CVE-2025-12695 (GCVE-0-2025-12695)
Vulnerability from cvelistv5
Published
2025-11-04 13:24
Modified
2025-11-04 14:19
Severity ?
5.9 (Medium) - CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:N/A:N
CWE
  • CWE-653 - Improper Isolation or Compartmentalization
Summary
The overly permissive sandbox configuration in DSPy allows attackers to steal sensitive files in cases when users build an AI agent which consumes user input and uses the “PythonInterpreter” class.
References
Impacted products
Vendor Product Version
Version: 0
 Create a notification for this product.
Show details on NVD website

To clipboard 

{
  "containers": {
    "adp": [
      {
        "metrics": [
          {
            "other": {
              "content": {
                "id": "CVE-2025-12695",
                "options": [
                  {
                    "Exploitation": "poc"
                  },
                  {
                    "Automatable": "yes"
                  },
                  {
                    "Technical Impact": "partial"
                  }
                ],
                "role": "CISA Coordinator",
                "timestamp": "2025-11-04T14:19:21.124251Z",
                "version": "2.0.3"
              },
              "type": "ssvc"
            }
          }
        ],
        "providerMetadata": {
          "dateUpdated": "2025-11-04T14:19:33.945Z",
          "orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
          "shortName": "CISA-ADP"
        },
        "title": "CISA ADP Vulnrichment"
      }
    ],
    "cna": {
      "affected": [
        {
          "collectionURL": "https://pypi.org/project/pip",
          "packageName": "dspy",
          "versions": [
            {
              "status": "affected",
              "version": "0",
              "versionType": "python"
            }
          ]
        }
      ],
      "descriptions": [
        {
          "lang": "en",
          "supportingMedia": [
            {
              "base64": false,
              "type": "text/html",
              "value": "\u003cp\u003eThe overly permissive sandbox configuration in DSPy allows attackers to steal sensitive files in cases when users build an AI agent which consumes user input and uses the \u201cPythonInterpreter\u201d class.\u003c/p\u003e"
            }
          ],
          "value": "The overly permissive sandbox configuration in DSPy allows attackers to steal sensitive files in cases when users build an AI agent which consumes user input and uses the \u201cPythonInterpreter\u201d class."
        }
      ],
      "metrics": [
        {
          "cvssV3_1": {
            "attackComplexity": "HIGH",
            "attackVector": "NETWORK",
            "availabilityImpact": "NONE",
            "baseScore": 5.9,
            "baseSeverity": "MEDIUM",
            "confidentialityImpact": "HIGH",
            "integrityImpact": "NONE",
            "privilegesRequired": "NONE",
            "scope": "UNCHANGED",
            "userInteraction": "NONE",
            "vectorString": "CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:N/A:N",
            "version": "3.1"
          },
          "format": "CVSS",
          "scenarios": [
            {
              "lang": "en",
              "value": "GENERAL"
            }
          ]
        }
      ],
      "problemTypes": [
        {
          "descriptions": [
            {
              "cweId": "CWE-653",
              "description": "CWE-653 Improper Isolation or Compartmentalization",
              "lang": "en",
              "type": "CWE"
            }
          ]
        }
      ],
      "providerMetadata": {
        "dateUpdated": "2025-11-04T13:24:42.358Z",
        "orgId": "48a46f29-ae42-4e1d-90dd-c1676c1e5e6d",
        "shortName": "JFROG"
      },
      "references": [
        {
          "tags": [
            "third-party-advisory"
          ],
          "url": "https://research.jfrog.com/vulnerabilities/dspy-sandbox-escape-arbitrary-file-read-jfsa-2025-001495652/"
        }
      ],
      "source": {
        "discovery": "EXTERNAL"
      },
      "title": "Insecure configuration in DSPy lead to arbitrary file read when running untrusted code inside the sandbox"
    }
  },
  "cveMetadata": {
    "assignerOrgId": "48a46f29-ae42-4e1d-90dd-c1676c1e5e6d",
    "assignerShortName": "JFROG",
    "cveId": "CVE-2025-12695",
    "datePublished": "2025-11-04T13:24:42.358Z",
    "dateReserved": "2025-11-04T13:05:46.639Z",
    "dateUpdated": "2025-11-04T14:19:33.945Z",
    "state": "PUBLISHED"
  },
  "dataType": "CVE_RECORD",
  "dataVersion": "5.2",
  "vulnerability-lookup:meta": {
    "nvd": "{\"cve\":{\"id\":\"CVE-2025-12695\",\"sourceIdentifier\":\"reefs@jfrog.com\",\"published\":\"2025-11-04T14:15:34.087\",\"lastModified\":\"2025-11-04T15:40:45.533\",\"vulnStatus\":\"Awaiting Analysis\",\"cveTags\":[],\"descriptions\":[{\"lang\":\"en\",\"value\":\"The overly permissive sandbox configuration in DSPy allows attackers to steal sensitive files in cases when users build an AI agent which consumes user input and uses the \u201cPythonInterpreter\u201d class.\"}],\"metrics\":{\"cvssMetricV31\":[{\"source\":\"reefs@jfrog.com\",\"type\":\"Secondary\",\"cvssData\":{\"version\":\"3.1\",\"vectorString\":\"CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:N/A:N\",\"baseScore\":5.9,\"baseSeverity\":\"MEDIUM\",\"attackVector\":\"NETWORK\",\"attackComplexity\":\"HIGH\",\"privilegesRequired\":\"NONE\",\"userInteraction\":\"NONE\",\"scope\":\"UNCHANGED\",\"confidentialityImpact\":\"HIGH\",\"integrityImpact\":\"NONE\",\"availabilityImpact\":\"NONE\"},\"exploitabilityScore\":2.2,\"impactScore\":3.6}]},\"weaknesses\":[{\"source\":\"reefs@jfrog.com\",\"type\":\"Primary\",\"description\":[{\"lang\":\"en\",\"value\":\"CWE-653\"}]}],\"references\":[{\"url\":\"https://research.jfrog.com/vulnerabilities/dspy-sandbox-escape-arbitrary-file-read-jfsa-2025-001495652/\",\"source\":\"reefs@jfrog.com\"}]}}",
    "vulnrichment": {
      "containers": "{\"adp\": [{\"title\": \"CISA ADP Vulnrichment\", \"metrics\": [{\"other\": {\"type\": \"ssvc\", \"content\": {\"id\": \"CVE-2025-12695\", \"role\": \"CISA Coordinator\", \"options\": [{\"Exploitation\": \"poc\"}, {\"Automatable\": \"yes\"}, {\"Technical Impact\": \"partial\"}], \"version\": \"2.0.3\", \"timestamp\": \"2025-11-04T14:19:21.124251Z\"}}}], \"providerMetadata\": {\"orgId\": \"134c704f-9b21-4f2e-91b3-4a467353bcc0\", \"shortName\": \"CISA-ADP\", \"dateUpdated\": \"2025-11-04T14:19:26.090Z\"}}], \"cna\": {\"title\": \"Insecure configuration in DSPy lead to arbitrary file read when running untrusted code inside the sandbox\", \"source\": {\"discovery\": \"EXTERNAL\"}, \"metrics\": [{\"format\": \"CVSS\", \"cvssV3_1\": {\"scope\": \"UNCHANGED\", \"version\": \"3.1\", \"baseScore\": 5.9, \"attackVector\": \"NETWORK\", \"baseSeverity\": \"MEDIUM\", \"vectorString\": \"CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:N/A:N\", \"integrityImpact\": \"NONE\", \"userInteraction\": \"NONE\", \"attackComplexity\": \"HIGH\", \"availabilityImpact\": \"NONE\", \"privilegesRequired\": \"NONE\", \"confidentialityImpact\": \"HIGH\"}, \"scenarios\": [{\"lang\": \"en\", \"value\": \"GENERAL\"}]}], \"affected\": [{\"versions\": [{\"status\": \"affected\", \"version\": \"0\", \"versionType\": \"python\"}], \"packageName\": \"dspy\", \"collectionURL\": \"https://pypi.org/project/pip\"}], \"references\": [{\"url\": \"https://research.jfrog.com/vulnerabilities/dspy-sandbox-escape-arbitrary-file-read-jfsa-2025-001495652/\", \"tags\": [\"third-party-advisory\"]}], \"descriptions\": [{\"lang\": \"en\", \"value\": \"The overly permissive sandbox configuration in DSPy allows attackers to steal sensitive files in cases when users build an AI agent which consumes user input and uses the \\u201cPythonInterpreter\\u201d class.\", \"supportingMedia\": [{\"type\": \"text/html\", \"value\": \"\u003cp\u003eThe overly permissive sandbox configuration in DSPy allows attackers to steal sensitive files in cases when users build an AI agent which consumes user input and uses the \\u201cPythonInterpreter\\u201d class.\u003c/p\u003e\", \"base64\": false}]}], \"problemTypes\": [{\"descriptions\": [{\"lang\": \"en\", \"type\": \"CWE\", \"cweId\": \"CWE-653\", \"description\": \"CWE-653 Improper Isolation or Compartmentalization\"}]}], \"providerMetadata\": {\"orgId\": \"48a46f29-ae42-4e1d-90dd-c1676c1e5e6d\", \"shortName\": \"JFROG\", \"dateUpdated\": \"2025-11-04T13:24:42.358Z\"}}}",
      "cveMetadata": "{\"cveId\": \"CVE-2025-12695\", \"state\": \"PUBLISHED\", \"dateUpdated\": \"2025-11-04T14:19:33.945Z\", \"dateReserved\": \"2025-11-04T13:05:46.639Z\", \"assignerOrgId\": \"48a46f29-ae42-4e1d-90dd-c1676c1e5e6d\", \"datePublished\": \"2025-11-04T13:24:42.358Z\", \"assignerShortName\": \"JFROG\"}",
      "dataType": "CVE_RECORD",
      "dataVersion": "5.2"
    }
  }
}


Log in or create an account to share your comment.




Tags
Taxonomy of the tags.




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.