{"id":"CVE-2026-67531","aliases":["EUVD-2026-53684","GHSA-mp29-fxh8-92px"],"status":"published","title":"FrontMCP: CodeCall sandbox escape -\u003e host RCE via live Zod schema exposure by getTool","description":"FrontMCP is a TypeScript-first framework for the Model Context Protocol (MCP). Prior to 1.5.7, the sandboxed codecall:execute tool exposes live host Zod schema instances to the script via getTool(), and because Zod v4 defines _zod as a non-configurable, non-writable own property, the ECMAScript Proxy invariants force the security membrane to hand back the raw host object, letting a script reach _zod.constr.constructor (the host Function constructor) and execute arbitrary code in the server process. A single tools/call is sufficient to escape the sandbox and achieve remote code execution as the server user, exposing everything the process holds such as OAuth client secrets, JWT_SECRET, session keys, database credentials, and cloud instance metadata. Because the framework's DEFAULT_AUTH_OPTIONS is public mode, an unconfigured server serves this to unauthenticated callers, and on authenticated servers an indirect prompt injection in tool output or fetched content can trigger it without a human attackerThis issue is fixed in version 1.5.7.","severity":{"version":"4.0","score":9.3,"vector":"CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N","level":"critical","assessor":"cna","from":"cve"},"epss":{"day":"2026-11-12","score":0.00735,"percentile":0.54923,"model":"v2026.06.15"},"facts":{"in_kev":false,"known_exploited":false,"public_exploit":false,"fix_available":false},"cwes":["CWE-94"],"published":"2026-08-05T22:54:36Z","modified":"2026-10-28T02:30:39Z","products":[{"vendor":"agentfront","product":"frontmcp"}],"references":[{"url":"https://github.com/agentfront/frontmcp/security/advisories/GHSA-mp29-fxh8-92px","tags":["x_refsource_CONFIRM","exploit","advisory"]},{"url":"https://github.com/agentfront/frontmcp/commit/209cddd19a8d4db0777f725b527818da7df6f67f","tags":["x_refsource_MISC","fix"]},{"url":"https://github.com/CVEProject/cvelistV5/tree/main/cves/2026/67xxx/CVE-2026-67531.json","tags":["advisory"]},{"url":"https://nvd.nist.gov/vuln/detail/CVE-2026-67531","tags":["advisory"]}],"sources":{"cve":[{"id":"CVE-2026-67531","title":"FrontMCP: CodeCall sandbox escape -\u003e host RCE via live Zod schema exposure by getTool","description":"FrontMCP is a TypeScript-first framework for the Model Context Protocol (MCP). Prior to 1.5.7, the sandboxed codecall:execute tool exposes live host Zod schema instances to the script via getTool(), and because Zod v4 defines _zod as a non-configurable, non-writable own property, the ECMAScript Proxy invariants force the security membrane to hand back the raw host object, letting a script reach _zod.constr.constructor (the host Function constructor) and execute arbitrary code in the server process. A single tools/call is sufficient to escape the sandbox and achieve remote code execution as the server user, exposing everything the process holds such as OAuth client secrets, JWT_SECRET, session keys, database credentials, and cloud instance metadata. Because the framework's DEFAULT_AUTH_OPTIONS is public mode, an unconfigured server serves this to unauthenticated callers, and on authenticated servers an indirect prompt injection in tool output or fetched content can trigger it without a human attackerThis issue is fixed in version 1.5.7.","status":"published","published":"2026-08-05T22:54:36.219Z","modified":"2026-08-06T18:44:39.455Z","cvss":[{"version":"4.0","score":9.3,"vector":"CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N","level":"critical","assessor":"cna"}],"cwes":["CWE-94"],"updated_at":"2026-11-02T14:20:06.562031Z"}],"euvd":[{"id":"EUVD-2026-53684","description":"FrontMCP is a TypeScript-first framework for the Model Context Protocol (MCP). Prior to 1.5.7, the sandboxed codecall:execute tool exposes live host Zod schema instances to the script via getTool(), and because Zod v4 defines _zod as a non-configurable, non-writable own property, the ECMAScript Proxy invariants force the security membrane to hand back the raw host object, letting a script reach _zod.constr.constructor (the host Function constructor) and execute arbitrary code in the server process. A single tools/call is sufficient to escape the sandbox and achieve remote code execution as the server user, exposing everything the process holds such as OAuth client secrets, JWT_SECRET, session keys, database credentials, and cloud instance metadata. Because the framework's DEFAULT_AUTH_OPTIONS is public mode, an unconfigured server serves this to unauthenticated callers, and on authenticated servers an indirect prompt injection in tool output or fetched content can trigger it without a human attackerThis issue is fixed in version 1.5.7.","status":"published","published":"2026-08-05T22:54:36Z","modified":"2026-08-06T18:44:39Z","cvss":[{"version":"4.0","score":9.3,"vector":"CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N","level":"critical"}],"updated_at":"2026-11-02T18:02:07.757838Z"}],"nvd":[{"id":"CVE-2026-67531","description":"FrontMCP is a TypeScript-first framework for the Model Context Protocol (MCP). Prior to 1.5.7, the sandboxed codecall:execute tool exposes live host Zod schema instances to the script via getTool(), and because Zod v4 defines _zod as a non-configurable, non-writable own property, the ECMAScript Proxy invariants force the security membrane to hand back the raw host object, letting a script reach _zod.constr.constructor (the host Function constructor) and execute arbitrary code in the server process. A single tools/call is sufficient to escape the sandbox and achieve remote code execution as the server user, exposing everything the process holds such as OAuth client secrets, JWT_SECRET, session keys, database credentials, and cloud instance metadata. Because the framework's DEFAULT_AUTH_OPTIONS is public mode, an unconfigured server serves this to unauthenticated callers, and on authenticated servers an indirect prompt injection in tool output or fetched content can trigger it without a human attackerThis issue is fixed in version 1.5.7.","status":"published","published":"2026-08-06T00:16:53.733Z","modified":"2026-10-24T16:10:00.153Z","cvss":[{"version":"4.0","score":9.3,"vector":"CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/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":"critical","assessor":"secondary"}],"cwes":["CWE-94"],"updated_at":"2026-11-02T14:12:35.031043Z"}],"osv":[{"id":"CVE-2026-67531","title":"FrontMCP: CodeCall sandbox escape -\u003e host RCE via live Zod schema exposure by getTool","description":"FrontMCP is a TypeScript-first framework for the Model Context Protocol (MCP). Prior to 1.5.7, the sandboxed codecall:execute tool exposes live host Zod schema instances to the script via getTool(), and because Zod v4 defines _zod as a non-configurable, non-writable own property, the ECMAScript Proxy invariants force the security membrane to hand back the raw host object, letting a script reach _zod.constr.constructor (the host Function constructor) and execute arbitrary code in the server process. A single tools/call is sufficient to escape the sandbox and achieve remote code execution as the server user, exposing everything the process holds such as OAuth client secrets, JWT_SECRET, session keys, database credentials, and cloud instance metadata. Because the framework's DEFAULT_AUTH_OPTIONS is public mode, an unconfigured server serves this to unauthenticated callers, and on authenticated servers an indirect prompt injection in tool output or fetched content can trigger it without a human attackerThis issue is fixed in version 1.5.7.","status":"published","published":"2026-08-05T22:54:36.219Z","modified":"2026-10-28T02:30:39.983019255Z","cvss":[{"version":"4.0","score":9.3,"vector":"CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N","level":"critical"}],"cwes":["CWE-94"],"updated_at":"2026-11-02T15:45:14.46177Z"}]}}