Your RSA-2048 keys break in 2030. Find every one of them before attackers do.
🐹
🐹 Go
Not in CISA KEV
CRITICAL severity

GHSA-xqhv-chqm-fhcc

CRITICAL

GHSA-xqhv-chqm-fhcc is a critical-severity (CVSS 9.6) Missing Authentication vulnerability in github.com/BishopFox/joro. O3 Security confirms whether GHSA-xqhv-chqm-fhcc is actually reachable in your code before you act, and blocks exploitation at runtime until you patch.

Joro: Unauthenticated Cross-Origin Plugin Upload Leads to RCE

Also known asCVE-2026-53649GO-2026-5947
Published
Jul 8, 2026
Updated
Jul 21, 2026
Affected
1 pkg
Patched
1 / 1
Exploits
None indexed
Exploitation data as of Sep 8, 2026 · OSV.dev, NVD, FIRST.org (EPSS)

Exploitation Status

No confirmed exploitation observed yet

  • A successful exploit gives an attacker total control of the affected component, not partial access.
  • CISA’s own triage has not observed active exploitation or public proof-of-concept code for this CVE as of its last assessment.

Exploitation and automatability from CISA’s SSVC triage for GHSA-xqhv-chqm-fhcc.

EPSS Exploitation Probability

via FIRST.org ↗
0.2%probability of exploitation in next 30 days
Lower Risk0.00%
Lower risk than most CVEs12th percentile — riskier than 12% of all scored CVEsHighest risk

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.

How urgent is this, really

GHSA-xqhv-chqm-fhcc plotted by exploitation likelihood (EPSS) against impact (CVSS). The shaded corner — EPSS 50%+ and CVSS 7.0+ — is where this CVE doesn't sit, though severity or exploitability alone can still warrant action.

Where this sits among everything scored

Of 369,023 CVEs with a current EPSS score, this one falls in the < 10% band (highlighted). Real counts from FIRST.org, not a sample — log-scaled since the landscape is heavily right-skewed.

Real-World Exposure

1 pkg affected
🐹github.com/BishopFox/joro

Real-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

Unauthenticated Cross-Origin Plugin Upload Leads to RCE (Joro ≤ v1.1.0)

Severity: Critical CVSS v3.1: 9.6 (AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H) Affected versions: Joro ≤ v1.1.0, proxy mode (default), Linux/macOS Reporter: cstover Date: 2026-05-27


Summary

Joro's default proxy mode (in versions <= 1.1.0) exposes a local API on 127.0.0.1:9090 that performs no authentication and applies a wildcard CORS policy. Because plugin uploads use the CORS-safelisted multipart/form-data content type, cross-origin JavaScript on any page the operator visits can reach privileged endpoints - including uploading a native plugin and triggering a restart - directly through the operator's browser, with no preflight or credentials. Since plugins execute on load, this yields unauthenticated remote code execution as the operator's user from a single page visit.


Root Cause

Three weaknesses combined into the exploit chain.

1. No authentication in proxy mode. internal/api/server.go applied AuthMiddleware only when listenerMode was true. In the default proxy mode every API endpoint — including plugin upload and system restart — accepted requests without any token, cookie, or credential.

2. Permissive CORS with an insufficient protection assumption. corsMiddleware set Access-Control-Allow-Origin: * unconditionally on all responses. SECURITY.md documented this as an intentional tradeoff on the basis that proxy mode binds to 127.0.0.1, which the document states "limits exposure to the local machine."

That assumption was incorrect. multipart/form-data is a CORS-safelisted Content-Type, so cross-origin JavaScript can POST files to the Joro API without triggering a preflight request — the browser allows it. Any web page the operator visited reached the localhost API through their browser without restriction. The localhost bind provided no protection against browser-mediated requests.

3. Plugin init() executed on plugin.Open() before symbol lookup. internal/plugins/loader.go called plugin.Open(), which ran the plugin's init() functions before any symbol lookup occurred. A plugin with no exports still executed its payload the moment Joro restarted.


Attack Chain

  1. The operator visits an attacker-controlled page in Firefox on their machine.
  2. JavaScript on the page fetches pwn.so from the attacker's server (same-origin, no CORS issue).
  3. JavaScript POSTs pwn.so to http://127.0.0.1:9090/api/v1/plugins/upload as multipart/form-data. Joro accepts it — no auth, no preflight.
  4. JavaScript POSTs to http://127.0.0.1:9090/api/v1/system/restart. Joro re-executes.
  5. On restart, plugin.Open("pwn.so") calls init(), which opens a goroutine and dials back to the attacker's listener.
  6. An interactive /bin/bash -i shell is obtained as the operator's user.

The plugin ABI matches without any access to the operator's machine. The same public v1.1.0 release tarball is downloaded and Joro's own --build-plugin feature is used, which reads runtime/debug.BuildInfo from the release binary and forwards every ABI-relevant flag. One .so works against every operator running that release.


Impact

Unauthenticated, remote, browser-mediated code execution as the operator's user. Because the exploit pivots through the operator's browser to the loopback-bound API, the network bind offers no protection, and a single ABI-matched plugin works against every operator running the affected release.

Fix

The chain is broken at multiple layers. Cross-origin browser access to the proxy-mode API is eliminated, the API is restricted to same-origin requests targeting a loopback host, and the UI/API is bound to loopback only.

1. Removed the wildcard CORS header and gated the proxy-mode API behind a same-origin guard

corsMiddleware (which set Access-Control-Allow-Origin: * on every response) was deleted, and proxy mode now wraps the API in originGuard instead. (internal/api/server.go, commit 5c0ca35)

 var handler http.Handler = mux
 if s.listenerMode {
+     // Listener/teamserver: bearer-token auth.
      handler = team.AuthMiddleware(s.teamToken, handler)
+} else {
+     // Proxy mode: restrict the API to same-origin browser requests.
+     handler = originGuard(uiBind, handler)
 }
-handler = corsMiddleware(handler)
-// corsMiddleware adds permissive CORS headers for dev usage.
-func corsMiddleware(next http.Handler) http.Handler {
-     return http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
-             w.Header().Set("Access-Control-Allow-Origin", "*")
-             w.Header().Set("Access-Control-Allow-Methods", "GET, POST, PUT, DELETE, OPTIONS")
-             w.Header().Set("Access-Control-Allow-Headers", "Content-Type, Authorization, X-Joro-Nickname")
-             if r.Method == http.MethodOptions {
-                     w.WriteHeader(http.StatusNoContent)
-                     return
-             }
-             next.ServeHTTP(w, r)
-     })
-}

2. Same-origin enforcement via Sec-Fetch-Site + Origin/Host

originGuard rejects state-changing requests (and the /ws upgrade) whose Sec-Fetch-Site indicates a cross-origin initiator or whose Origin host does not match the request Host. Non-browser local tooling (no browser headers) is still allowed. (internal/api/originguard.go, commit 5c0ca35)

func isMutating(method string) bool {
      switch method {
      case http.MethodPost, http.MethodPut, http.MethodDelete, http.MethodPatch:
              return true
      default:
              return false
      }
}

func sameOrigin(r *http.Request) bool {
      switch r.Header.Get("Sec-Fetch-Site") {
      case "", "same-origin", "none":
              // Same-origin, a direct navigation, or a non-browser client.
      default: // "cross-site", "same-site"
              return false
      }
      if origin := r.Header.Get("Origin"); origin != "" {
              if origin == "null" {
                      return false // opaque/sandboxed cross-origin context
              }
              u, err := url.Parse(origin)
              if err != nil || !strings.EqualFold(reqHostname(u.Host), reqHostname(r.Host)) {
                      return false
              }
      }
      return true
}

3. Tightened the WebSocket origin check

The WebSocket upgrader previously accepted every origin (CheckOrigin: return true). It now rejects cross-origin handshakes while still permitting non-browser clients. (internal/api/ws.go, commit 5c0ca35)

var upgrader = websocket.Upgrader{
-     CheckOrigin: func(r *http.Request) bool { return true },
+     CheckOrigin: func(r *http.Request) bool {
+             origin := r.Header.Get("Origin")
+             if origin == "" {
+                     return true
+             }
+             if origin == "null" {
+                     return false
+             }
+             u, err := url.Parse(origin)
+             if err != nil {
+                     return false
+             }
+             return strings.EqualFold(reqHostname(u.Host), reqHostname(r.Host))
+     },
 }

4. Bound the proxy-mode UI/API to loopback and removed the wildcard host exception

The same-origin check alone can be defeated by DNS rebinding under a wildcard bind, because a rebound host (e.g. attacker.com) carries consistent Origin/Host/Sec-Fetch-Site headers. Two coordinated changes close this: the proxy-mode UI/API now binds to 127.0.0.1 regardless of --bind (which governs only the proxy port), and hostAllowed no longer has a wildcard exception, so the host must be loopback or the exact bind address. (internal/api/server.go and internal/api/originguard.go, commit 871936f)

+// In proxy mode the UI/API binds to loopback only: --bind governs the proxy
+// port, and remote collaboration is listener/teamserver mode (bearer-token auth).
+uiBind := s.cfg.BindAddr
+if !s.listenerMode {
+     uiBind = "127.0.0.1"
+}
+
 var handler http.Handler = mux
 ...
 s.srv = &http.Server{
-     Addr:              fmt.Sprintf("%s:%d", s.cfg.BindAddr, s.cfg.UIPort),
+     Addr:              fmt.Sprintf("%s:%d", uiBind, s.cfg.UIPort),
 func hostAllowed(reqHost, bindAddr string) bool {
      h := reqHostname(reqHost)
      if h == "" {
              return false
      }
      switch h {
      case "localhost", "127.0.0.1", "::1":
              return true
      }
-     switch bindAddr {
-     case "", "0.0.0.0", "::":
-             return true
-     }
      return strings.EqualFold(h, reqHostname(bindAddr))
 }

Affected Packages

1 total 1 fixed
EcosystemPackageVulnerable rangeFix
🐹Gogithub.com/BishopFox/joroall versions0.0.0-20260601151442-5c0ca35db828

Detection & mitigation playbook

Open-source dependency
  1. Detect

    Scan your dependency tree (package-lock.json, pnpm-lock.yaml, requirements.txt, go.sum, etc.) for github.com/BishopFox/joro. 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.

  2. Fix

    Update github.com/BishopFox/joro to 0.0.0-20260601151442-5c0ca35db828 or later, then make sure no transitive (indirect) dependency still pins the vulnerable range — O3 confirms GHSA-xqhv-chqm-fhcc is resolved across your whole dependency graph.

  3. 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.

  4. How O3 protects you

    O3 pinpoints whether GHSA-xqhv-chqm-fhcc 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-xqhv-chqm-fhcc. Runtime protection reduces exposure until a permanent patch is applied and verified — it complements patching, it doesn't replace it.

Frequently Asked Questions

# Unauthenticated Cross-Origin Plugin Upload Leads to RCE (Joro ≤ v1.1.0) **Severity:** Critical **CVSS v3.1:** 9.6 (AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H) **Affected versions:** Joro ≤ v1.1.0, proxy mode (default), Linux/macOS **Reporter:** cstover **Date:** 2026-05-27 --- ## Summary Joro's default proxy mode (in versions <= 1.1.0) exposes a local API on `127.0.0.1:9090` that performs no authentication and applies a wildcard CORS policy. Because plugin uploads use the CORS-safelisted `multipart/form-data` content type, cross-origin JavaScript on any page the operator visits can reach privil
O3 Security · Impact-Aware SCA

Is GHSA-xqhv-chqm-fhcc in your dependencies?

O3 detects GHSA-xqhv-chqm-fhcc across Go dependencies and uses function-level reachability to confirm whether the vulnerable code path is actually reachable — not just present. No false positives.

GHSA-xqhv-chqm-fhcc: joro Remote Code… | O3 Security