CVE-2026-44427 — registry
Fix: modelcontextprotocol/registry#1227CVE-2026-44427 is a Open Redirect vulnerability in github.com/modelcontextprotocol/registry. A fix is available for github.com/modelcontextprotocol/registry — see the affected versions and patch details below.
MCP Registry: Open Redirect
Exploitation Status
Proof-of-concept exploit code exists
- CISA’s SSVC triage found public proof-of-concept exploit code for this CVE, though no confirmed active exploitation.
Exploitation and automatability from CISA’s SSVC triage for CVE-2026-44427.
EPSS Exploitation Probability
EPSS (Exploit Prediction Scoring System) is a daily probability model maintained by FIRST.org. It estimates the likelihood a CVE will be exploited in production environments within the next 30 days, derived from real-world threat intelligence signals.
Real-World Exposure
github.com/modelcontextprotocol/registryReal-time download stats are indexed for npm and PyPI packages. This vulnerability affects Go packages — download data is not available via public APIs for these ecosystems.
Description
Summary
The TrailingSlashMiddleware in internal/api/server.go is vulnerable to an open redirect attack. An attacker can craft a URL with a protocol-relative path (e.g., //evil.com/) that, after trailing slash removal, results in a Location header of //evil.com — which browsers interpret as an absolute URL to an external domain.
Details
The TrailingSlashMiddleware strips trailing slashes from request paths and issues a 308 Permanent Redirect to the cleaned path. However, it does not validate or sanitize the resulting path before using it as the redirect target.
When a request is made with a path like //evil.com/, the middleware processes it as follows:
PoC
- Start the registry server locally or identify a deployed instance
- Send a request with a double-slash path followed by an external domain:
curl -v https://<registry-host>//evil.com/<img width="3066" height="969" alt="image" src="https://github.com/user-attachments/assets/a5305f00-29bf-4708-952a-478d608f2074" /> - Observe the 308 Permanent Redirect response with Location: //evil.com:
- When accessed in a browser, the user is redirected to https://evil.com
Impact
Phishing: Attackers can abuse the trusted registry domain to redirect users to credential-harvesting pages Malware distribution: Redirect users to sites serving malicious downloads Trust abuse: Links originating from the official MCP Registry domain carry implicit trust
Affected Packages
| Ecosystem | Package | Vulnerable range | Fix |
|---|---|---|---|
| 🐹Go | github.com/modelcontextprotocol/registry | ≥ 1.1.0&&< 1.7.5 | 1.7.5go get github.com/modelcontextprotocol/registry@v1.7.5 |
Detection & mitigation playbook
Open-source dependencyDetect
Scan your dependency tree (package-lock.json, pnpm-lock.yaml, requirements.txt, go.sum, etc.) for github.com/modelcontextprotocol/registry, including transitive dependencies — a direct dependency you never call can still pull in a vulnerable version.
Fix
Update github.com/modelcontextprotocol/registry to 1.7.5 or later, then make sure no transitive (indirect) dependency still pins the vulnerable range — O3 confirms CVE-2026-44427 is resolved across your whole dependency graph.
Workarounds
If you can't upgrade right away: gate or disable the affected feature, validate untrusted input at the boundary, and avoid passing attacker-controlled data into the vulnerable path. O3's runtime protection blocks exploitation in production as an interim safeguard until the upgrade lands.
How O3 protects you
O3 Security's impact-aware SCA analyses which vulnerable code paths your application actually calls, so a match like CVE-2026-44427 can be triaged on real exposure rather than presence alone.
Tailored to CVE-2026-44427. Runtime protection reduces exposure until a permanent patch is applied and verified — it complements patching, it doesn't replace it.
Frequently Asked Questions
Is CVE-2026-44427 in your dependencies?
O3 Security finds CVE-2026-44427 across Go dependencies, including transitive ones, and its impact-aware SCA ranks findings by whether your code actually calls the vulnerable path.