vulnerabilities

CVE-2026-6213

A vulnerability in Remote Spark SparkView before build 1122 allows an attacker to bypasses the local connection check and achieve …

CVE-2024-9310

By utilizing software-defined radios and a custom low-latency processing pipeline, RF signals with spoofed location data can be transmitted to …

CVE-2025-24369

Anubis is a tool that allows administrators to protect bots against AI scrapers through bot-checking heuristics and a proof-of-work challenge …

CVE-2025-0117

A reliance on untrusted input for a security decision in the GlobalProtect app on Windows devices potentially enables a locally …

CVE-2026-9561

Eclipse Kura versions prior to 5.6.2 trust the client-supplied X-Forwarded-For HTTP header as the authoritative source of the client IP …

CVE-2026-39807

Reliance on Untrusted Inputs in a Security Decision vulnerability in mtrudel bandit allows unauthenticated transport-state spoofing on plaintext HTTP connections. …

CVE-2026-82533