cve.bz

CVE-2026-48710 Kludex Starlette HTTP Request/Response Smuggling Vulnerability

6.5 mediumKEVExploit referenceEPSS 70%Fix available

Medium (6.5) in Kludex Starlette. In CISA KEV since 2026-09-02. EPSS 70% estimated probability of exploitation activity in the next 30 days. Fixed in starlette 1.0.1.

On this page
Starlette is a lightweight ASGI framework/toolkit. Prior to version 1.0.1, the HTTP `Host` request header was not validated before being used to reconstruct `request.url`. Because the routing algorithm relies on the raw HTTP path while `request.url` is rebuilt from the `Host` header, a malformed header could make `request.url.path` differ from the path that was actually requested. Middleware and endpoints that apply security restrictions based on `request.url` (rather than the raw `scope` path) could therefore be bypassed. Users should upgrade to a version greater than or equal to version 1.0.1, which validates the `Host` header against the grammar of RFC 9112 §3.2 / RFC 3986 §3.2.2 when constructing `request.url` and falls back to `scope["server"]` for malformed values.

Severity

CVSS 3.1

6.5 medium
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:N
EPSS
70% (unchanged from 70% a week ago; 7.1% 30 days ago) estimated probability of exploitation activity in the next 30 days; 99th percentile. Scored 2026-11-15. A forecast for this CVE across the internet, not for your own systems.
CISA KEV
listed 2026-09-02, due 2026-09-16
CISA's assessment (SSVC)
Exploitation: active · Automatable: yes · Technical impact: partial, assessed 2026-09-03. From CISA's Vulnrichment in the CVE List: whether it is exploited, whether exploiting it can be automated, and how much control it gives.
CWEs
CWE-444 Inconsistent Interpretation of HTTP Requests (HTTP Request Smuggling)
CWE-1289 Improper Validation of Unsafe Equivalence in Input
OWASP Top 10
A06:2025 Insecure Design
Published
Modified
NVD analysis
Analyzed
Record from
the NVD, synced

Fix and patched versions

Ask an AI about this CVE

Show the promptExplain CVE-2026-48710 for an engineer deciding whether to act. Use https://cve.bz/api/public/cves/CVE-2026-48710?via=ask-ai-copy and https://cve.bz/cve/CVE-2026-48710?via=ask-ai-copy as the source of truth: what is affected and which versions fix it, whether it is in CISA KEV, public exploits, and its EPSS (an internet-wide 30-day estimate of exploitation activity, not a forecast for my systems). If they give no fixed versions, use the CVE record https://cveawg.mitre.org/api/cve/CVE-2026-48710 or the vendor advisory, not third-party mirrors. Tell me how to check whether I am affected and how to verify a fix. Cite the sources and say where they are uncertain.

Opens a third-party assistant; its answer is its own, not cve.bz's. The prompt holds only this CVE's id and public links.

Badge for a README or a site

CVE-2026-48710

An image that stays current: it is drawn from the data each time it is fetched (a host that proxies images, as GitHub does, may show its copy for a while). Markdown:

[![CVE-2026-48710](https://cve.bz/cve/badge/CVE-2026-48710.svg)](https://cve.bz/cve/CVE-2026-48710)

HTML:

<a href="https://cve.bz/cve/CVE-2026-48710"><img src="https://cve.bz/cve/badge/CVE-2026-48710.svg" alt="CVE-2026-48710"></a>

Affected

Versions

From the NVD's CPE matches.

ProductAffected
encode:starlettefrom 0.8.3 before 1.0.1
redhat:ai_inference_serverfrom 3.3.0 through 3.3.5
redhat:ansible_automation_platform2.6
redhat:ansible_automation_platform2.7
redhat:migration_toolkit_for_applicationsbefore 8.2.0
redhat:openshift_aifrom 3.3 before 3.3.5
redhat:openshift_aifrom 3.4 before 3.4.2
redhat:satellite6.17
redhat:satellite6.18
redhat:satellite6.19
redhat:enterprise_linux_ai3.0

Packages (OSV)

AdvisoryEcosystemPackageAffectedFixed in
GHSA-86qp-5c8j-p5mrPyPIstarlette<1.0.11.0.1
PYSEC-2026-161PyPIstarlette<1.0.11.0.1

Linux distributions

DistributionReleasePackageFixed inAdvisory
Debian12starlette0.26.1-1+deb12u1DEBIAN-CVE-2026-48710
Debian13starlette0.46.1-3+deb13u2DEBIAN-CVE-2026-48710
Debian14starlette1.1.0-1DEBIAN-CVE-2026-48710
Ubuntu22.04:LTS, 24.04:LTS, 25.10 and 2 morestarletteno fix listedUBUNTU-CVE-2026-48710
SUSELinux Enterprise Server 16.0, Linux Enterprise Server for SAP applications 16.0python-starlette0.41.3-160000.3.1SUSE-SU-2026:22151-1
openSUSELeap 16.0python-starlette0.41.3-160000.3.1openSUSE-SU-2026:20975-1
openSUSETumbleweedpython-google-adk2.2.0-1.1openSUSE-SU-2026:11044-1
openSUSETumbleweedpython-starlette1.2.0-1.1openSUSE-SU-2026:11026-1
Alpinev3.24py3-starlette1.1.0-r0ALPINE-CVE-2026-48710

From the distributions' data via OSV.dev: the Debian Security Tracker; Ubuntu Security Notices © Canonical Ltd., CC BY-SA 4.0; SUSE security data, CC BY 4.0; the Alpine secdb, CC BY-SA 4.0. Modified: reformatted and linked to CVE records. Credits. On the distributions' own pages: Debian · Ubuntu · SUSE · openSUSE · Alpine.

Red Hat

Red Hat rates this Moderate.

Red Hat's statement

The impact of this vulnerability is rated Moderate. A malformed Host header can manipulate the reconstructed request.url.path, introducing a discrepancy with the canonical ASGI request path (scope["path"]). However, the vulnerability does not directly grant access, modify routing, or bypass security controls on its own. A path-based authentication or authorization bypass is possible only if an application depends on request.url.path for access control decisions rather than the canonical ASGI path or standard framework mechanisms. The impact relies heavily on downstream application design rather than a default library failure.

Red Hat Satellite / Insights (IoP): The starlette package is included as a dependency of Connexion, but this flaw is rated Low impact because these services do not use path-based authentication middleware and all services are deployed behind reverse proxies.

OpenShift Lightspeed (OLS): The starlette package is included as a dependency, but this flaw is rated Low impact because authentication is enforced via Kubernetes TokenReview/SubjectAccessReview through FastAPI dependency injection, not path-based middleware; request.url.path is only used for metrics and response header application.

RHOAI Kubeflow Training images: starlette is being used as indirect dependency of mlflow, but this flaw is rated as Low impact because mlflow is used as client library only to store training metrics on remote server, it doesn't expose any endpoint.

RHOAI odh-automl: The impact is reduced to Low since starlette is only used in the backend and it's transient dependency and starlette is not used at all to serve any endpoints.

RHOAI odh-kserve-autogluon-server-rhel9: While the Starlette dependency is present in the build, the vulnerable code in starlette is not reachable. This reduces the impact to low.

Red Hat's mitigation

Deploying an RFC-compliant reverse proxy (such as nginx, Apache, HAProxy, or Caddy) in front of the ASGI server will reject malformed Host headers before they reach the application. This is the most straightforward mitigation that does not require code changes.

If custom middleware is present, it should be updated to use `request.scope["path"]` instead of `request.url.path` for any security decisions. The ASGI scope path is derived from the HTTP request line and is not influenced by the Host header, so it reflects the actual request target.

ProductPackageStateAdvisory
Red Hat AI Inference Server 3.3rhaiis/vllm-rocm-rhel9:1782353093Fixed 2026-06-25RHSA-2026:30088
Red Hat AI Inference Server 3.3rhaiis/vllm-cuda-rhel9:1782352847Fixed 2026-06-25RHSA-2026:30089
Red Hat AI Inference Server 3.4rhaii/vllm-spyre-rhel9:1789681201Fixed 2026-09-21RHSA-2026:69464
Red Hat AI Inference Server 3.4rhaii/vllm-cpu-rhel9:1789681128Fixed 2026-09-21RHSA-2026:69466
Red Hat AI Inference Server 3.4rhaii/vllm-cuda-rhel9:1789681126Fixed 2026-09-21RHSA-2026:69467
Red Hat AI Inference Server 3.4rhaii/vllm-cpu-rhel9:1790075793Fixed 2026-09-23RHSA-2026:70965
Red Hat AI Inference Server 3.4rhaii/vllm-spyre-rhel9:1790076141Fixed 2026-09-23RHSA-2026:70969
Red Hat AI Inference Server 3.4rhaii/vllm-cuda-rhel9:1790090131Fixed 2026-09-23RHSA-2026:70979
Red Hat Ansible Automation Platform 2.6ansible-automation-platform-26/lightspeed-chatbot-rhel9:1780102732Fixed 2026-06-09RHSA-2026:24866
Red Hat Ansible Automation Platform 2.7ansible-automation-platform-27/lightspeed-chatbot-rhel9:1780099185Fixed 2026-06-04RHSA-2026:23346
Exploit Intelligenceexploit-intelligence-tech-preview/vulnerability-analysis-rhel9Not affected
OpenShift Lightspeedopenshift-lightspeed/lightspeed-ocp-rag-rhel9Not affected
OpenShift Lightspeedopenshift-lightspeed/lightspeed-service-api-rhel9Affected
Red Hat AI Inference Serverrhaiis/vllm-cpu-rhel9Will not fix
Red Hat AI Inference Serverrhaiis/vllm-neuron-rhel9Out of support scope
Red Hat AI Inference Serverrhaiis/vllm-spyre-rhel9Affected
Red Hat AI Inference Serverrhaiis/vllm-tpu-rhel9Affected
Red Hat AI Inference Serverrhaii/vllm-gaudi-rhel9Out of support scope
Red Hat AI Inference Serverrhaii/vllm-neuron-rhel9Out of support scope
Red Hat AI Inference Serverrhaii/vllm-rocm-rhel9Affected
71 more in the JSON and on Red Hat's page.

Contains Red Hat Security Data © Red Hat, Inc., licensed under CC BY 4.0. Source: https://access.redhat.com/security/data. Modified: reformatted. Red Hat's page for CVE-2026-48710.

GitHub PoC

Repositories on GitHub named after this CVE or listed for it by the PoC-in-GitHub index. For those marked Harbor Check: matches CVE, Flawharbor read the code and found it implements this CVE's exploitation technique: read, not run, and its dependencies only checked for warning signs, so a pointer and no guarantee. The rest are unverified: some hold a working proof of concept, some nothing useful, and some malware made to look like one. Read the code before you run any of it.

Detection rules

Ways to detect this CVE, not exploit code: the nuclei template that checks a system for it, and the Suricata and Snort rules that name it. A rule set may ship a rule turned off, so check yours by its sid.

Attack patterns (CAPEC)

  • CAPEC-33 HTTP Request Smuggling Likelihood: Medium · Severity: High
  • CAPEC-273 HTTP Response Smuggling Likelihood: Medium · Severity: High

The attack patterns MITRE's CAPEC relates to this record's weaknesses, as MITRE names them. Credits.

References

Check your stackJSON