ToolHive is a utility designed to simplify the deployment and management of Model Context Protocol (MCP) servers. Prior to 0.31.0, remote.Handler.Authenticate in pkg/auth/remote/handler.go invokes discovery.DetectAuthenticationFromServer in pkg/auth/discovery/discovery.go, whose host-side HTTP clients trust remote-server-controlled authentication discovery destinations, follow redirects without host or scheme restrictions, and do not consistently block private addresses. A malicious or compromised remote MCP server can place a resource_metadata URL in WWW-Authenticate for ParseWWWAuthenticate to extract, after which FetchResourceMetadata, OIDC issuer discovery, and well-known discovery can issue GET requests to link-local, RFC1918, or other internal services outside the server's container. The user connects to a server that the user intends to use, but the server controls the internal destination; this path does not depend on the separate NAT64 guard omission because the affected clients did not invoke IsPrivateIP, and it differs from the DCR resolver path that already refused redirects. The demonstrated primitive reaches internal-only HTTP services and reachability or error oracles, and can retrieve AWS metadata credentials where IMDSv1 accepts an unauthenticated GET, while IMDSv2 and GCP metadata prerequisites are not satisfied by the demonstrated request. This vulnerability is fixed in 0.31.0.

Affected: Go: github.com/stacklok/toolhive

# Security Advisory: SSRF in remote MCP server authentication discovery

**Severity:** High. **CWE:** CWE-918. **Affected:** ToolHive through the latest release v0.29.3 and current `main` (HEAD b672d82f, 2026-06-12; re-verified 2026-06-14). `FetchResourceMetadata` and the discovery clients remain unguarded; no commits to `pkg/auth/discovery` or `pkg/auth/remote` address this. Originally identified at commit 05f11b53; all line references below are against HEAD b672d82f.

## Summary ToolHive's remote MCP server authentication discovery issues outbound HTTP requests to URLs the remote MCP server controls, with no private-IP or loopback guard and no restriction on redirects. ToolHive's core security model treats every MCP server as untrusted: the README states it "runs every MCP server in an isolated container" with "no local credentials," and it ships an egress proxy for network isolation. This discovery code runs host-side, in the ToolHive process, before and outside that per-server container sandbox. A malicious or compromised remote MCP server, added by a user through ToolHive's normal remote-server workflow, can therefore drive the ToolHive host itself to fetch arbitrary internal URLs, including cloud instance metadata, which bypasses the isolation ToolHive exists to provide. The user never selects a malicious target; they connect to a server they intend to use, and the attack is carried entirely in that server's discovery response.

ToolHive already establishes this bounda…