GHSA-5P9G-J988-PCWV
Vulnerability from github – Published: 2026-07-30 14:44 – Updated: 2026-07-30 14:44Summary
Vulnerability: Missing Session Ownership Validation in the Ruby MCP SDK's Streamable and SSE HTTP transport implementation. Any attacker with a stolen session ID can execute tools with the victim's session. This is a silent attack - the victim's session is compromised and being used for unauthorized actions, but it is hard to know for the victim
Details
https://github.com/modelcontextprotocol/ruby-sdk/blob/main/lib/mcp/server/transports/streamable_http_transport.rb#L260-L278
Victim starts a legitimate MCP session and receives session ID abc-123 Attacker obtains the session ID (various means - network sniffing, logs, etc. out of scope for this analysis) Attacker sends POST to /messages/abc-123 with a tool call Server accepts the request (no ownership validation!) Server executes the tool and sends response to victim's SSE stream Victim receives attacker's response, thinking it's legitimate
PoC
attacker_client.py legitimate_client.py
Prerequisites
- Python 3.8+
- Install dependencies:
requests
Running the Demo
- Terminal 1: Start the Ruby MCP Server
ruby streamable_http_server.rb
Makes use of https://github.com/modelcontextprotocol/ruby-sdk/blob/main/examples/streamable_http_server.rb This server has a tool call notification_tool which the clients call
- Terminal 2: Start Victim Client
python3 legitimate_client.py
- Terminal 3 - Attacker Client:
Copy the session ID from Terminal 1 and run:
python3 attacker_client.py abc-123-def-456
- Back to Terminal 2 - Victim sees the injected response:
Impact
- Integrity: HIGH - Attacker can execute unauthorized tools and modify state
- Availability: LOW - Attacker can disrupt victim's session with injected responses
Additional Details
Session Hijacking Protection in MCP Implementations
The MCP specification recommends - "MCP servers SHOULD bind session IDs to user-specific information".
User Binding - Comparison other SDKs
csharp-sdk
- https://github.com/modelcontextprotocol/csharp-sdk/blob/main/src/ModelContextProtocol.AspNetCore/SseHandler.cs#L36
- https://github.com/modelcontextprotocol/csharp-sdk/blob/main/src/ModelContextProtocol.AspNetCore/SseHandler.cs#L95-L98
Go-sdk
- https://github.com/modelcontextprotocol/go-sdk/blob/main/mcp/streamable.go#L66
- https://github.com/modelcontextprotocol/go-sdk/blob/main/mcp/streamable.go#L316-L320
Comparison with the Stream Replacement vulnerability
Stream Replacement https://github.com/modelcontextprotocol/ruby-sdk/security/advisories/GHSA-qvqr-5cv7-wh35
- Target: SSE stream connection
- Attack Vector: GET /mcp
- What happens to Victim: The connection is disconnected and doesn't receive reponses
Session Poisoning
- Target: Tool execution
- Attack Vector: POST /mcp
- What happens to Victim: The connection stays connected and receives responses of tool calls by attacker
{
"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-2026-67431"
],
"database_specific": {
"cwe_ids": [
"CWE-284"
],
"github_reviewed": true,
"github_reviewed_at": "2026-07-30T14:44:28Z",
"nvd_published_at": "2026-07-29T20:17:12Z",
"severity": "HIGH"
},
"details": "### Summary\n**Vulnerability**: Missing Session Ownership Validation in the Ruby MCP SDK\u0027s Streamable and SSE HTTP transport implementation. Any attacker with a stolen session ID can execute tools with the victim\u0027s session. This is a silent attack - the victim\u0027s session is compromised and being used for unauthorized actions, but it is hard to know for the victim\n\n### Details\nhttps://github.com/modelcontextprotocol/ruby-sdk/blob/main/lib/mcp/server/transports/streamable_http_transport.rb#L260-L278\n\n**Victim** starts a legitimate MCP session and receives session ID abc-123\n**Attacker** obtains the session ID (various means - network sniffing, logs, etc. out of scope for this analysis)\n**Attacker** sends POST to /messages/abc-123 with a tool call\n**Server** accepts the request (no ownership validation!)\n**Server** executes the tool and sends response to victim\u0027s SSE stream\n**Victim** receives attacker\u0027s response, thinking it\u0027s legitimate\n\n### PoC\n[attacker_client.py](https://github.com/user-attachments/files/26044817/attacker_client.py)\n[legitimate_client.py](https://github.com/user-attachments/files/26044818/legitimate_client.py)\n\n**Prerequisites**\n\n1. Python 3.8+\n2. Install dependencies: `requests`\n\n**Running the Demo**\n\n1. **Terminal 1:** Start the Ruby MCP Server\n\n`ruby streamable_http_server.rb`\n\nMakes use of https://github.com/modelcontextprotocol/ruby-sdk/blob/main/examples/streamable_http_server.rb\nThis server has a tool call notification_tool which the clients call\n\n2. **Terminal 2:** Start Victim Client\n\n`python3 legitimate_client.py`\n\n3. **Terminal 3** - Attacker Client:\n\nCopy the session ID from Terminal 1 and run:\n\n```bash\npython3 attacker_client.py abc-123-def-456\n```\n4. **Back to Terminal 2** - Victim sees the injected response:\n\n### Impact\n\n- **Integrity:** HIGH - Attacker can execute unauthorized tools and modify state\n- **Availability:** LOW - Attacker can disrupt victim\u0027s session with injected responses\n\n### Additional Details\nSession Hijacking Protection in MCP Implementations\n\nThe MCP specification recommends - \"[MCP servers SHOULD bind session IDs to user-specific information](https://modelcontextprotocol.io/docs/tutorials/security/security_best_practices#mitigation-4)\".\n\n#### User Binding - Comparison other SDKs\n**csharp-sdk**\n\n- https://github.com/modelcontextprotocol/csharp-sdk/blob/main/src/ModelContextProtocol.AspNetCore/SseHandler.cs#L36\n- https://github.com/modelcontextprotocol/csharp-sdk/blob/main/src/ModelContextProtocol.AspNetCore/SseHandler.cs#L95-L98\n\n**Go-sdk**\n\n- https://github.com/modelcontextprotocol/go-sdk/blob/main/mcp/streamable.go#L66\n- https://github.com/modelcontextprotocol/go-sdk/blob/main/mcp/streamable.go#L316-L320\n\n#### Comparison with the Stream Replacement vulnerability \n\n**Stream Replacement**\nhttps://github.com/modelcontextprotocol/ruby-sdk/security/advisories/GHSA-qvqr-5cv7-wh35\n\n- **Target**: SSE stream connection\n- **Attack Vector**: GET /mcp\n- **What happens to Victim**: The connection is disconnected and doesn\u0027t receive reponses \n\n**Session Poisoning**\n\n- **Target**: Tool execution\n- **Attack Vector**: POST /mcp\n- **What happens to Victim**: The connection stays connected and receives responses of tool calls by attacker",
"id": "GHSA-5p9g-j988-pcwv",
"modified": "2026-07-30T14:44:28Z",
"published": "2026-07-30T14:44:28Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/modelcontextprotocol/ruby-sdk/security/advisories/GHSA-5p9g-j988-pcwv"
},
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2026-67431"
},
{
"type": "WEB",
"url": "https://github.com/modelcontextprotocol/ruby-sdk/commit/35466605319a34e4c7808712ae9bb1ca1afb2356"
},
{
"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:N/VC:N/VI:H/VA:L/SC:N/SI:N/SA:N",
"type": "CVSS_V4"
}
],
"summary": "MCP Ruby SDK: Ruby SSE Session Poisoning"
}
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.