Remote code execution via bypassed browser-request guard
Published Nov 26, 2025 · Updated Aug 18, 2026
Code injection in Anyscale Ray before 2.52.0 allows remote attackers to execute arbitrary code through browser-based DNS rebinding. The dashboard middleware treats only User-Agent values starting with Mozilla as browser traffic, while Firefox and Safari fetch requests can override that header and reach the unauthenticated Jobs API. A developer must run the dashboard and visit attacker-controlled content; successful requests submit a job entrypoint that runs with Ray's host privileges.
Summary
What happened
Code injection in Anyscale Ray before 2.52.0 allows remote attackers to execute arbitrary code through browser-based DNS rebinding. The dashboard middleware treats only User-Agent values starting with Mozilla as browser traffic, while Firefox and Safari fetch requests can override that header and reach the unauthenticated Jobs API. A developer must run the dashboard and visit attacker-controlled content; successful requests submit a job entrypoint that runs with Ray's host privileges.
The record
- CVE
- CVE-2025-62593
- Published
- Nov 26, 2025
- Updated
- Aug 18, 2026
- Vendor
- Anyscale
- Product
- Ray
- Classifications
- CWE-352, CWE-94, T1189
- Attack vector
- network
- Privileges
- unauthenticated
Timeline
How it unfolded
- Nov 26, 2025CVE publishedPublication date reported by the CVE source.
- Aug 17, 2026Exploitation reportedCISA Known Exploited Vulnerabilities listing
- Aug 18, 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=< 2.52.0
What conditions does exploitation require?
What is affected?
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
- Boreas37 Ray RCE proof of conceptfunctional · validated
- Singularity Ray Jobs RCE payloadfunctional · demonstrated
Labels summarize the accepted research assessment. They do not indicate a test against your environment.
Reported exploitation
- CISA Known Exploited Vulnerabilities listingknown exploited catalog
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