GHSA-RWRF-2PQF-9J8J

Vulnerability from github – Published: 2026-10-27 21:05 – Updated: 2026-10-27 21:05
VLAI
Summary
Model Context Protocol Python SDK: Client fetched server-chosen $ref URLs while validating tool results against output schemas
Details

Summary

In affected versions, when a connected MCP server declared an outputSchema for a tool, the SDK's client validated each call_tool result with jsonschema's default resolver, which fetches any $ref it cannot resolve locally by opening the URL. A server could therefore put an http://, https:// or file:// reference in a tool's output schema and make the client:

  • request that URL, or read that local file, from the client's own network position on every call of the tool (the content was used only as a schema and never sent back to the server), and
  • stall: the fetch ran synchronously on the event loop with no timeout, so a target that accepted the connection and then went quiet froze every session in the process for as long as it held the connection.

Am I affected?

Yes, if your application uses the SDK as an MCP client (ClientSession over any transport, ClientSessionGroup, or the 2.x Client) and calls tools on servers you do not fully control. There was no setting that turned the validation off.

Affected versions: 1.10.0 through 1.29.1, and 2.0.0 through 2.1.1.

Not affected: applications that only build servers. If nothing in your code uses ClientSession, ClientSessionGroup or the 2.x Client class, you are not affected.

Fix

Upgrade to mcp 1.30.0 (1.x) or 2.2.0 (2.x) or later. The client now resolves references only within the schema document the server sent (plus the standard JSON Schema metaschemas), so result validation performs no network or file access. A tool whose output schema points outside its own document fails at call time with RuntimeError (Invalid schema for tool ...); in-document references such as #/$defs/... work as before.

Until you can upgrade, call tools only on servers you trust. As a partial stopgap, skip tools whose outputSchema contains an $id, or a $ref / $dynamicRef that does not start with #.

Severity
6.5 (Medium)
CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:N/I:N/A:H
Show details on source website
To clipboard 

{
  "affected": [
    {
      "package": {
        "ecosystem": "PyPI",
        "name": "mcp"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "2.0.0"
            },
            {
              "fixed": "2.2.0"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    },
    {
      "package": {
        "ecosystem": "PyPI",
        "name": "mcp"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "1.10.0"
            },
            {
              "fixed": "1.30.0"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "aliases": [
    "CVE-2026-116686"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-918"
    ],
    "github_reviewed": true,
    "github_reviewed_at": "2026-10-27T21:05:29Z",
    "nvd_published_at": null,
    "severity": "MODERATE"
  },
  "details": "### Summary\nIn affected versions, when a connected MCP server declared an `outputSchema` for a tool, the SDK\u0027s client validated each `call_tool` result with `jsonschema`\u0027s default resolver, which fetches any `$ref` it cannot resolve locally by opening the URL. A server could therefore put an `http://`, `https://` or `file://` reference in a tool\u0027s output schema and make the client:\n\n- request that URL, or read that local file, from the client\u0027s own network position on every call of the tool (the content was used only as a schema and never sent back to the server), and\n- stall: the fetch ran synchronously on the event loop with no timeout, so a target that accepted the connection and then went quiet froze every session in the process for as long as it held the connection.\n\n### Am I affected?\nYes, if your application uses the SDK as an MCP client (`ClientSession` over any transport, `ClientSessionGroup`, or the 2.x `Client`) and calls tools on servers you do not fully control. There was no setting that turned the validation off.\n\nAffected versions: 1.10.0 through 1.29.1, and 2.0.0 through 2.1.1.\n\nNot affected: applications that only build servers. If nothing in your code uses `ClientSession`, `ClientSessionGroup` or the 2.x `Client` class, you are not affected.\n\n### Fix\nUpgrade to `mcp` 1.30.0 (1.x) or 2.2.0 (2.x) or later. The client now resolves references only within the schema document the server sent (plus the standard JSON Schema metaschemas), so result validation performs no network or file access. A tool whose output schema points outside its own document fails at call time with `RuntimeError` (`Invalid schema for tool ...`); in-document references such as `#/$defs/...` work as before.\n\nUntil you can upgrade, call tools only on servers you trust. As a partial stopgap, skip tools whose `outputSchema` contains an `$id`, or a `$ref` / `$dynamicRef` that does not start with `#`.",
  "id": "GHSA-rwrf-2pqf-9j8j",
  "modified": "2026-10-27T21:05:29Z",
  "published": "2026-10-27T21:05:29Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/modelcontextprotocol/python-sdk/security/advisories/GHSA-rwrf-2pqf-9j8j"
    },
    {
      "type": "WEB",
      "url": "https://github.com/modelcontextprotocol/python-sdk/pull/3394"
    },
    {
      "type": "WEB",
      "url": "https://github.com/modelcontextprotocol/python-sdk/pull/3396"
    },
    {
      "type": "WEB",
      "url": "https://github.com/modelcontextprotocol/python-sdk/commit/5ebdfed4060d8e5b11b8e2348f86bad4d1fb8836"
    },
    {
      "type": "WEB",
      "url": "https://github.com/modelcontextprotocol/python-sdk/commit/6705402e246dc4eb1fcdf4d9902b78d3c9c36e1b"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/modelcontextprotocol/python-sdk"
    },
    {
      "type": "WEB",
      "url": "https://github.com/modelcontextprotocol/python-sdk/releases/tag/v1.30.0"
    },
    {
      "type": "WEB",
      "url": "https://github.com/modelcontextprotocol/python-sdk/releases/tag/v2.2.0"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:N/I:N/A:H",
      "type": "CVSS_V3"
    }
  ],
  "summary": "Model Context Protocol Python SDK: Client fetched server-chosen $ref URLs while validating tool results against output schemas"
}



Log in or create an account to share your comment.




Tags
Taxonomy of the tags.




Drag nodes to arrange them. Double-click a related vulnerability to open it, or a source to visit its link. The Physics button on the left tunes the layout and the simulation; the legend at the bottom left hides or shows nodes by kind and by sighting type.

Forecast uses a logistic model when the trend is rising, or an exponential decay model when the trend is falling. Fitted via linearized least squares.

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.
Vulnerability timeline

Sightings and recorded dates, grouped by day · UTC

JSON

Loading timeline…




    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.