Mcpm: Starlette: Unvalidated request path concatenated into authority poisons request.url.hostname (CVE-2026-54282)
Description
A vulnerability in mcpm's Starlette component allows an attacker to manipulate the reconstructed request URL's hostname by sending a request path that does not start with a slash. This causes the URL parser to misinterpret the path as part of the authority section, leading to an attacker-controlled hostname in request.url.hostname. This can mislead code that relies on request.url.hostname for security decisions. The issue affects versions where the request path is not validated before URL reconstruction. A patch is available that prevents the path from crossing into the URL authority.
CVSS v3.1
Score 3.7low
Affected software
Run on your own infrastructure? Check whether these packages are installed with threat-finder — our free open-source scanner.
AI-Powered Analysis
Machine-generated threat intelligence
Technical Analysis
In affected versions of mcpm's Starlette, the HTTP request path is concatenated directly after the host without validating that it begins with a '/'. When a request path lacks a leading slash (e.g., '@google.com'), the reconstructed request URL is parsed such that the hostname becomes attacker-controlled (e.g., 'google.com'). This occurs because the URL is rebuilt as '{scheme}://{host}{path}' and re-parsed, causing the path to be interpreted as userinfo and hostname per RFC 3986. This vulnerability can mislead code that uses request.url.hostname, request.url.netloc, or request.url for security-sensitive decisions. Exploitation requires an ASGI server forwarding a request-target lacking a leading slash into scope['path']. The exposure is limited to middleware or handlers that inspect request.url before routing or in 404/exception handlers. A patch is available that ensures the path does not cross into the URL authority.
Potential Impact
Applications using affected versions that rely on request.url.hostname or related properties for security decisions such as host-based authorization, redirects, SSRF targets, cache keys, or audit logs may be misled by attacker-controlled hostnames. However, the exploitability is limited because the malformed path does not match registered routes, resulting in 404 responses and no endpoint execution. The vulnerability primarily affects middleware or exception handlers that inspect the URL before routing. There is no known exploitation in the wild. The CVSS score is 3.7 (low severity) reflecting limited impact and complexity.
Mitigation Recommendations
Upgrade to a patched version of mcpm that prevents the request path from crossing into the URL authority. The patch ensures that malformed paths such as '@google.com' are correctly interpreted as part of the path (e.g., 'http://localhost/@google.com') with the hostname remaining trustworthy. No other mitigations are indicated. Check vendor advisories for the exact patched versions.
Technical Details
- Gcve Source
- db.gcve.eu
- Osv Id
- CLEANSTART-2026-TE29052
- Osv Schema Version
- 1.7.3
- Ecosystems
- ["Alpine"]
- State
- PUBLISHED
Threat ID: 6a984559acd9273b496ab7c7
Added to database: 09/02/2026, 15:48:41 UTC
Last enriched: 10/04/2026, 21:38:09 UTC
Last updated: 10/17/2026, 18:48:20 UTC
Views: 56
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.