{"dataType": "CVE_RECORD", "dataVersion": "5.2", "cveMetadata": {"cveId": "CVE-2026-64651", "assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa", "state": "PUBLISHED", "assignerShortName": "GitHub_M", "dateReserved": "2026-07-20T17:11:30.896Z", "datePublished": "2026-07-20T20:26:07.497Z", "dateUpdated": "2026-07-20T20:26:07.497Z"}, "containers": {"cna": {"title": "AI SDK OpenCode Harness Tool Relay Authorization Bypass", "problemTypes": [{"descriptions": [{"cweId": "CWE-863", "lang": "en", "description": "CWE-863: Incorrect Authorization", "type": "CWE"}]}], "metrics": [{"cvssV4_0": {"attackVector": "LOCAL", "attackComplexity": "LOW", "attackRequirements": "NONE", "privilegesRequired": "NONE", "userInteraction": "NONE", "vulnConfidentialityImpact": "NONE", "vulnIntegrityImpact": "LOW", "vulnAvailabilityImpact": "NONE", "subConfidentialityImpact": "HIGH", "subIntegrityImpact": "HIGH", "subAvailabilityImpact": "NONE", "baseScore": 6.3, "baseSeverity": "MEDIUM", "vectorString": "CVSS:4.0/AV:L/AC:L/AT:N/PR:N/UI:N/VC:N/VI:L/VA:N/SC:H/SI:H/SA:N", "version": "4.0"}}], "references": [{"name": "https://github.com/vercel/ai/security/advisories/GHSA-g48p-5rr5-8rgq", "tags": ["x_refsource_CONFIRM"], "url": "https://github.com/vercel/ai/security/advisories/GHSA-g48p-5rr5-8rgq"}, {"name": "https://github.com/vercel/ai/pull/17105", "tags": ["x_refsource_MISC"], "url": "https://github.com/vercel/ai/pull/17105"}], "affected": [{"vendor": "vercel", "product": "@ai-sdk/harness-opencode", "versions": [{"version": "< 1.0.29", "status": "affected"}]}], "providerMetadata": {"orgId": "a0819718-46f1-4df5-94e2-005712e83aaa", "shortName": "GitHub_M", "dateUpdated": "2026-07-20T20:26:07.497Z"}, "descriptions": [{"lang": "en", "value": "The `@ai-sdk/harness-opencode` tool connects HarnessAgent to OpenCode through a sandboxed bridge. Prior to version 1.0.28, the tool relay authorizes requests from any process whose command line contains an allowed helper script path (`host-tool-mcp.mjs`). This allows untrusted code executing in the sandbox to invoke arbitrary host-exposed tools including secret lookups, deployment operations, and cloud API calls without a corresponding model-authorized tool-call event. Exploitation requires a Linux environment (the vulnerable fallback checks `process.platform === 'linux'` and reads `/proc`); an active harness session with one or more host-provided tools; and untrusted code executing in the sandbox (e.g. a malicious dependency, build script, or lifecycle hook) The fix in version 1.0.28 removes the process-path authorization fallback entirely. Relay requests are now only accepted after exact, short-lived, one-time authorization matching the tool name and input from a bridge-observed model event. Some workarounds are available. Do not run the OpenCode harness on untrusted repositories or with untrusted dependencies, and/or limit host-exposed tools to non-sensitive operations when working with untrusted code."}], "source": {"advisory": "GHSA-g48p-5rr5-8rgq", "discovery": "UNKNOWN"}}}}

{}

{}

{}

{}