CVE-2026-48817: CWE-470: Use of Externally-Controlled Input to Select Classes or Code ('Unsafe Reflection') in Kludex starlette
Starlette versions prior to 1.1.0 have a vulnerability where HTTPEndpoint dispatches requests by dynamically invoking methods based on the HTTP method name without restricting to standard HTTP verbs. This allows an attacker to invoke internal methods named after non-standard HTTP verbs if the route is registered without an explicit methods= argument. This can lead to unauthorized invocation of internal helper methods that bypass authorization checks. The issue is fixed in version 1.1.0.
AI Analysis
Technical Summary
In Starlette versions 1.0.1 and below, the HTTPEndpoint class dispatches requests by lowercasing the HTTP method and using getattr to find a handler method. However, if a route is registered without specifying the methods= argument, all HTTP methods are accepted, including non-standard ones. If a non-standard HTTP method matches an attribute name on the endpoint subclass, that attribute is invoked as a handler, potentially exposing internal methods not intended to be HTTP handlers. This unsafe reflection can lead to unauthorized access to internal functionality. The vulnerability is addressed in Starlette 1.1.0 by restricting method dispatching.
Potential Impact
An attacker can invoke internal methods of an HTTPEndpoint subclass that were never intended to be exposed as HTTP handlers, bypassing authorization checks applied to intended public handlers. This can lead to unauthorized actions or information disclosure within applications using affected Starlette versions if routes are registered without method restrictions.
Mitigation Recommendations
Upgrade Starlette to version 1.1.0 or later, where this issue is fixed. Until then, ensure that all Route registrations explicitly specify the methods= argument to restrict allowed HTTP methods and prevent invocation of unintended internal methods.
CVE-2026-48817: CWE-470: Use of Externally-Controlled Input to Select Classes or Code ('Unsafe Reflection') in Kludex starlette
Description
Starlette versions prior to 1.1.0 have a vulnerability where HTTPEndpoint dispatches requests by dynamically invoking methods based on the HTTP method name without restricting to standard HTTP verbs. This allows an attacker to invoke internal methods named after non-standard HTTP verbs if the route is registered without an explicit methods= argument. This can lead to unauthorized invocation of internal helper methods that bypass authorization checks. The issue is fixed in version 1.1.0.
CVSS v3.1
Score 5.3medium
Affected software
Run on your own infrastructure? Check whether these packages are installed with threat-finder — our free open-source scanner.
Weaknesses
AI-Powered Analysis
Machine-generated threat intelligence
Technical Analysis
In Starlette versions 1.0.1 and below, the HTTPEndpoint class dispatches requests by lowercasing the HTTP method and using getattr to find a handler method. However, if a route is registered without specifying the methods= argument, all HTTP methods are accepted, including non-standard ones. If a non-standard HTTP method matches an attribute name on the endpoint subclass, that attribute is invoked as a handler, potentially exposing internal methods not intended to be HTTP handlers. This unsafe reflection can lead to unauthorized access to internal functionality. The vulnerability is addressed in Starlette 1.1.0 by restricting method dispatching.
Potential Impact
An attacker can invoke internal methods of an HTTPEndpoint subclass that were never intended to be exposed as HTTP handlers, bypassing authorization checks applied to intended public handlers. This can lead to unauthorized actions or information disclosure within applications using affected Starlette versions if routes are registered without method restrictions.
Mitigation Recommendations
Upgrade Starlette to version 1.1.0 or later, where this issue is fixed. Until then, ensure that all Route registrations explicitly specify the methods= argument to restrict allowed HTTP methods and prevent invocation of unintended internal methods.
Technical Details
- Data Version
- 5.2
- Assigner Short Name
- GitHub_M
- Date Reserved
- 2026-05-22T20:57:10.976Z
- Cvss Version
- 3.1
- State
- PUBLISHED
- Remediation Level
- null
Threat ID: 6a33019af198dc38c1fe1944
Added to database: 6/17/2026, 8:20:42 PM
Last enriched: 6/17/2026, 8:35:14 PM
Last updated: 6/17/2026, 9:28:19 PM
Views: 5
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.