EUVD-2026-84951
GHSA-jcjp-gr26-563j
Severity by source
CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:N/VC:H/VI:H/VA:N/SC:N/SI:N/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
AC:H because exploitation depends on the app minting tokens for 2+ audiences in one process and the attacker being positioned at Service B; C:H/I:H reflect token capture enabling full impersonation, no availability impact.
Primary rating from Vendor (Google).
CVSS VectorVendor: Google
Lifecycle Timeline
2DescriptionCVE.org
A caching flaw in the toolbox-core package of the mcp-toolbox-sdk-python SDK causes the same Google ID token to be cached and reused across different audiences. If an application uses the SDK to authenticate to two or more different audiences within the same process, the module-level token cache fails to key its cached tokens by the requested audience. Consequently, a valid, unexpired token minted for a sensitive service (Service A) can be retrieved from the cache and sent to a secondary service (Service B). An attacker who operates, compromises, or monitors traffic to Service B can capture this token and replay it to impersonate the victim application against Service A.
AnalysisAI
Cross-audience Google ID token reuse in the toolbox-core module of mcp-toolbox-sdk-python causes a credential intended for one service to be silently handed to a different service within the same process. Because the module-level token cache is not keyed by audience, an application that fetches ID tokens for two or more audiences in one process can send a Service A token to Service B; an attacker who operates, compromises, or passively monitors Service B can capture that token and replay it against Service A to impersonate the victim application. …
Unlock full vulnerability intelligence
- Risk assessment & exploitation conditions
- Attack chain visualization
- Remediation with exact patch versions
- Threat intelligence from 22 sources
- Personal watchlist & email alerts
No credit card · 7-day full trial
Attack ChainAIDerived
Hypothetical attack flow derived from CVE metadata
Vulnerability AssessmentAI
| Exploitation | Exploitation requires the victim application to use the toolbox-core module of mcp-toolbox-sdk-python to fetch Google ID tokens (via get_google_id_token / get_google_token_from_aud) for two or more different audiences within the SAME process, so that the shared module-level _token_cache serves a Service-A token for a Service-B request. … Additional conditions and limiting factors are described in the full assessment. |
| Risk Assessment | This is a genuine credential-leak flaw whose real-world priority is moderate and highly conditional despite the headline CVSS 4.0 base score of 9.1. … Full risk analysis with EPSS, KEV, and SSVC signal comparison available after sign-in. |
| Exploit Scenario | Full exploit scenario with step-by-step reproduction available after sign-in. |
| Remediation | Apply the vendor fix: upgrade toolbox-core to the mcp-toolbox-sdk-python release that includes the audience-keyed cache change from PR #675 (https://github.com/googleapis/mcp-toolbox-sdk-python/pull/675). … Detailed patch versions, workarounds, and compensating controls in full report. |
Recommended ActionAI
Within 24 hours, inventory all deployments of mcp-toolbox-sdk-python/toolbox-core that obtain Google ID tokens for more than one audience in a single process, isolate or temporarily disable multi-audience token reuse paths, rotate any Google ID tokens or service-account credentials that could have crossed audience boundaries, and review identity/API logs for unexpected token use against Service A; within 7 days, apply the patch available per vendor advisory for mcp-toolbox-sdk-python/toolbox-core, verify the exact fixed release in the advisory, deploy to staging before production, and confirm the token cache is keyed by audience with regression tests for cross-audience separation; within 30 days, complete production remediation of all affected services, validate no remaining multi-audience cache paths, add dependency scanning and CI checks for this SDK, restrict network/identity visibility so Service B traffic cannot be passively monitored by untrusted parties, and update secure design standards for audience-scoped credential caching.
Sign in for detailed remediation steps and compensating controls.
Threat intelligence, references, and detailed analysis are available after sign-in.
Wazuh SIEM platform versions 4.4.0 through 4.9.0 contain an unsafe deserialization vulnerability in the DistributedAPI t
BentoML version 1.4.2 and earlier contains an unauthenticated remote code execution vulnerability through insecure deser
pgAdmin 4 contains critical remote code execution vulnerabilities in the Query Tool download and Cloud Deployment endpoi
The renderLocalView function in render/views.py in graphite-web in Graphite 0.9.5 through 0.9.10 uses the pickle Python
BentoML is a Python library for building online serving systems optimized for AI apps and model inference. Rated critica
OpenSSL before 0.9.8za, 1.0.0 before 1.0.0m, and 1.0.1 before 1.0.1h does not properly restrict processing of ChangeCiph
pyLoad download manager version prior to 0.5.0b3.dev77 exposes the Flask SECRET_KEY through an unauthenticated endpoint.
Langflow (a visual LLM pipeline builder) contains a critical unauthenticated code execution vulnerability (CVE-2026-3301
In Mercurial before 4.1.3, "hg serve --stdio" allows remote authenticated users to launch the Python debugger, and conse
Unauthenticated remote code execution affects Kestra OSS (the open-source event-driven orchestration platform) prior to
Unauthenticated remote code execution in Marimo ≤0.20.4 allows attackers to execute arbitrary system commands via the `/
pyLoad is the free and open-source Download Manager written in pure Python. Rated medium severity (CVSS 5.3), this vulne
Same technique Information Disclosure
View allShare
External POC / Exploit Code
Leaving vuln.today