Description

MCPHub is a unified hub for centrally managing and dynamically orchestrating multiple MCP servers/APIs into separate endpoints with flexible routing strategies. Prior to version 1.0.29, MCPHub's PUT /api/system-config endpoint (handler updateSystemConfig) performs no authorization check. It is protected only by the app-wide authentication middleware and a rate limiter — it never inspects req.user.isAdmin. This issue has been patched in version 1.0.29.

Executive Summary

Technical Details

Mitigations

References

Additional Metadata

{
  "score": "8.8",
  "pubDate": "2026-08-31T18:17:19.763Z",
  "pubdate": "2026-08-31T18:17:19.763Z",
  "executiveSummary": "MCPHub is susceptible to a critical broken access control vulnerability originating from the lack of authorization validation within its system configuration interface.\nThe vulnerability resides in the PUT /api/system-config endpoint, which fails to verify if the authenticated user possesses administrative privileges before processing configuration updates.\nBy bypassing the implicit trust model assumed by the application's global authentication middleware, any authenticated user—regardless of their privilege level—can modify core system configurations.\nThis flaw allows unauthorized actors to manipulate orchestration settings, potentially leading to unauthorized redirection of traffic, service disruption, or full control over the connected MCP server integrations.\nThe risk is high, as it transforms a restricted administrative function into a publicly accessible endpoint for all authenticated entities within the environment.\nThe issue affects all MCPHub versions prior to 1.0.29. Mitigation requires immediate upgrading to the patched version, as the underlying architecture does not enforce role-based access control (RBAC) at the route level for this specific request handler.",
  "technicalDetails": "The vulnerability is identified as a broken access control flaw within the MCPHub API surface. The root cause is the failure of the updateSystemConfig handler to implement server-side authorization checks.\nWhile the application employs global authentication middleware and a rate limiter, the updateSystemConfig function does not inspect the 'req.user.isAdmin' property or any equivalent authorization context.\nConsequently, the authorization logic relies solely on the presence of a valid session token rather than validating the user's role or permissions against the action being requested.\nAn attacker can exploit this by authenticating as a low-privileged user and issuing a crafted PUT request to the /api/system-config endpoint. Because the application logic fails to perform a secondary validation step after the initial authentication handshake, the request is processed as if it originated from an administrative source.\nThe attack flow follows these steps: 1) The attacker authenticates to MCPHub using a standard user account. 2) The attacker intercepts or crafts a PUT request targeting /api/system-config, containing arbitrary or malicious configuration payloads. 3) The application's middleware successfully authenticates the user but does not verify administrative status. 4) The updateSystemConfig handler receives the request and executes the update logic without further verification. 5) The system state is modified according to the attacker's inputs.\nSuccessful exploitation allows for the modification of critical system parameters, such as dynamic routing tables, upstream API endpoint addresses, or security headers. This could lead to an adversary rerouting internal API requests to rogue servers, effectively enabling man-in-the-middle (MITM) attacks or data exfiltration from integrated MCP services.\nThe vulnerability exists in the handler logic directly associated with the /api/system-config route. The flaw is present in all versions prior to 1.0.29. The exposure is network-wide for any user with a valid authentication token, meaning the impact scales with the number of permitted users in the environment.\nPost-exploitation impact includes unauthorized system reconfiguration, persistent service denial, and the potential to compromise sensitive data flows handled by the hub's orchestration capabilities."
}