FKIE_CVE-2025-9959

Vulnerability from fkie_nvd - Published: 2025-09-03 17:15 - Updated: 2026-09-26 00:10
Severity
7.6 (High) - CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:L/I:H/A:L
Summary
Incomplete validation of dunder attributes allows an attacker to escape from the Local Python execution environment sandbox, enforced by smolagents. The attack requires a Prompt Injection in order to trick the agent to create malicious code.
References
reefs@jfrog.comhttps://github.com/huggingface/smolagents/pull/1551
reefs@jfrog.comhttps://research.jfrog.com/vulnerabilities/smolagents-local-python-sandbox-escape-jfsa-2025-001434277/
Impacted products
Vendor Product Version
To clipboard 

{
  "affected": [
    {
      "affectedData": [
        {
          "collectionURL": "https://pypi.org/project/pip",
          "packageName": "smolagents",
          "versions": [
            {
              "lessThan": "1.21.0",
              "status": "affected",
              "version": "0",
              "versionType": "python"
            }
          ]
        }
      ],
      "source": "reefs@jfrog.com"
    }
  ],
  "cveTags": [],
  "descriptions": [
    {
      "lang": "en",
      "value": "Incomplete validation of dunder attributes allows an attacker to escape from the Local Python execution environment sandbox, enforced by smolagents. The attack requires a Prompt Injection in order to trick the agent to create malicious code."
    },
    {
      "lang": "es",
      "value": "La validaci\u00f3n incompleta de atributos dunder permite a un atacante escapar de la sandbox del entorno de ejecuci\u00f3n de Python local, impuesta por smolagents. El ataque requiere una inyecci\u00f3n de prompt para enga\u00f1ar al agente y que cree c\u00f3digo malicioso."
    }
  ],
  "id": "CVE-2025-9959",
  "lastModified": "2026-09-26T00:10:00.127",
  "metrics": {
    "cvssMetricV31": [
      {
        "cvssData": {
          "attackComplexity": "LOW",
          "attackVector": "NETWORK",
          "availabilityImpact": "LOW",
          "baseScore": 7.6,
          "baseSeverity": "HIGH",
          "confidentialityImpact": "LOW",
          "integrityImpact": "HIGH",
          "privilegesRequired": "NONE",
          "scope": "UNCHANGED",
          "userInteraction": "REQUIRED",
          "vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:L/I:H/A:L",
          "version": "3.1"
        },
        "exploitabilityScore": 2.8,
        "impactScore": 4.7,
        "source": "reefs@jfrog.com",
        "type": "Secondary"
      }
    ],
    "ssvcV203": [
      {
        "source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
        "ssvcData": {
          "id": "CVE-2025-9959",
          "options": [
            {
              "exploitation": "none"
            },
            {
              "automatable": "no"
            },
            {
              "technicalImpact": "partial"
            }
          ],
          "role": "CISA Coordinator",
          "timestamp": "2025-09-03T18:48:07.608510Z",
          "version": "2.0.3"
        }
      }
    ]
  },
  "published": "2025-09-03T17:15:35.737",
  "references": [
    {
      "source": "reefs@jfrog.com",
      "url": "https://github.com/huggingface/smolagents/pull/1551"
    },
    {
      "source": "reefs@jfrog.com",
      "url": "https://research.jfrog.com/vulnerabilities/smolagents-local-python-sandbox-escape-jfsa-2025-001434277/"
    }
  ],
  "sourceIdentifier": "reefs@jfrog.com",
  "vulnStatus": "Deferred",
  "weaknesses": [
    {
      "description": [
        {
          "lang": "en",
          "value": "CWE-94"
        }
      ],
      "source": "reefs@jfrog.com",
      "type": "Secondary"
    }
  ]
}



Log in or create an account to share your comment.




Tags
Taxonomy of the tags.





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.



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.