{"id":"CVE-2026-45001","aliases":["EUVD-2026-29146","GHSA-7jm2-g593-4qrc","GHSA-9fc9-8v4x-f5cp"],"status":"published","title":"OpenClaw \u003c 2026.4.20 - Gateway Config Mutation Guard Bypass via Agent Tool Access","description":"OpenClaw before 2026.4.20 contains a guard bypass vulnerability in the agent-facing gateway config.patch and config.apply endpoints that fails to protect operator-trusted settings including sandbox policy, plugin enablement, gateway auth/TLS, hook routing, MCP server configuration, SSRF policy, and filesystem hardening. A prompt-injected model with access to the owner-only gateway tool can persist unauthorized changes to protected operator settings.","severity":{"version":"4.0","score":6,"vector":"CVSS:4.0/AV:N/AC:L/AT:P/PR:L/UI:N/VC:L/VI:H/VA:N/SC:N/SI:N/SA:N","level":"medium","assessor":"cna","from":"cve"},"epss":{"day":"2026-11-10","score":0.00218,"percentile":0.12337,"model":"v2026.06.15"},"facts":{"in_kev":false,"known_exploited":false,"public_exploit":false,"fix_available":true},"cwes":["CWE-1220","CWE-285","CWE-862"],"published":"2026-04-25T23:51:11Z","modified":"2026-10-28T02:30:32Z","fixes":[{"ecosystem":"npm","package":"openclaw","fixed_in":["2026.4.20"],"sources":["ghsa","osv"]}],"affected":[{"source":"ghsa","ecosystem":"npm","package":"openclaw","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"2026.4.20"}]}]},{"source":"osv","ecosystem":"npm","package":"openclaw","ranges":[{"type":"SEMVER","events":[{"introduced":"0"},{"fixed":"2026.4.20"}]}]}],"products":[{"vendor":"openclaw","product":"openclaw"}],"references":[{"url":"https://github.com/openclaw/openclaw/security/advisories/GHSA-7jm2-g593-4qrc","tags":["vendor-advisory","Third Party Advisory","advisory","web"]},{"url":"https://github.com/openclaw/openclaw/commit/fe30b31a97a917ecc6e92f6c85378b6b20352422","tags":["patch","Patch","fix","web"]},{"url":"https://www.vulncheck.com/advisories/openclaw-gateway-config-mutation-guard-bypass-via-agent-tool-access","tags":["third-party-advisory","Patch","Third Party Advisory","advisory"]},{"url":"https://github.com/advisories/GHSA-7jm2-g593-4qrc"},{"url":"https://nvd.nist.gov/vuln/detail/CVE-2026-45001","tags":["advisory"]},{"url":"https://github.com/advisories/GHSA-9fc9-8v4x-f5cp"},{"url":"https://github.com/CVEProject/cvelistV5/tree/main/cves/2026/45xxx/CVE-2026-45001.json","tags":["advisory"]},{"url":"https://github.com/openclaw/openclaw","tags":["package"]}],"sources":{"cve":[{"id":"CVE-2026-45001","title":"OpenClaw \u003c 2026.4.20 - Gateway Config Mutation Guard Bypass via Agent Tool Access","description":"OpenClaw before 2026.4.20 contains a guard bypass vulnerability in the agent-facing gateway config.patch and config.apply endpoints that fails to protect operator-trusted settings including sandbox policy, plugin enablement, gateway auth/TLS, hook routing, MCP server configuration, SSRF policy, and filesystem hardening. A prompt-injected model with access to the owner-only gateway tool can persist unauthorized changes to protected operator settings.","status":"published","published":"2026-05-11T16:46:39.888Z","modified":"2026-05-11T17:51:23.651Z","cvss":[{"version":"4.0","score":6,"vector":"CVSS:4.0/AV:N/AC:L/AT:P/PR:L/UI:N/VC:L/VI:H/VA:N/SC:N/SI:N/SA:N","level":"medium","assessor":"cna"},{"version":"3.1","score":7.1,"vector":"CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:L/I:H/A:N","level":"high","assessor":"cna"}],"cwes":["CWE-862"],"updated_at":"2026-11-02T13:46:43.922571Z"}],"euvd":[{"id":"EUVD-2026-29146","description":"OpenClaw before 2026.4.20 contains a guard bypass vulnerability in the agent-facing gateway config.patch and config.apply endpoints that fails to protect operator-trusted settings including sandbox policy, plugin enablement, gateway auth/TLS, hook routing, MCP server configuration, SSRF policy, and filesystem hardening. A prompt-injected model with access to the owner-only gateway tool can persist unauthorized changes to protected operator settings.","status":"published","published":"2026-05-11T16:46:39Z","modified":"2026-05-11T17:51:23Z","cvss":[{"version":"4.0","score":6,"vector":"CVSS:4.0/AV:N/AC:L/AT:P/PR:L/UI:N/VC:L/VI:H/VA:N/SC:N/SI:N/SA:N","level":"medium"}],"updated_at":"2026-11-02T19:05:10.736891Z"}],"ghsa":[{"id":"GHSA-7jm2-g593-4qrc","title":"OpenClaw: Agent gateway config mutations could change protected operator settings","description":"## Affected Packages / Versions\n\n- Package: `openclaw` (npm)\n- Affected versions: `\u003c 2026.4.20`\n- Patched version: `2026.4.20`\n\n## Impact\n\nThe agent-facing `gateway config.patch` / `config.apply` guard did not cover several operator-trusted settings, including sandbox policy, plugin enablement, gateway auth/TLS, hook routing, MCP server configuration, SSRF policy, and filesystem hardening. A prompt-injected model with access to the owner-only gateway tool could persist changes to those settings.\n\nThis is a model-to-operator guard bypass, not a remote unauthenticated gateway compromise. Severity is medium.\n\n## Fix\n\nOpenClaw now blocks model-driven gateway config mutations for the broader operator-trusted path set and covers per-agent overrides and array-entry patching.\n\nFix commit:\n\n- `fe30b31a97a917ecc6e92f6c85378b6b20352422`\n\n## Release\n\nFixed in OpenClaw `2026.4.20`.","status":"published","published":"2026-04-25T23:51:11Z","modified":"2026-04-25T23:51:11Z","cvss":[{"version":"4.0","score":6,"vector":"CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:P/VC:N/VI:H/VA:L/SC:N/SI:N/SA:N","level":"medium"}],"cwes":["CWE-285","CWE-1220"],"updated_at":"2026-11-02T14:39:58.897998Z"},{"id":"GHSA-9fc9-8v4x-f5cp","title":"OpenClaw before 2026.4.20 contains a guard bypass vulnerability in the agent-facing gateway...","description":"OpenClaw before 2026.4.20 contains a guard bypass vulnerability in the agent-facing gateway config.patch and config.apply endpoints that fails to protect operator-trusted settings including sandbox policy, plugin enablement, gateway auth/TLS, hook routing, MCP server configuration, SSRF policy, and filesystem hardening. A prompt-injected model with access to the owner-only gateway tool can persist unauthorized changes to protected operator settings.","status":"published","published":"2026-05-11T18:31:47Z","modified":"2026-05-11T18:31:59Z","cvss":[{"version":"4.0","score":6,"vector":"CVSS:4.0/AV:N/AC:L/AT:P/PR:L/UI:N/VC:L/VI:H/VA:N/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":"medium"},{"version":"3.1","score":7.1,"vector":"CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:L/I:H/A:N","level":"high"}],"cwes":["CWE-862"],"updated_at":"2026-11-02T15:07:59.393178Z"}],"nvd":[{"id":"CVE-2026-45001","description":"OpenClaw before 2026.4.20 contains a guard bypass vulnerability in the agent-facing gateway config.patch and config.apply endpoints that fails to protect operator-trusted settings including sandbox policy, plugin enablement, gateway auth/TLS, hook routing, MCP server configuration, SSRF policy, and filesystem hardening. A prompt-injected model with access to the owner-only gateway tool can persist unauthorized changes to protected operator settings.","status":"published","published":"2026-05-11T18:16:40.237Z","modified":"2026-06-17T10:51:35.173Z","cvss":[{"version":"4.0","score":6,"vector":"CVSS:4.0/AV:N/AC:L/AT:P/PR:L/UI:N/VC:L/VI:H/VA:N/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":"medium","assessor":"secondary"},{"version":"3.1","score":7.1,"vector":"CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:L/I:H/A:N","level":"high","assessor":"secondary"}],"cwes":["CWE-862"],"updated_at":"2026-11-02T14:12:18.239974Z"}],"osv":[{"id":"CVE-2026-45001","title":"OpenClaw \u003c 2026.4.20 - Gateway Config Mutation Guard Bypass via Agent Tool Access","description":"OpenClaw before 2026.4.20 contains a guard bypass vulnerability in the agent-facing gateway config.patch and config.apply endpoints that fails to protect operator-trusted settings including sandbox policy, plugin enablement, gateway auth/TLS, hook routing, MCP server configuration, SSRF policy, and filesystem hardening. A prompt-injected model with access to the owner-only gateway tool can persist unauthorized changes to protected operator settings.","status":"published","published":"2026-05-11T16:46:39.888Z","modified":"2026-10-28T02:30:32.015589177Z","cvss":[{"version":"4.0","score":6,"vector":"CVSS:4.0/AV:N/AC:L/AT:P/PR:L/UI:N/VC:L/VI:H/VA:N/SC:N/SI:N/SA:N","level":"medium"}],"cwes":["CWE-862"],"updated_at":"2026-11-02T15:45:05.172514Z"},{"id":"GHSA-7jm2-g593-4qrc","title":"OpenClaw: Agent gateway config mutations could change protected operator settings","description":"## Affected Packages / Versions\n\n- Package: `openclaw` (npm)\n- Affected versions: `\u003c 2026.4.20`\n- Patched version: `2026.4.20`\n\n## Impact\n\nThe agent-facing `gateway config.patch` / `config.apply` guard did not cover several operator-trusted settings, including sandbox policy, plugin enablement, gateway auth/TLS, hook routing, MCP server configuration, SSRF policy, and filesystem hardening. A prompt-injected model with access to the owner-only gateway tool could persist changes to those settings.\n\nThis is a model-to-operator guard bypass, not a remote unauthenticated gateway compromise. Severity is medium.\n\n## Fix\n\nOpenClaw now blocks model-driven gateway config mutations for the broader operator-trusted path set and covers per-agent overrides and array-entry patching.\n\nFix commit:\n\n- `fe30b31a97a917ecc6e92f6c85378b6b20352422`\n\n## Release\n\nFixed in OpenClaw `2026.4.20`.","status":"published","published":"2026-04-25T23:51:11Z","modified":"2026-07-08T08:12:14.559661984Z","cvss":[{"version":"4.0","score":6,"vector":"CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:P/VC:N/VI:H/VA:L/SC:N/SI:N/SA:N","level":"medium"}],"cwes":["CWE-1220","CWE-285"],"updated_at":"2026-11-02T15:46:41.002603Z"}]}}