Unauthorized host tool invocation via process-path fallback
Published Jul 20, 2026 · Updated Jul 20, 2026
Authorization bypass in Vercel @ai-sdk/harness-opencode 1.0.27 and earlier allows local users to invoke host-exposed tools. The tool relay treats any Linux process whose command line contains the host-tool-mcp.mjs helper path as authorized, without matching a model-authorized tool event. An active harness session with exposed host tools and untrusted sandbox code is required, after which the code can perform secret lookups, deployments, or cloud API calls.
Summary
What happened
Authorization bypass in Vercel @ai-sdk/harness-opencode 1.0.27 and earlier allows local users to invoke host-exposed tools. The tool relay treats any Linux process whose command line contains the host-tool-mcp.mjs helper path as authorized, without matching a model-authorized tool event. An active harness session with exposed host tools and untrusted sandbox code is required, after which the code can perform secret lookups, deployments, or cloud API calls.
The record
- CVE
- CVE-2026-64651
- Published
- Jul 20, 2026
- Updated
- Jul 20, 2026
- Vendor
- Vercel
- Product
- @ai-sdk/harness-opencode
- Classifications
- T1068
- Attack vector
- local
- Privileges
- unauthenticated
Timeline
How it unfolded
- Jul 20, 2026CVE publishedPublication date reported by the CVE source.
- Jul 20, 2026Record updatedLatest update available in the CVE record.
Exploitability
Present is not the same as exploitable
Compare your product and version with the public record. A matching version still requires validation against your environment.
Is a vulnerable build present?
Compare these published version ranges with your installed build and any vendor patches.
- Affected versionversion=< 1.0.29
What conditions does exploitation require?
What is affected?
Published CVSS scores
CVSS describes severity. EPSS estimates exploitation probability.
Attacks
What attackers are doing with it
Daily unique IPs observed by Shadowserver honeypots for known exploited vulnerabilities (KEVs). Missing observations do not establish an absence of attacks.
Weakness, pattern, technique
Public exploit references
No public exploit references are available in this record.
Labels summarize the accepted research assessment. They do not indicate a test against your environment.
Technologies
Your stack
See the directory against your own environment.
Your stack
Check the software in your environment
Book a demo to see how Hinoki identifies affected software and validates exploitability in your environment.
Book a demo