{"dataType": "CVE_RECORD", "dataVersion": "5.2", "cveMetadata": {"cveId": "CVE-2026-33068", "assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa", "state": "PUBLISHED", "assignerShortName": "GitHub_M", "dateReserved": "2026-03-17T19:27:06.343Z", "datePublished": "2026-03-20T08:17:47.794Z", "dateUpdated": "2026-03-20T13:48:36.014Z"}, "containers": {"cna": {"title": "Claude Code has a Workspace Trust Dialog Bypass via Repo-Controlled Settings File", "problemTypes": [{"descriptions": [{"cweId": "CWE-807", "lang": "en", "description": "CWE-807: Reliance on Untrusted Inputs in a Security Decision", "type": "CWE"}]}], "metrics": [{"cvssV4_0": {"attackVector": "NETWORK", "attackComplexity": "LOW", "attackRequirements": "PRESENT", "privilegesRequired": "NONE", "userInteraction": "PASSIVE", "vulnConfidentialityImpact": "HIGH", "vulnIntegrityImpact": "HIGH", "vulnAvailabilityImpact": "HIGH", "subConfidentialityImpact": "NONE", "subIntegrityImpact": "NONE", "subAvailabilityImpact": "NONE", "baseScore": 7.7, "baseSeverity": "HIGH", "vectorString": "CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:P/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N", "version": "4.0"}}], "references": [{"name": "https://github.com/anthropics/claude-code/security/advisories/GHSA-mmgp-wc2j-qcv7", "tags": ["x_refsource_CONFIRM"], "url": "https://github.com/anthropics/claude-code/security/advisories/GHSA-mmgp-wc2j-qcv7"}], "affected": [{"vendor": "anthropics", "product": "claude-code", "versions": [{"version": "< 2.1.53", "status": "affected"}]}], "providerMetadata": {"orgId": "a0819718-46f1-4df5-94e2-005712e83aaa", "shortName": "GitHub_M", "dateUpdated": "2026-03-20T08:17:47.794Z"}, "descriptions": [{"lang": "en", "value": "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."}], "source": {"advisory": "GHSA-mmgp-wc2j-qcv7", "discovery": "UNKNOWN"}}, "adp": [{"metrics": [{"other": {"type": "ssvc", "content": {"timestamp": "2026-03-20T13:48:28.263574Z", "id": "CVE-2026-33068", "options": [{"Exploitation": "none"}, {"Automatable": "no"}, {"Technical Impact": "total"}], "role": "CISA Coordinator", "version": "2.0.3"}}}], "title": "CISA ADP Vulnrichment", "providerMetadata": {"orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0", "shortName": "CISA-ADP", "dateUpdated": "2026-03-20T13:48:36.014Z"}}]}}

{"dataType": "CVE_RECORD", "containers": {"adp": [{"title": "CISA ADP Vulnrichment", "metrics": [{"other": {"type": "ssvc", "content": {"id": "CVE-2026-33068", "role": "CISA Coordinator", "options": [{"Exploitation": "none"}, {"Automatable": "no"}, {"Technical Impact": "total"}], "version": "2.0.3", "timestamp": "2026-03-20T13:48:28.263574Z"}}}], "providerMetadata": {"orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0", "shortName": "CISA-ADP", "dateUpdated": "2026-03-20T13:48:31.483Z"}}], "cna": {"title": "Claude Code has a Workspace Trust Dialog Bypass via Repo-Controlled Settings File", "source": {"advisory": "GHSA-mmgp-wc2j-qcv7", "discovery": "UNKNOWN"}, "metrics": [{"cvssV4_0": {"version": "4.0", "baseScore": 7.7, "attackVector": "NETWORK", "baseSeverity": "HIGH", "vectorString": "CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:P/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N", "userInteraction": "PASSIVE", "attackComplexity": "LOW", "attackRequirements": "PRESENT", "privilegesRequired": "NONE", "subIntegrityImpact": "NONE", "vulnIntegrityImpact": "HIGH", "subAvailabilityImpact": "NONE", "vulnAvailabilityImpact": "HIGH", "subConfidentialityImpact": "NONE", "vulnConfidentialityImpact": "HIGH"}}], "affected": [{"vendor": "anthropics", "product": "claude-code", "versions": [{"status": "affected", "version": "< 2.1.53"}]}], "references": [{"url": "https://github.com/anthropics/claude-code/security/advisories/GHSA-mmgp-wc2j-qcv7", "name": "https://github.com/anthropics/claude-code/security/advisories/GHSA-mmgp-wc2j-qcv7", "tags": ["x_refsource_CONFIRM"]}], "descriptions": [{"lang": "en", "value": "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."}], "problemTypes": [{"descriptions": [{"lang": "en", "type": "CWE", "cweId": "CWE-807", "description": "CWE-807: Reliance on Untrusted Inputs in a Security Decision"}]}], "providerMetadata": {"orgId": "a0819718-46f1-4df5-94e2-005712e83aaa", "shortName": "GitHub_M", "dateUpdated": "2026-03-20T08:17:47.794Z"}}}, "cveMetadata": {"cveId": "CVE-2026-33068", "state": "PUBLISHED", "dateUpdated": "2026-03-20T13:48:36.014Z", "dateReserved": "2026-03-17T19:27:06.343Z", "assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa", "datePublished": "2026-03-20T08:17:47.794Z", "assignerShortName": "GitHub_M"}, "dataVersion": "5.2"}

{"cveTags": [], "descriptions": [{"lang": "en", "value": "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."}], "id": "CVE-2026-33068", "lastModified": "2026-03-20T13:37:50.737", "metrics": {"cvssMetricV40": [{"cvssData": {"Automatable": "NOT_DEFINED", "Recovery": "NOT_DEFINED", "Safety": "NOT_DEFINED", "attackComplexity": "LOW", "attackRequirements": "PRESENT", "attackVector": "NETWORK", "availabilityRequirement": "NOT_DEFINED", "baseScore": 7.7, "baseSeverity": "HIGH", "confidentialityRequirement": "NOT_DEFINED", "exploitMaturity": "NOT_DEFINED", "integrityRequirement": "NOT_DEFINED", "modifiedAttackComplexity": "NOT_DEFINED", "modifiedAttackRequirements": "NOT_DEFINED", "modifiedAttackVector": "NOT_DEFINED", "modifiedPrivilegesRequired": "NOT_DEFINED", "modifiedSubAvailabilityImpact": "NOT_DEFINED", "modifiedSubConfidentialityImpact": "NOT_DEFINED", "modifiedSubIntegrityImpact": "NOT_DEFINED", "modifiedUserInteraction": "NOT_DEFINED", "modifiedVulnAvailabilityImpact": "NOT_DEFINED", "modifiedVulnConfidentialityImpact": "NOT_DEFINED", "modifiedVulnIntegrityImpact": "NOT_DEFINED", "privilegesRequired": "NONE", "providerUrgency": "NOT_DEFINED", "subAvailabilityImpact": "NONE", "subConfidentialityImpact": "NONE", "subIntegrityImpact": "NONE", "userInteraction": "PASSIVE", "valueDensity": "NOT_DEFINED", "vectorString": "CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:P/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X", "version": "4.0", "vulnAvailabilityImpact": "HIGH", "vulnConfidentialityImpact": "HIGH", "vulnIntegrityImpact": "HIGH", "vulnerabilityResponseEffort": "NOT_DEFINED"}, "source": "security-advisories@github.com", "type": "Secondary"}]}, "published": "2026-03-20T09:16:15.017", "references": [{"source": "security-advisories@github.com", "url": "https://github.com/anthropics/claude-code/security/advisories/GHSA-mmgp-wc2j-qcv7"}], "sourceIdentifier": "security-advisories@github.com", "vulnStatus": "Awaiting Analysis", "weaknesses": [{"description": [{"lang": "en", "value": "CWE-807"}], "source": "security-advisories@github.com", "type": "Primary"}]}

{}

{"affected": [{"configurations": [{"platform": null, "status": "affected", "versions": {"scheme": "semver", "value": "[0,2.1.53)"}}], "enrichment": {"confidence": 100.0, "confidence_source": "matching", "scores": [{"score": 100.0, "source": "matching"}]}, "original": {"product": "claude-code", "source": "cna", "vendor": "anthropics"}, "product": "claude_code", "vendor": "anthropics"}], "analysis": {"en": {"generated_at": "2026-03-20T09:22:54.246589+00:00", "value": {"mitigation_remediation": ["Apply the vendor patch to update Claude Code to version 2.1.53 or later."], "summary": {"action": "Immediate Patch", "impact": "Execution of malicious code without user consent"}, "threat_synthesis": {"affected_systems": "The affected product is anthropics:claude-code, specifically versions prior to 2.1.53. The vulnerability was addressed in release 2.1.53, which introduced checks to resolve permission mode before displaying the workspace trust dialog.", "description_and_impact": "Claude Code, an agentic coding tool, had a flaw in versions prior to 2.1.53 where the permission mode from settings files\u2014including the repository\u2011controlled .claude/settings.json\u2014was determined before the workspace trust confirmation dialog was shown. A malicious repository could set permissions.defaultMode to bypassPermissions in its committed settings file, causing the trust dialog to be silently skipped on first open. This allowed an attacker to place a user into a permissive mode without any visible confirmation, enabling tool execution without explicit user consent. The weakness is classified under CWE\u2011807, reflecting improper authorization or trust handling.", "risk_and_exploitability": "The CVSS base score is 7.7, indicating high severity. EPSS data is unavailable, and the vulnerability is not listed in the CISA KEV catalog. Exploitation requires a repository containing a committed .claude/settings.json that sets permissions.defaultMode to bypassPermissions. The attack vector is essentially a malicious repository that the user opens in Claude Code; no user interaction beyond opening the repository is required because the trust dialog is suppressed automatically. Consequently, the risk is high within environments that accept untrusted repositories without additional verification."}}}}, "created": "2026-03-20T10:36:38.292011+00:00", "updated": "2026-03-20T16:27:44.370788+00:00", "vendors": ["anthropics", "anthropics$PRODUCT$claude_code"]}