CVE-2026-92206 (GCVE-0-2026-92206)
Vulnerability from cvelistv5
Published
2026-10-21 17:17
Modified
2026-10-30 15:35
CWE
  • CWE-470 - Use of Externally-Controlled Input to Select Classes or Code ('Unsafe Reflection')
Summary
CrewAI crewAI Framework Agent Loading Unsafe Reflection Remote Code Execution Vulnerability. This vulnerability allows remote attackers to execute arbitrary code on affected installations of CrewAI crewAI. User interaction is required to exploit this vulnerability in that the target must load a malicious agent configuration from the repository. The specific flaw exists within the load_agent_from_repository function. The process does not properly restrict a user-supplied argument before using it to import a module. An attacker can leverage this vulnerability to execute code in the context of the service account. Was ZDI-CAN-28387.
References
Impacted products
Vendor Product Version
CrewAI CrewAI Version: Latest (as of 2025-10-09)
Create a notification for this product.
Show details on NVD website


{
  "containers": {
    "adp": [
      {
        "metrics": [
          {
            "other": {
              "content": {
                "id": "CVE-2026-92206",
                "options": [
                  {
                    "Exploitation": "none"
                  },
                  {
                    "Automatable": "no"
                  },
                  {
                    "Technical Impact": "total"
                  }
                ],
                "role": "CISA Coordinator",
                "timestamp": "2026-10-30T15:34:31.928024Z",
                "version": "2.0.3"
              },
              "type": "ssvc"
            }
          }
        ],
        "providerMetadata": {
          "dateUpdated": "2026-10-30T15:35:08.028Z",
          "orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
          "shortName": "CISA-ADP"
        },
        "title": "CISA ADP Vulnrichment"
      }
    ],
    "cna": {
      "affected": [
        {
          "defaultStatus": "unknown",
          "product": "CrewAI",
          "vendor": "CrewAI",
          "versions": [
            {
              "status": "affected",
              "version": "Latest (as of 2025-10-09)"
            }
          ]
        }
      ],
      "dateAssigned": "2026-09-15T18:15:32.955Z",
      "datePublic": "2026-09-16T15:11:00.748Z",
      "descriptions": [
        {
          "lang": "en",
          "value": "CrewAI crewAI Framework Agent Loading Unsafe Reflection Remote Code Execution Vulnerability. This vulnerability allows remote attackers to execute arbitrary code on affected installations of CrewAI crewAI. User interaction is required to exploit this vulnerability in that the target must load a malicious agent configuration from the repository.\n\nThe specific flaw exists within the load_agent_from_repository function. The process does not properly restrict a user-supplied argument before using it to import a module. An attacker can leverage this vulnerability to execute code in the context of the service account. Was ZDI-CAN-28387."
        }
      ],
      "metrics": [
        {
          "cvssV3_0": {
            "baseScore": 8.8,
            "baseSeverity": "HIGH",
            "vectorString": "CVSS:3.0/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H",
            "version": "3.0"
          },
          "format": "CVSS"
        }
      ],
      "problemTypes": [
        {
          "descriptions": [
            {
              "cweId": "CWE-470",
              "description": "CWE-470: Use of Externally-Controlled Input to Select Classes or Code (\u0027Unsafe Reflection\u0027)",
              "lang": "en",
              "type": "CWE"
            }
          ]
        }
      ],
      "providerMetadata": {
        "dateUpdated": "2026-10-21T17:17:33.262Z",
        "orgId": "99f1926a-a320-47d8-bbb5-42feb611262e",
        "shortName": "zdi"
      },
      "references": [
        {
          "name": "ZDI-26-706",
          "tags": [
            "x_research-advisory"
          ],
          "url": "https://www.zerodayinitiative.com/advisories/ZDI-26-706/"
        }
      ],
      "source": {
        "lang": "en",
        "value": "Peter Girnus (@gothburz), Demeng Chen, and Brandon Niemczyk  of Trend Zero Day Initiative"
      },
      "title": "CrewAI crewAI Framework Agent Loading Unsafe Reflection Remote Code Execution Vulnerability"
    }
  },
  "cveMetadata": {
    "assignerOrgId": "99f1926a-a320-47d8-bbb5-42feb611262e",
    "assignerShortName": "zdi",
    "cveId": "CVE-2026-92206",
    "datePublished": "2026-10-21T17:17:33.262Z",
    "dateReserved": "2026-09-15T18:15:32.922Z",
    "dateUpdated": "2026-10-30T15:35:08.028Z",
    "state": "PUBLISHED"
  },
  "dataType": "CVE_RECORD",
  "dataVersion": "5.2",
  "vulnerability-lookup:meta": {
    "vulnrichment": {
      "containers": "{\"cna\": {\"providerMetadata\": {\"orgId\": \"99f1926a-a320-47d8-bbb5-42feb611262e\", \"shortName\": \"zdi\", \"dateUpdated\": \"2026-10-21T17:17:33.262Z\"}, \"title\": \"CrewAI crewAI Framework Agent Loading Unsafe Reflection Remote Code Execution Vulnerability\", \"descriptions\": [{\"lang\": \"en\", \"value\": \"CrewAI crewAI Framework Agent Loading Unsafe Reflection Remote Code Execution Vulnerability. This vulnerability allows remote attackers to execute arbitrary code on affected installations of CrewAI crewAI. User interaction is required to exploit this vulnerability in that the target must load a malicious agent configuration from the repository.\\n\\nThe specific flaw exists within the load_agent_from_repository function. The process does not properly restrict a user-supplied argument before using it to import a module. An attacker can leverage this vulnerability to execute code in the context of the service account. Was ZDI-CAN-28387.\"}], \"affected\": [{\"vendor\": \"CrewAI\", \"product\": \"CrewAI\", \"versions\": [{\"version\": \"Latest (as of 2025-10-09)\", \"status\": \"affected\"}], \"defaultStatus\": \"unknown\"}], \"problemTypes\": [{\"descriptions\": [{\"lang\": \"en\", \"cweId\": \"CWE-470\", \"description\": \"CWE-470: Use of Externally-Controlled Input to Select Classes or Code (\u0027Unsafe Reflection\u0027)\", \"type\": \"CWE\"}]}], \"references\": [{\"url\": \"https://www.zerodayinitiative.com/advisories/ZDI-26-706/\", \"name\": \"ZDI-26-706\", \"tags\": [\"x_research-advisory\"]}], \"dateAssigned\": \"2026-09-15T18:15:32.955Z\", \"datePublic\": \"2026-09-16T15:11:00.748Z\", \"source\": {\"lang\": \"en\", \"value\": \"Peter Girnus (@gothburz), Demeng Chen, and Brandon Niemczyk  of Trend Zero Day Initiative\"}, \"metrics\": [{\"format\": \"CVSS\", \"cvssV3_0\": {\"version\": \"3.0\", \"vectorString\": \"CVSS:3.0/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H\", \"baseScore\": 8.8, \"baseSeverity\": \"HIGH\"}}]}, \"adp\": [{\"title\": \"CISA ADP Vulnrichment\", \"metrics\": [{\"other\": {\"type\": \"ssvc\", \"content\": {\"id\": \"CVE-2026-92206\", \"role\": \"CISA Coordinator\", \"options\": [{\"Exploitation\": \"none\"}, {\"Automatable\": \"no\"}, {\"Technical Impact\": \"total\"}], \"version\": \"2.0.3\", \"timestamp\": \"2026-10-30T15:34:31.928024Z\"}}}], \"providerMetadata\": {\"orgId\": \"134c704f-9b21-4f2e-91b3-4a467353bcc0\", \"shortName\": \"CISA-ADP\", \"dateUpdated\": \"2026-10-30T15:35:03.588Z\"}}]}",
      "cveMetadata": "{\"cveId\": \"CVE-2026-92206\", \"assignerOrgId\": \"99f1926a-a320-47d8-bbb5-42feb611262e\", \"state\": \"PUBLISHED\", \"assignerShortName\": \"zdi\", \"dateReserved\": \"2026-09-15T18:15:32.922Z\", \"datePublished\": \"2026-10-21T17:17:33.262Z\", \"dateUpdated\": \"2026-10-30T15:35:08.028Z\"}",
      "dataType": "CVE_RECORD",
      "dataVersion": "5.2"
    }
  }
}


Log in or create an account to share your comment.




Tags
Taxonomy of the tags.


Loading…

Loading…

Loading…

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.


Loading…

Loading…