GHSA-53WG-R69P-V3R7

Vulnerability from github – Published: 2026-01-16 21:09 – Updated: 2026-01-21 16:20
VLAI
Summary
GraphQL Modules has a Race Condition issue
Details

Summary

Originally reported as an issue #2613 but should be elevated to a security issue as the ExecutionContext is often used to pass authentication tokens from incoming requests to services loading data from backend APIs.

Details

When 2 or more parallel requests are made which trigger the same service, the context of the requests is mixed up in the service when the context is injected via @ExecutionContext()

PoC

In a new project/folder, create and install the following package.json:

{
  "name": "GHSA-53wg-r69p-v3r7",
  "scripts": {
    "test": "jest"
  },
  "dependencies": {
    "graphql-modules": "2.4.0"
  },
  "devDependencies": {
    "@babel/plugin-proposal-class-properties": "^7.18.6",
    "@babel/plugin-proposal-decorators": "^7.28.6",
    "babel-plugin-parameter-decorator": "^1.0.16",
    "jest": "^29.7.0",
    "reflect-metadata": "^0.2.2"
  }
}

with:

npm i

configure babel.config.json using:

{
  "plugins": [
    ["@babel/plugin-proposal-decorators", { "legacy": true }],
    "babel-plugin-parameter-decorator",
    "@babel/plugin-proposal-class-properties"
  ]
}

then write the following test GHSA-53wg-r69p-v3r7.spec.ts:

require("reflect-metadata");
const {
  createApplication,
  createModule,
  Injectable,
  Scope,
  ExecutionContext,
  gql,
  testkit,
} = require("graphql-modules");

test("accessing a singleton provider context during another asynchronous execution", async () => {
  @Injectable({ scope: Scope.Singleton })
  class IdentifierProvider {
    @ExecutionContext()
    context;

    getId() {
      return this.context.identifier;
    }
  }

  const { promise: gettingBefore, resolve: gotBefore } = createDeferred();

  const { promise: waitForGettingAfter, resolve: getAfter } = createDeferred();

  const mod = createModule({
    id: "mod",
    providers: [IdentifierProvider],
    typeDefs: gql`
      type Query {
        getAsyncIdentifiers: Identifiers!
      }

      type Identifiers {
        before: String!
        after: String!
      }
    `,
    resolvers: {
      Query: {
        async getAsyncIdentifiers(_0, _1, context) {
          const before = context.injector.get(IdentifierProvider).getId();
          gotBefore();
          await waitForGettingAfter;
          const after = context.injector.get(IdentifierProvider).getId();
          return { before, after };
        },
      },
    },
  });

  const app = createApplication({
    modules: [mod],
  });

  const document = gql`
    {
      getAsyncIdentifiers {
        before
        after
      }
    }
  `;

  const firstResult$ = testkit.execute(app, {
    contextValue: {
      identifier: "first",
    },
    document,
  });

  await gettingBefore;

  const secondResult$ = testkit.execute(app, {
    contextValue: {
      identifier: "second",
    },
    document,
  });

  getAfter();

  await expect(firstResult$).resolves.toEqual({
    data: {
      getAsyncIdentifiers: {
        before: "first",
        after: "first",
      },
    },
  });

  await expect(secondResult$).resolves.toEqual({
    data: {
      getAsyncIdentifiers: {
        before: "second",
        after: "second",
      },
    },
  });
});

function createDeferred() {
  let resolve, reject;
  const promise = new Promise((res, rej) => {
    resolve = res;
    reject = rej;
  });
  return {
    promise,
    resolve,
    reject,
  };
}

and execute using:

npm test

Your project tree should look like this:

GHSA-53wg-r69p-v3r7
    package.json
    package-lock.json
    babel.config.json
    GHSA-53wg-r69p-v3r7.spec.js

Expected vs. Actual Outcome

- Expected  - 1
+ Received  + 1

  Object {
    "data": Object {
      "getAsyncIdentifiers": Object {
-       "after": "first",
+       "after": "second",
        "before": "first",
      },
    },
  }

Impact

Any application that uses services that inject the context using @ExecutionContext() from a singleton provider are at risk. The more traffic an application has, the higher the chance for parallel requests, the higher the risk.

Show details on source website

{
  "affected": [
    {
      "package": {
        "ecosystem": "npm",
        "name": "graphql-modules"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "2.2.1"
            },
            {
              "fixed": "2.4.1"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    },
    {
      "package": {
        "ecosystem": "npm",
        "name": "graphql-modules"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "3.0.0"
            },
            {
              "fixed": "3.1.1"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "aliases": [
    "CVE-2026-23735"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-362"
    ],
    "github_reviewed": true,
    "github_reviewed_at": "2026-01-16T21:09:08Z",
    "nvd_published_at": "2026-01-16T20:15:51Z",
    "severity": "HIGH"
  },
  "details": "### Summary\nOriginally reported as an issue #2613 but should be elevated to a security issue as the ExecutionContext is often used to pass authentication tokens from incoming requests to services loading data from backend APIs.\n\n### Details\nWhen 2 or more parallel requests are made which trigger the same service, the context of the requests is mixed up in the service when the context is injected via `@ExecutionContext()`\n\n### PoC\n\nIn a new project/folder, create and install the following `package.json`:\n\n```json\n{\n  \"name\": \"GHSA-53wg-r69p-v3r7\",\n  \"scripts\": {\n    \"test\": \"jest\"\n  },\n  \"dependencies\": {\n    \"graphql-modules\": \"2.4.0\"\n  },\n  \"devDependencies\": {\n    \"@babel/plugin-proposal-class-properties\": \"^7.18.6\",\n    \"@babel/plugin-proposal-decorators\": \"^7.28.6\",\n    \"babel-plugin-parameter-decorator\": \"^1.0.16\",\n    \"jest\": \"^29.7.0\",\n    \"reflect-metadata\": \"^0.2.2\"\n  }\n}\n```\n\nwith:\n\n```\nnpm i\n```\n\nconfigure `babel.config.json` using:\n\n```json\n{\n  \"plugins\": [\n    [\"@babel/plugin-proposal-decorators\", { \"legacy\": true }],\n    \"babel-plugin-parameter-decorator\",\n    \"@babel/plugin-proposal-class-properties\"\n  ]\n}\n```\n\nthen write the following test `GHSA-53wg-r69p-v3r7.spec.ts`:\n\n```js\nrequire(\"reflect-metadata\");\nconst {\n  createApplication,\n  createModule,\n  Injectable,\n  Scope,\n  ExecutionContext,\n  gql,\n  testkit,\n} = require(\"graphql-modules\");\n\ntest(\"accessing a singleton provider context during another asynchronous execution\", async () =\u003e {\n  @Injectable({ scope: Scope.Singleton })\n  class IdentifierProvider {\n    @ExecutionContext()\n    context;\n\n    getId() {\n      return this.context.identifier;\n    }\n  }\n\n  const { promise: gettingBefore, resolve: gotBefore } = createDeferred();\n\n  const { promise: waitForGettingAfter, resolve: getAfter } = createDeferred();\n\n  const mod = createModule({\n    id: \"mod\",\n    providers: [IdentifierProvider],\n    typeDefs: gql`\n      type Query {\n        getAsyncIdentifiers: Identifiers!\n      }\n\n      type Identifiers {\n        before: String!\n        after: String!\n      }\n    `,\n    resolvers: {\n      Query: {\n        async getAsyncIdentifiers(_0, _1, context) {\n          const before = context.injector.get(IdentifierProvider).getId();\n          gotBefore();\n          await waitForGettingAfter;\n          const after = context.injector.get(IdentifierProvider).getId();\n          return { before, after };\n        },\n      },\n    },\n  });\n\n  const app = createApplication({\n    modules: [mod],\n  });\n\n  const document = gql`\n    {\n      getAsyncIdentifiers {\n        before\n        after\n      }\n    }\n  `;\n\n  const firstResult$ = testkit.execute(app, {\n    contextValue: {\n      identifier: \"first\",\n    },\n    document,\n  });\n\n  await gettingBefore;\n\n  const secondResult$ = testkit.execute(app, {\n    contextValue: {\n      identifier: \"second\",\n    },\n    document,\n  });\n\n  getAfter();\n\n  await expect(firstResult$).resolves.toEqual({\n    data: {\n      getAsyncIdentifiers: {\n        before: \"first\",\n        after: \"first\",\n      },\n    },\n  });\n\n  await expect(secondResult$).resolves.toEqual({\n    data: {\n      getAsyncIdentifiers: {\n        before: \"second\",\n        after: \"second\",\n      },\n    },\n  });\n});\n\nfunction createDeferred() {\n  let resolve, reject;\n  const promise = new Promise((res, rej) =\u003e {\n    resolve = res;\n    reject = rej;\n  });\n  return {\n    promise,\n    resolve,\n    reject,\n  };\n}\n```\n\nand execute using:\n\n```\nnpm test\n```\n\nYour project tree should look like this:\n\n```\nGHSA-53wg-r69p-v3r7\n    package.json\n    package-lock.json\n    babel.config.json\n    GHSA-53wg-r69p-v3r7.spec.js\n```\n\n#### Expected vs. Actual Outcome\n\n```diff\n- Expected  - 1\n+ Received  + 1\n\n  Object {\n    \"data\": Object {\n      \"getAsyncIdentifiers\": Object {\n-       \"after\": \"first\",\n+       \"after\": \"second\",\n        \"before\": \"first\",\n      },\n    },\n  }\n```\n\n### Impact\n\nAny application that uses services that inject the context using `@ExecutionContext()` from a singleton provider are at risk. The more traffic an application has, the higher the chance for parallel requests, the higher the risk.",
  "id": "GHSA-53wg-r69p-v3r7",
  "modified": "2026-01-21T16:20:01Z",
  "published": "2026-01-16T21:09:08Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/graphql-hive/graphql-modules/security/advisories/GHSA-53wg-r69p-v3r7"
    },
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-23735"
    },
    {
      "type": "WEB",
      "url": "https://github.com/graphql-hive/graphql-modules/issues/2613"
    },
    {
      "type": "WEB",
      "url": "https://github.com/graphql-hive/graphql-modules/pull/2521"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/graphql-hive/graphql-modules"
    },
    {
      "type": "WEB",
      "url": "https://github.com/graphql-hive/graphql-modules/releases/tag/release-1768575025568"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:H/VA:N/SC:N/SI:N/SA:N",
      "type": "CVSS_V4"
    }
  ],
  "summary": "GraphQL Modules has a Race Condition issue"
}



Log in or create an account to share your comment.




Tags
Taxonomy of the tags.


Loading…

Loading…

Loading…

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.

Loading…

Loading…

Loading…

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.


Loading…