SGLang contains an SSRF and local file read in the multimodal generation endpoint /v1/chat/completions due to unsanitized image_url, allowing access… (CVE-2026-15974)
SGLang has a vulnerability in its multimodal generation endpoint /v1/chat/completions where an unsanitized image_url parameter allows server-side request forgery (SSRF) and local file reading. This flaw enables attackers to access internal metadata, secrets, and internal services. No affected versions or patch information is provided, and no known exploits are reported in the wild.
AI Analysis
Technical Summary
The vulnerability in SGLang's /v1/chat/completions endpoint arises from improper sanitization of the image_url parameter, leading to SSRF and local file read capabilities. This allows an attacker to make unauthorized requests to internal resources and read sensitive files, potentially exposing internal metadata and secrets. There is no CVSS score or detailed vendor advisory available, and no patch or remediation information is provided. The service is not cloud-hosted, so remediation responsibility lies with the user or vendor.
Potential Impact
An attacker exploiting this vulnerability could access internal metadata, secrets, and internal services that should not be exposed externally. This can lead to information disclosure and potentially further compromise of the affected system. No evidence of active exploitation is currently known.
Mitigation Recommendations
Patch status is not yet confirmed — check the vendor advisory for current remediation guidance. Until a fix is available, users should consider restricting access to the vulnerable endpoint, validating and sanitizing input parameters such as image_url, and monitoring for suspicious activity related to SSRF attempts.
SGLang contains an SSRF and local file read in the multimodal generation endpoint /v1/chat/completions due to unsanitized image_url, allowing access… (CVE-2026-15974)
Description
SGLang has a vulnerability in its multimodal generation endpoint /v1/chat/completions where an unsanitized image_url parameter allows server-side request forgery (SSRF) and local file reading. This flaw enables attackers to access internal metadata, secrets, and internal services. No affected versions or patch information is provided, and no known exploits are reported in the wild.
AI-Powered Analysis
Machine-generated threat intelligence
Technical Analysis
The vulnerability in SGLang's /v1/chat/completions endpoint arises from improper sanitization of the image_url parameter, leading to SSRF and local file read capabilities. This allows an attacker to make unauthorized requests to internal resources and read sensitive files, potentially exposing internal metadata and secrets. There is no CVSS score or detailed vendor advisory available, and no patch or remediation information is provided. The service is not cloud-hosted, so remediation responsibility lies with the user or vendor.
Potential Impact
An attacker exploiting this vulnerability could access internal metadata, secrets, and internal services that should not be exposed externally. This can lead to information disclosure and potentially further compromise of the affected system. No evidence of active exploitation is currently known.
Mitigation Recommendations
Patch status is not yet confirmed — check the vendor advisory for current remediation guidance. Until a fix is available, users should consider restricting access to the vulnerable endpoint, validating and sanitizing input parameters such as image_url, and monitoring for suspicious activity related to SSRF attempts.
Technical Details
- Gcve Source
- db.gcve.eu
- Osv Id
- GHSA-hg26-hr8r-fxc9
- Osv Schema Version
- 1.4.0
- Aliases
- ["CVE-2026-15974"]
- Ecosystems
- []
- Database Specific Severity
- null
- Cvss Version
- null
Threat ID: 6a6bdd8b9c2644c7f8da72e6
Added to database: 07/30/2026, 23:26:03 UTC
Last enriched: 07/30/2026, 23:33:40 UTC
Last updated: 07/31/2026, 01:31:19 UTC
Views: 4
Community Reviews
0 reviewsCrowdsource mitigation strategies, share intel context, and vote on the most helpful responses. Sign in to add your voice and help keep defenders ahead.
Want to contribute mitigation steps or threat intel context? Sign in or create an account to join the community discussion.
Actions
Updates to AI analysis require Pro Console access. Upgrade inside Console → Billing.
Need more coverage?
Upgrade to Pro Console for AI refresh and higher limits.
For incident response and remediation, OffSeq services can help resolve threats faster.
Latest Threats
Check if your credentials are on the dark web
Instant breach scanning across billions of leaked records. Free tier available.