GHSA-33F5-2C5Q-WGWJ
Vulnerability from github – Published: 2026-09-16 22:13 – Updated: 2026-09-16 22:13Summary
The rmcp library does not validate the resource parameter in OAuth Protected Resource metadata (RFC 9728), allowing a malicious MCP server to redirect OAuth flows to a legitimate authorization server and steal the resulting access tokens.
Details
RFC 9728 specifies two MUST requirements for resource parameter validation:
- Section 7.3: the client MUST ensure that the resource identifier URL it is using as the prefix for the metadata request exactly matches the resource value in the returned metadata document.
- Section 3.3: if the resource value returned is not identical to the URL the client used, the data MUST NOT be used.
In the current implementation (crates/rmcp/src/transport/auth.rs), the ResourceServerMetadata struct (lines 390–394) does not include a resource field:
struct ResourceServerMetadata {
authorization_server: Option<String>,
authorization_servers: Option<Vec<String>>,
scopes_supported: Option<Vec<String>>,
}
And discover_oauth_server_via_resource_metadata() (lines 1446–1465) proceeds without any resource URL validation.
Recommended fix
- Add the
resourcefield to the struct:rust struct ResourceServerMetadata { resource: Option<String>, // RFC 9728 REQUIRED field authorization_server: Option<String>, authorization_servers: Option<Vec<String>>, scopes_supported: Option<Vec<String>>, } -
Add validation logic after fetching metadata: ```rust let Some(resource_metadata) = self .fetch_resource_metadata_from_url(&resource_metadata_url) .await? else { return Ok(None); };
// RFC 9728: validate that the resource identifier matches our target server if let Some(resource) = &resource_metadata.resource { if resource.trim_end_matches('/') != self.base_url.as_str().trim_end_matches('/') { return Err(AuthError::MetadataError(format!( "Resource metadata mismatch: expected '{}', got '{}'", self.base_url, resource ))); } } ```
PoC
- Attacker sets up a malicious MCP server at
fake-mcp.com/mcp. -
At
fake-mcp.com/mcp/.well-known/oauth-protected-resource, the attacker serves metadata declaring:- resource:
real-mcp.com/mcp(the legitimate server) - authorization_servers: the legitimate authorization server(s) of
real-mcp.com/mcp
- resource:
-
Victim configures any MCP client using
rmcpto connect tofake-mcp.com/mcp. -
rmcpfetches the protected resource metadata and, without validating that theresourcefield (real-mcp.com/mcp) differs from the configured server (fake-mcp.com/mcp), initiates an OAuth flow with the legitimate authorization server. - The victim sees a legitimate authorization prompt and completes the flow.
- The resulting access token — valid for
real-mcp.com/mcp— is sent tofake-mcp.com/mcpin subsequent requests. - The attacker captures the token and can impersonate the victim on
real-mcp.com/mcp.
Impact
This is an access token theft vulnerability via OAuth resource metadata spoofing. All MCP clients built on rmcp that rely on OAuth-protected MCP servers are affected. An attacker who tricks a user into connecting to a malicious MCP server can steal valid access tokens for any legitimate MCP server, enabling full impersonation of the victim.
Credit
Jian Cui, Minsun Shim, Zhou Li, Xiaojing Liao University of Illinois Urbana-Champaign (UIUC) University of California, Irvine (UCI)
{
"affected": [
{
"package": {
"ecosystem": "crates.io",
"name": "rmcp"
},
"ranges": [
{
"events": [
{
"introduced": "0"
},
{
"fixed": "2.0.0"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"aliases": [
"CVE-2026-63127"
],
"database_specific": {
"cwe_ids": [
"CWE-345"
],
"github_reviewed": true,
"github_reviewed_at": "2026-09-16T22:13:26Z",
"nvd_published_at": "2026-09-16T15:17:39Z",
"severity": "HIGH"
},
"details": "### Summary\nThe `rmcp` library does not validate the `resource` parameter in OAuth Protected Resource metadata (RFC 9728), allowing a malicious MCP server to redirect OAuth flows to a legitimate authorization server and steal the resulting access tokens.\n\n### Details\nRFC 9728 specifies two MUST requirements for resource parameter validation:\n- Section 7.3: the client MUST ensure that the resource identifier URL it is using as the prefix for the metadata request exactly matches the `resource` value in the returned metadata document.\n- Section 3.3: if the `resource` value returned is not identical to the URL the client used, the data MUST NOT be used.\n\nIn the current implementation (`crates/rmcp/src/transport/auth.rs`), the `ResourceServerMetadata` struct (lines 390\u2013394) does not include a resource field:\n```rust\nstruct ResourceServerMetadata {\n authorization_server: Option\u003cString\u003e,\n authorization_servers: Option\u003cVec\u003cString\u003e\u003e,\n scopes_supported: Option\u003cVec\u003cString\u003e\u003e,\n}\n```\nAnd discover_oauth_server_via_resource_metadata() (lines 1446\u20131465) proceeds without any resource URL validation.\n\n#### Recommended fix\n1. Add the `resource` field to the struct:\n ```rust\n struct ResourceServerMetadata {\n resource: Option\u003cString\u003e, // RFC 9728 REQUIRED field\n authorization_server: Option\u003cString\u003e,\n authorization_servers: Option\u003cVec\u003cString\u003e\u003e,\n scopes_supported: Option\u003cVec\u003cString\u003e\u003e,\n }\n ```\n2. Add validation logic after fetching metadata:\n ```rust\n let Some(resource_metadata) = self\n .fetch_resource_metadata_from_url(\u0026resource_metadata_url)\n .await?\n else {\n return Ok(None);\n };\n \n // RFC 9728: validate that the resource identifier matches our target server\n if let Some(resource) = \u0026resource_metadata.resource {\n if resource.trim_end_matches(\u0027/\u0027) != self.base_url.as_str().trim_end_matches(\u0027/\u0027) {\n return Err(AuthError::MetadataError(format!(\n \"Resource metadata mismatch: expected \u0027{}\u0027, got \u0027{}\u0027\",\n self.base_url, resource\n )));\n }\n }\n ```\n\n### PoC\n\n1. Attacker sets up a malicious MCP server at `fake-mcp.com/mcp`. \n2. At `fake-mcp.com/mcp/.well-known/oauth-protected-resource`, the attacker serves metadata declaring:\n - resource: `real-mcp.com/mcp` (the legitimate server)\n - authorization_servers: the legitimate authorization server(s) of `real-mcp.com/mcp`\n\n\n3. Victim configures any MCP client using `rmcp` to connect to `fake-mcp.com/mcp`.\n\n4. `rmcp` fetches the protected resource metadata and, without validating that the `resource` field (`real-mcp.com/mcp`) differs from the configured server (`fake-mcp.com/mcp`), initiates an OAuth flow with the legitimate authorization server.\n5. The victim sees a legitimate authorization prompt and completes the flow.\n6. The resulting access token \u2014 valid for `real-mcp.com/mcp` \u2014 is sent to `fake-mcp.com/mcp` in subsequent requests.\n7. The attacker captures the token and can impersonate the victim on `real-mcp.com/mcp`.\n\n### Impact\nThis is an access token theft vulnerability via OAuth resource metadata spoofing. All MCP clients built on `rmcp` that rely on OAuth-protected MCP servers are affected. An attacker who tricks a user into connecting to a malicious MCP server can steal valid access tokens for any legitimate MCP server, enabling full impersonation of the victim.\n\n#### Credit\nJian Cui, Minsun Shim, Zhou Li, Xiaojing Liao\nUniversity of Illinois Urbana-Champaign (UIUC)\nUniversity of California, Irvine (UCI)",
"id": "GHSA-33f5-2c5q-wgwj",
"modified": "2026-09-16T22:13:26Z",
"published": "2026-09-16T22:13:26Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/modelcontextprotocol/rust-sdk/security/advisories/GHSA-33f5-2c5q-wgwj"
},
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-63127"
},
{
"type": "WEB",
"url": "https://github.com/modelcontextprotocol/rust-sdk/pull/937"
},
{
"type": "WEB",
"url": "https://github.com/modelcontextprotocol/rust-sdk/commit/c1a8b29ff2cc45e7820b900dae42cbb4958089ec"
},
{
"type": "PACKAGE",
"url": "https://github.com/modelcontextprotocol/rust-sdk"
},
{
"type": "WEB",
"url": "https://github.com/modelcontextprotocol/rust-sdk/releases/tag/rmcp-v2.0.0"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:L/A:N",
"type": "CVSS_V3"
}
],
"summary": "RMCP: Missing Resource Field Validation in OAuth Protected Resource Metadata Discovery"
}
Sightings
| Author | Source | Type | Date | Other |
|---|
Nomenclature
- Seen: The vulnerability was mentioned, discussed, or observed by the user.
- Confirmed: The vulnerability has been validated from an analyst's perspective.
- Published Proof of Concept: A public proof of concept is available for this vulnerability.
- Exploited: The vulnerability was observed as exploited by the user who reported the sighting.
- Patched: The vulnerability was observed as successfully patched by the user who reported the sighting.
- Not exploited: The vulnerability was not observed as exploited by the user who reported the sighting.
- Not confirmed: The user expressed doubt about the validity of the vulnerability.
- Not patched: The vulnerability was not observed as successfully patched by the user who reported the sighting.
The approach is described in our paper Mapping CVEs to MITRE ATT&CK Techniques: A Curated Gold-Set Classifier and the Limits of LLM-Assisted Label Expansion.
Browse all ATT&CK techniques and the vulnerabilities related to each.
Related by attack behaviour
Vulnerabilities whose description is nearest to this one in the vector space of the CIRCL/vulnerability-attack-technique-biencoder model. This is a similarity search over the bi-encoder space (plain cosine), not a classification, and it has no measured accuracy.