CVE-2026-79745
nicheMissing Role Check in MCPHub Lets Non-Admins Tamper With Global Prompts and Resources
MCPHub prior to version 1.0.32 contains a missing-authorization flaw (CWE-862) in its built-in prompt and resource controllers, which perform no role checking and never read the requesting user's identity. Any authenticated non-admin user can send POST/PUT requests to the /api/prompts* and /api/resources* routes and create, overwrite, or shadow global prompt templates and resources, because those DAO records are consulted first for every user session. An attacker thereby gains unauthorized modification of data served to all other users of the hub, including the ability to plant stored prompt injections that reach other users' LLM sessions. All deployments of MCPHub running versions before 1.0.32 are affected, since the mutating routes are attached to the authenticated router without an admin gate. No public proof-of-concept, CISA KEV listing, or reported in-the-wild exploitation is known; EPSS currently estimates only a 0.2% chance of exploitation within 30 days.
What to do: Upgrade MCPHub to version 1.0.32, which adds the missing role checks on the prompt and resource write routes. As an interim mitigation, restrict POST/PUT access to /api/prompts* and /api/resources* to administrators at a reverse proxy or firewall layer, and audit existing stored prompts and resources for unauthorized creations or tampering that could inject content into other users' LLM sessions.
| MCPHub | all versions prior to 1.0.32 |
Order-of-magnitude estimate by the model from install counts, market share and public scan data it knows; verify before quoting.
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 1.0.32, the built-in prompt and resource controllers perform no role checking. The mutating POST/PUT /api/prompts* and POST/PUT /api/resources* routes are attached to the authenticated router with no admin gate, and the handlers never read req.user. The DAO singletons they write are consulted first — ahead of any connected MCP server — for every session in handleGetPromptRequest / handleReadResourceRequest. A non-admin can therefore create, overwrite, and shadow global prompt templates and resources that all other users are served. The scored impact is the unauthorized integrity violation (creation/tampering/shadowing of globally-served records); stored prompt injection into other users' LLM sessions is a downstream consequence of that tampering. This issue has been patched in version 1.0.32.
- Weakness
- CWE-862
- Vector
- CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:H/A:L
In the news0 stories
No ingested article mentions this CVE yet.