How severe is CVE-2026-33068?

CVE-2026-33068 has a CVSS 3.x base score of 8.8, rated high severity. It is exploitable over network with low attack complexity, requires no privileges and user interaction. Impact on confidentiality is high, integrity high, and availability high.

Is CVE-2026-33068 being actively exploited?

It is not currently listed in CISA's KEV catalog. Its EPSS exploit-prediction score is 0% (26th percentile), an estimate of the probability of exploitation in the next 30 days.

What products are affected by CVE-2026-33068?

CVE-2026-33068 affects Anthropic Claude Code. See the affected-products list for the exact vulnerable versions.

How do I fix CVE-2026-33068?

Review the linked vendor and NVD advisories for patched versions and mitigations, then upgrade or apply the recommended workaround. Given its high severity, prioritise patching exposed systems.

Does CVE-2026-33068 have an EU (EUVD) identifier?

Yes. CVE-2026-33068 is tracked in the ENISA EU Vulnerability Database (EUVD) as EUVD-2026-13631.

When was CVE-2026-33068 published?

CVE-2026-33068 was published on 2026-03-20 and last updated on 2026-06-17.

CVE-2026-33068

CVE-2026-33068 is a high-severity vulnerability in Anthropic Claude Code with a CVSS 3.x base score of 8.8. It is not currently listed as actively exploited by CISA, and its EPSS exploit-prediction score is low. The underlying weakness is classified as CWE-807.

Key facts

Severity: High (CVSS 3.x base score 8.8)
CVSS v4: 7.7
EPSS exploit prediction: 0% (26th percentile)
Actively exploited: Not listed in CISA KEV
EU (EUVD) id: EUVD-2026-13631
Weakness: CWE-807
Affected product: Anthropic Claude Code
Published: 2026-03-20
Last modified: 2026-06-17

Description

Claude Code is an agentic coding tool. Versions prior to 2.1.53 resolved the permission mode from settings files, including the repo-controlled .claude/settings.json, before determining whether to display the workspace trust confirmation dialog. A malicious repository could set permissions.defaultMode to bypassPermissions in its committed .claude/settings.json, causing the trust dialog to be silently skipped on first open. This allowed a user to be placed into a permissive mode without seeing the trust confirmation prompt, making it easier for an attacker-controlled repository to gain tool execution without explicit user consent. This issue has been patched in version 2.1.53.

Frequently asked questions

What is CVE-2026-33068?: Claude Code is an agentic coding tool. Versions prior to 2.1.53 resolved the permission mode from settings files, including the repo-controlled .claude/settings.json, before determining whether to display the workspace trust confirmation dialog. A malicious repository could set permissions.defaultMode to bypassPermissions in its committed .claude/settings.json, causing the trust dialog to be silently skipped on first open. This allowed a user to be placed into a permissive mode without seeing the trust confirmation prompt, making it easier for an attacker-controlled repository to gain tool execution without explicit user consent. This issue has been patched in version 2.1.53.
How severe is CVE-2026-33068?: CVE-2026-33068 has a CVSS 3.x base score of 8.8, rated high severity. It is exploitable over network with low attack complexity, requires no privileges and user interaction. Impact on confidentiality is high, integrity high, and availability high.
Is CVE-2026-33068 being actively exploited?: It is not currently listed in CISA's KEV catalog. Its EPSS exploit-prediction score is 0% (26th percentile), an estimate of the probability of exploitation in the next 30 days.
What products are affected by CVE-2026-33068?: CVE-2026-33068 affects Anthropic Claude Code. See the affected-products list for the exact vulnerable versions.
How do I fix CVE-2026-33068?: Review the linked vendor and NVD advisories for patched versions and mitigations, then upgrade or apply the recommended workaround. Given its high severity, prioritise patching exposed systems.
Does CVE-2026-33068 have an EU (EUVD) identifier?: Yes. CVE-2026-33068 is tracked in the ENISA EU Vulnerability Database (EUVD) as EUVD-2026-13631.
When was CVE-2026-33068 published?: CVE-2026-33068 was published on 2026-03-20 and last updated on 2026-06-17.

References

https://github.com/anthropics/claude-code/security/advisories/GHSA-mmgp-wc2j-qcv7

Affected products (1)

cpe:2.3:a:anthropic:claude_code:*:*:*:*:*:node.js:*:*

More vulnerabilities in Anthropic Claude Code

CVE-2026-39861 — Critical (CVSS 10.0): Claude Code is an agentic coding tool. Prior to version 2.1.64, Claude Code's sandbox did not prevent sandboxed…
CVE-2026-25725 — Critical (CVSS 10.0): Claude Code is an agentic coding tool. Prior to version 2.1.2, Claude Code's bubblewrap sandboxing mechanism failed to…
CVE-2025-66032 — Critical (CVSS 9.8): Claude Code is an agentic coding tool. Prior to 1.0.93, Due to errors in parsing shell commands related to $IFS and…
CVE-2025-64755 — Critical (CVSS 9.8): Claude Code is an agentic coding tool. Prior to version 2.0.31, due to an error in sed command parsing, it was possible…
CVE-2025-65099 — Critical (CVSS 9.8): Claude Code is an agentic coding tool. Prior to version 1.0.39, when running on a machine with Yarn 3.0 or above,…
CVE-2025-59828 — Critical (CVSS 9.8): Claude Code is an agentic coding tool. Prior to Claude Code version 1.0.39, when using Claude Code with Yarn versions…

All CVEs affecting Anthropic Claude Code →

Other CWE-807 vulnerabilities

CVE-2025-13926 — Critical (CVSS 9.8): An attacker could use data obtained by sniffing the network traffic to forge packets in order to make arbitrary…
CVE-2026-32975 — Critical (CVSS 9.8): OpenClaw before 2026.3.12 contains a weak authorization vulnerability in Zalouser allowlist mode that matches mutable…
CVE-2025-12488 — Critical (CVSS 9.8): oobabooga text-generation-webui trust_remote_code Reliance on Untrusted Inputs Remote Code Execution Vulnerability.…
CVE-2025-12487 — Critical (CVSS 9.8): oobabooga text-generation-webui trust_remote_code Reliance on Untrusted Inputs Remote Code Execution Vulnerability.…
CVE-2025-49827 — Critical (CVSS 9.8): Conjur provides secrets management and application identity for infrastructure. Conjur OSS versions 1.19.5 through…
CVE-2025-1126 — Critical (CVSS 9.3): A Reliance on Untrusted Inputs in a Security Decision vulnerability has been identified in the Lexmark Print Management…

Browse all CWE-807 vulnerabilities →