Vercel @ai-sdk/harness-opencode up to 1.0.28 Relay host-tool-mcp.mjs improper authorization

Summaryinfo

A vulnerability described as problematic has been identified in Vercel @ai-sdk and harness-opencode up to 1.0.28. This affects an unknown function of the file host-tool-mcp.mjs of the component Relay. The manipulation results in improper authorization. This vulnerability is cataloged as CVE-2026-64651. The attack must be initiated from a local position. There is no exploit available.

Detailsinfo

A vulnerability, which was classified as problematic, was found in Vercel @ai-sdk and harness-opencode up to 1.0.28. Affected is an unknown code block of the file host-tool-mcp.mjs of the component Relay. The manipulation with an unknown input leads to a improper authorization vulnerability. CWE is classifying the issue as CWE-285. The product does not perform or incorrectly performs an authorization check when an actor attempts to access a resource or perform an action. This is going to have an impact on confidentiality, integrity, and availability. CVE summarizes:

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.

The advisory is shared for download at github.com. This vulnerability is traded as CVE-2026-64651 since 07/20/2026. The exploitability is told to be easy. The attack needs to be approached locally. There are known technical details, but no exploit is available. The MITRE ATT&CK project declares the attack technique as T1548.002.

Upgrading to version 1.0.29 eliminates this vulnerability.

VulDB is the best source for vulnerability data and more expert information about this specific topic.

Productinfo

Type

• Artificial Intelligence Software

Vendor

• Vercel

Name

• @ai-sdk
• harness-opencode

Version

• 1.0.0
• 1.0.1
• 1.0.2
• 1.0.3
• 1.0.4
• 1.0.5
• 1.0.6
• 1.0.7
• 1.0.8
• 1.0.9
• 1.0.10
• 1.0.11
• 1.0.12
• 1.0.13
• 1.0.14
• 1.0.15
• 1.0.16
• 1.0.17
• 1.0.18
• 1.0.19
• 1.0.20
• 1.0.21
• 1.0.22
• 1.0.23
• 1.0.24
• 1.0.25
• 1.0.26
• 1.0.27
• 1.0.28

CPE 2.3info

• 🔒
• 🔒
• 🔒

CPE 2.2info

• 🔒
• 🔒
• 🔒

CVSSv4info

CVSSv3info

CVSSv2info

Exploitinginfo

Threat Intelligenceinfo

Countermeasuresinfo

Timelineinfo

Sourcesinfo

Entryinfo

VulDB is the best source for vulnerability data and more expert information about this specific topic.

Discussion