{"id":"CVE-2026-33068","aliases":["EUVD-2026-13631","GHSA-mmgp-wc2j-qcv7"],"status":"published","title":"Claude Code has a Workspace Trust Dialog Bypass via Repo-Controlled Settings File","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.","severity":{"version":"4.0","score":7.7,"vector":"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","level":"high","assessor":"cna","from":"cve"},"epss":{"day":"2026-11-10","score":0.00351,"percentile":0.29082,"model":"v2026.06.15"},"facts":{"in_kev":false,"known_exploited":false,"public_exploit":false,"fix_available":true},"cwes":["CWE-807"],"published":"2026-03-19T12:42:09Z","modified":"2026-10-28T02:30:25Z","fixes":[{"ecosystem":"npm","package":"@anthropic-ai/claude-code","fixed_in":["2.1.53"],"sources":["ghsa","osv"]}],"affected":[{"source":"ghsa","ecosystem":"npm","package":"@anthropic-ai/claude-code","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"2.1.53"}]}]},{"source":"osv","ecosystem":"npm","package":"@anthropic-ai/claude-code","ranges":[{"type":"SEMVER","events":[{"introduced":"0"},{"fixed":"2.1.53"}]}]}],"products":[{"vendor":"anthropic","product":"claude_code"},{"vendor":"anthropics","product":"claude-code"}],"references":[{"url":"https://github.com/anthropics/claude-code/security/advisories/GHSA-mmgp-wc2j-qcv7","tags":["x_refsource_CONFIRM","Vendor Advisory","advisory","web"]},{"url":"https://nvd.nist.gov/vuln/detail/CVE-2026-33068","tags":["advisory"]},{"url":"https://github.com/advisories/GHSA-mmgp-wc2j-qcv7"},{"url":"https://github.com/CVEProject/cvelistV5/tree/main/cves/2026/33xxx/CVE-2026-33068.json","tags":["advisory"]},{"url":"https://github.com/anthropics/claude-code","tags":["package"]}],"sources":{"cve":[{"id":"CVE-2026-33068","title":"Claude Code has a Workspace Trust Dialog Bypass via Repo-Controlled Settings File","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.","status":"published","published":"2026-03-20T08:17:47.794Z","modified":"2026-03-20T13:48:36.014Z","cvss":[{"version":"4.0","score":7.7,"vector":"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","level":"high","assessor":"cna"}],"cwes":["CWE-807"],"updated_at":"2026-11-02T14:20:08.009733Z"}],"euvd":[{"id":"EUVD-2026-13631","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.","status":"published","published":"2026-03-20T08:17:47Z","modified":"2026-03-20T13:48:36Z","cvss":[{"version":"4.0","score":7.7,"vector":"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","level":"high"}],"updated_at":"2026-11-02T19:25:32.430758Z"}],"ghsa":[{"id":"GHSA-mmgp-wc2j-qcv7","title":"Claude Code has a Workspace Trust Dialog Bypass via Repo-Controlled Settings File","description":"Claude Code 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.\n\nUsers on standard Claude Code auto-update have received this fix already. Users performing manual updates are advised to update to the latest version.\n\nThank you to hackerone.com/cantina_xyz for reporting this issue.","status":"published","published":"2026-03-19T12:42:09Z","modified":"2026-03-20T21:24:22Z","cvss":[{"version":"4.0","score":7.7,"vector":"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","level":"high"}],"cwes":["CWE-807"],"updated_at":"2026-11-02T14:39:39.567266Z"}],"nvd":[{"id":"CVE-2026-33068","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.","status":"published","published":"2026-03-20T09:16:15.017Z","modified":"2026-06-17T10:36:53.807Z","cvss":[{"version":"4.0","score":7.7,"vector":"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","level":"high","assessor":"secondary"},{"version":"3.1","score":8.8,"vector":"CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H","level":"high","assessor":"primary"}],"cwes":["CWE-807"],"updated_at":"2026-11-02T14:12:10.126802Z"}],"osv":[{"id":"CVE-2026-33068","title":"Claude Code has a Workspace Trust Dialog Bypass via Repo-Controlled Settings File","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.","status":"published","published":"2026-03-20T08:17:47.794Z","modified":"2026-10-28T02:30:25.275252333Z","cvss":[{"version":"4.0","score":7.7,"vector":"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","level":"high"}],"cwes":["CWE-807"],"updated_at":"2026-11-02T15:45:01.548715Z"},{"id":"GHSA-mmgp-wc2j-qcv7","title":"Claude Code has a Workspace Trust Dialog Bypass via Repo-Controlled Settings File","description":"Claude Code 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.\n\nUsers on standard Claude Code auto-update have received this fix already. Users performing manual updates are advised to update to the latest version.\n\nThank you to hackerone.com/cantina_xyz for reporting this issue.","status":"published","published":"2026-03-19T12:42:09Z","modified":"2026-03-20T21:33:50.318924Z","cvss":[{"version":"4.0","score":7.7,"vector":"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","level":"high"}],"cwes":["CWE-807"],"updated_at":"2026-11-02T15:46:52.170928Z"}]}}