Logo
-

Byte Open Security

(ByteOS Network)

Log In

Sign Up

ByteOS

Security
Vulnerability Details
Registries
Custom Views
Weaknesses
Attack Patterns
Filters & Tools
Vulnerability Details :

CVE-2026-44428

Summary
Assigner-GitHub_M
Assigner Org ID-a0819718-46f1-4df5-94e2-005712e83aaa
Published At-14 May, 2026 | 21:09
Updated At-15 May, 2026 | 14:22
Rejected At-
Credits

MCP Registry: GitHub OIDC tokens replayable across registry deployments due to shared audience

The MCP Registry provides MCP clients with a list of MCP servers, like an app store for MCP servers. Prior to 1.7.6, the client-side and server-side GitHub OIDC flow is bound only to a global audience string, not to the specific registry instance being targeted. On the client side, the publisher always appends audience=mcp-registry when requesting the GitHub Actions ID token, regardless of the selected --registry URL. On the server side, the exchange endpoint validates only that same fixed audience and then derives publish permissions directly from repository_owner. As a result, a token legitimately obtained while interacting with one registry deployment remains acceptable to any other deployment that shares the same code and audience string. This vulnerability is fixed in 1.7.6.

Vendors
-
Not available
Products
-
Metrics (CVSS)
VersionBase scoreBase severityVector
Weaknesses
Attack Patterns
Solution/Workaround
References
HyperlinkResource Type
EPSS History
Score
Latest Score
-
N/A
No data available for selected date range
Percentile
Latest Percentile
-
N/A
No data available for selected date range
Stakeholder-Specific Vulnerability Categorization (SSVC)
▼Common Vulnerabilities and Exposures (CVE)
cve.org
Assigner:GitHub_M
Assigner Org ID:a0819718-46f1-4df5-94e2-005712e83aaa
Published At:14 May, 2026 | 21:09
Updated At:15 May, 2026 | 14:22
Rejected At:
▼CVE Numbering Authority (CNA)
MCP Registry: GitHub OIDC tokens replayable across registry deployments due to shared audience

The MCP Registry provides MCP clients with a list of MCP servers, like an app store for MCP servers. Prior to 1.7.6, the client-side and server-side GitHub OIDC flow is bound only to a global audience string, not to the specific registry instance being targeted. On the client side, the publisher always appends audience=mcp-registry when requesting the GitHub Actions ID token, regardless of the selected --registry URL. On the server side, the exchange endpoint validates only that same fixed audience and then derives publish permissions directly from repository_owner. As a result, a token legitimately obtained while interacting with one registry deployment remains acceptable to any other deployment that shares the same code and audience string. This vulnerability is fixed in 1.7.6.

Affected Products
Vendor
modelcontextprotocol
Product
registry
Versions
Affected
  • < 1.7.6
Problem Types
TypeCWE IDDescription
CWECWE-918CWE-918: Server-Side Request Forgery (SSRF)
Type: CWE
CWE ID: CWE-918
Description: CWE-918: Server-Side Request Forgery (SSRF)
Metrics
VersionBase scoreBase severityVector
4.02.1LOW
CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:A/VC:N/VI:L/VA:N/SC:N/SI:L/SA:N
Version: 4.0
Base score: 2.1
Base severity: LOW
Vector:
CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:A/VC:N/VI:L/VA:N/SC:N/SI:L/SA:N
Metrics Other Info
Impacts
CAPEC IDDescription
Solutions
Configurations
Workarounds
Exploits
Credits
Timeline
EventDate
Replaced By
Rejected Reason
References
HyperlinkResource
https://github.com/modelcontextprotocol/registry/security/advisories/GHSA-95c3-6vvw-4mrq
x_refsource_CONFIRM
Hyperlink: https://github.com/modelcontextprotocol/registry/security/advisories/GHSA-95c3-6vvw-4mrq
Resource:
x_refsource_CONFIRM
▼Authorized Data Publishers (ADP)
CISA ADP Vulnrichment
Affected Products
Metrics
VersionBase scoreBase severityVector
Metrics Other Info
Impacts
CAPEC IDDescription
Solutions
Configurations
Workarounds
Exploits
Credits
Timeline
EventDate
Replaced By
Rejected Reason
References
HyperlinkResource
Information is not available yet
▼National Vulnerability Database (NVD)
nvd.nist.gov
Source:security-advisories@github.com
Published At:14 May, 2026 | 22:16
Updated At:15 May, 2026 | 17:23

The MCP Registry provides MCP clients with a list of MCP servers, like an app store for MCP servers. Prior to 1.7.6, the client-side and server-side GitHub OIDC flow is bound only to a global audience string, not to the specific registry instance being targeted. On the client side, the publisher always appends audience=mcp-registry when requesting the GitHub Actions ID token, regardless of the selected --registry URL. On the server side, the exchange endpoint validates only that same fixed audience and then derives publish permissions directly from repository_owner. As a result, a token legitimately obtained while interacting with one registry deployment remains acceptable to any other deployment that shares the same code and audience string. This vulnerability is fixed in 1.7.6.

CISA Catalog
Date AddedDue DateVulnerability NameRequired Action
N/A
Date Added: N/A
Due Date: N/A
Vulnerability Name: N/A
Required Action: N/A
Metrics
TypeVersionBase scoreBase severityVector
Secondary4.02.1LOW
CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:A/VC:N/VI:L/VA:N/SC:N/SI:L/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X
Primary3.14.7MEDIUM
CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:N/I:L/A:N
Type: Secondary
Version: 4.0
Base score: 2.1
Base severity: LOW
Vector:
CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:A/VC:N/VI:L/VA:N/SC:N/SI:L/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X
Type: Primary
Version: 3.1
Base score: 4.7
Base severity: MEDIUM
Vector:
CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:N/I:L/A:N
CPE Matches
lfprojects
lfprojects
>>mcp_registry>>Versions before 1.7.6(exclusive)
cpe:2.3:a:lfprojects:mcp_registry:*:*:*:*:*:*:*:*
Weaknesses
CWE IDTypeSource
CWE-918Primarysecurity-advisories@github.com
CWE ID: CWE-918
Type: Primary
Source: security-advisories@github.com
Evaluator Description
Evaluator Impact

Evaluator Solution

Vendor Statements
References
HyperlinkSourceResource
https://github.com/modelcontextprotocol/registry/security/advisories/GHSA-95c3-6vvw-4mrqsecurity-advisories@github.com
Mitigation
Vendor Advisory
Hyperlink: https://github.com/modelcontextprotocol/registry/security/advisories/GHSA-95c3-6vvw-4mrq
Source: security-advisories@github.com
Resource:
Mitigation
Vendor Advisory

Change History

0
Information is not available yet

Similar CVEs

5Records found
CVE-2026-44430
Matching Score-6
Assigner-GitHub, Inc.
ShareView Details
Matching Score-6
Assigner-GitHub, Inc.
CVSS Score-6.3||MEDIUM
EPSS-0.04% / 12.50%
||
7 Day CHG~0.00%
Published-14 May, 2026 | 21:02
Updated-15 May, 2026 | 19:16
Rejected-Not Available
Known To Be Used In Ransomware Campaigns?-Not Available
KEV Added-Not Available
KEV Action Due Date-Not Available
MCP Registry: Unauthenticated SSRF: HTTP namespace verification dials 6to4 / NAT64 / site-local IPv6 addresses, bypassing private-address allowlist

The MCP Registry provides MCP clients with a list of MCP servers, like an app store for MCP servers. Prior to 1.7.7, the Registry's HTTP-based namespace verification (POST /v0/auth/http, POST /v0.1/auth/http) uses safeDialContext (internal/api/handlers/v0/auth/http.go:67-110) to refuse dialling private/internal addresses when fetching the well-known public-key file from a publisher-supplied domain. The blocklist (isBlockedIP, lines 125-133) relies entirely on Go stdlib's IsLoopback / IsPrivate / IsLinkLocalUnicast / IsMulticast / IsUnspecified plus a manual CGNAT range. None of these cover IPv6 6to4 (2002::/16), NAT64 (64:ff9b::/96 and 64:ff9b:1::/48 per RFC 8215), or deprecated site-local (fec0::/10) — all of which encode arbitrary IPv4 in the address bits and tunnel to RFC1918 / cloud-metadata services on dual-stack / NAT64-enabled hosts. This vulnerability is fixed in 1.7.7.

Action-Not Available
Vendor-lfprojectsmodelcontextprotocol
Product-mcp_registryregistry
CWE ID-CWE-918
Server-Side Request Forgery (SSRF)
CVE-2025-52967
Matching Score-6
Assigner-MITRE Corporation
ShareView Details
Matching Score-6
Assigner-MITRE Corporation
CVSS Score-5.8||MEDIUM
EPSS-0.25% / 47.99%
||
7 Day CHG~0.00%
Published-23 Jun, 2025 | 00:00
Updated-23 Jun, 2025 | 20:16
Rejected-Not Available
Known To Be Used In Ransomware Campaigns?-Not Available
KEV Added-Not Available
KEV Action Due Date-Not Available

gateway_proxy_handler in MLflow before 3.1.0 lacks gateway_path validation.

Action-Not Available
Vendor-lfprojects
Product-MLflow
CWE ID-CWE-918
Server-Side Request Forgery (SSRF)
CVE-2023-6974
Matching Score-6
Assigner-Protect AI (formerly huntr.dev)
ShareView Details
Matching Score-6
Assigner-Protect AI (formerly huntr.dev)
CVSS Score-8.6||HIGH
EPSS-2.59% / 85.76%
||
7 Day CHG~0.00%
Published-20 Dec, 2023 | 05:25
Updated-02 Aug, 2024 | 08:50
Rejected-Not Available
Known To Be Used In Ransomware Campaigns?-Not Available
KEV Added-Not Available
KEV Action Due Date-Not Available
Server-Side Request Forgery (SSRF)

A malicious user could use this issue to access internal HTTP(s) servers and in the worst case (ie: aws instance) it could be abuse to get a remote code execution on the victim machine.

Action-Not Available
Vendor-lfprojectsmlflow
Product-mlflowmlflow/mlflow
CWE ID-CWE-918
Server-Side Request Forgery (SSRF)
CVE-2021-41084
Matching Score-4
Assigner-GitHub, Inc.
ShareView Details
Matching Score-4
Assigner-GitHub, Inc.
CVSS Score-8.7||HIGH
EPSS-0.45% / 63.88%
||
7 Day CHG~0.00%
Published-21 Sep, 2021 | 17:20
Updated-04 Aug, 2024 | 02:59
Rejected-Not Available
Known To Be Used In Ransomware Campaigns?-Not Available
KEV Added-Not Available
KEV Action Due Date-Not Available
Response Splitting from unsanitized headers in http4s

http4s is an open source scala interface for HTTP. In affected versions http4s is vulnerable to response-splitting or request-splitting attacks when untrusted user input is used to create any of the following fields: Header names (`Header.name`å), Header values (`Header.value`), Status reason phrases (`Status.reason`), URI paths (`Uri.Path`), URI authority registered names (`URI.RegName`) (through 0.21). This issue has been resolved in versions 0.21.30, 0.22.5, 0.23.4, and 1.0.0-M27 perform the following. As a matter of practice http4s services and client applications should sanitize any user input in the aforementioned fields before returning a request or response to the backend. The carriage return, newline, and null characters are the most threatening.

Action-Not Available
Vendor-typelevelhttp4s
Product-http4shttp4s
CWE ID-CWE-918
Server-Side Request Forgery (SSRF)
CWE ID-CWE-74
Improper Neutralization of Special Elements in Output Used by a Downstream Component ('Injection')
CVE-2025-8675
Matching Score-4
Assigner-Drupal.org
ShareView Details
Matching Score-4
Assigner-Drupal.org
CVSS Score-4.7||MEDIUM
EPSS-0.05% / 15.70%
||
7 Day CHG~0.00%
Published-15 Aug, 2025 | 16:27
Updated-21 Aug, 2025 | 19:53
Rejected-Not Available
Known To Be Used In Ransomware Campaigns?-Not Available
KEV Added-Not Available
KEV Action Due Date-Not Available
AI SEO Link Advisor - Less critical - Server-side Request Forgery - SA-CONTRIB-2025-095

Server-Side Request Forgery (SSRF) vulnerability in Drupal AI SEO Link Advisor allows Server Side Request Forgery.This issue affects AI SEO Link Advisor: from 0.0.0 before 1.0.6.

Action-Not Available
Vendor-ai_seo_link_advisor_projectThe Drupal Association
Product-ai_seo_link_advisorAI SEO Link Advisor
CWE ID-CWE-918
Server-Side Request Forgery (SSRF)
Details not found