LiteLLM is a proxy server (AI Gateway) to call LLM APIs in OpenAI (or native) format. Prior to 1.84.0, LiteLLM's MCP Streamable HTTP endpoint allowed an unauthenticated attacker to use a fabricated Authorization header to trigger an OAuth2 passthrough fallback path that replaced failed LiteLLM key validation with an empty UserAPIKeyAuth() object, allowing requests to reach MCP tooling without a valid LiteLLM key. This issue is fixed in version 1.84.0.
Data as of 2 Nov 2026 · NVD, CISA KEV, EPSS
Defenders still triage vulnerabilities one CVE at a time, while increasingly automated attackers combine them at machine speed. Omenwave models the plausible pathways by which this CVE could be chained with others — surfacing them for research and awareness, with evidence for each step.
The access level the source grants already meets or exceeds the access level the destination requires — the position carries over.
LiteLLM is a proxy server (AI Gateway) to call LLM APIs in OpenAI (or native) format. From version 1.74.2 to before version 1.83.7, two endpoints use…
The access level the source grants already meets or exceeds the access level the destination requires — the position carries over.
Improper control of generation of code ('code injection') in Microsoft Office SharePoint allows an authorized attacker to execute code over a network.
The access level the source grants already meets or exceeds the access level the destination requires — the position carries over.
Microsoft Exchange Server Elevation of Privilege Vulnerability
The access level the source grants already meets or exceeds the access level the destination requires — the position carries over.
A remote code execution vulnerability exists in Microsoft Exchange software when the software fails to properly handle objects in memory, aka 'Micros…
The access level the source grants already meets or exceeds the access level the destination requires — the position carries over.
Roundcube Webmail before 1.5.10 and 1.6.x before 1.6.11 allows remote code execution by authenticated users because the _from parameter in a URL is n…
Modelled plausible chaining potential derived from public CVE metadata (CVSS, CWE, CPE). Not a confirmed real-world exploit chain and contains no exploitation detail. For research and awareness only.
Methodology · Interactive explorer · JSON: /cve/CVE-2026-59822