Vulnerability-Lookup

CVE-2026-45312 (GCVE-0-2026-45312)

Vulnerability from cvelistv5 – Published: 2026-05-29 12:24 – Updated: 2026-06-02 01:02

Title

RAGFlow: Server-Side Template Injection in Prompt Generator leads to Remote Code Execution

Summary

RAGFlow is an open-source RAG (Retrieval-Augmented Generation) engine. In 0.24.0 and earlier, a Jinja2 template injection in the prompt generator (rag/prompts/generator.py) allows any authenticated user to execute arbitrary OS commands on the server. Any normal user can register, create a Canvas workflow with a DuckDuckGo + LLM component chain, and trigger the SSTI.

Severity

9.9 (Critical)


                        
                          CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H

SSVC

Exploitation: poc Automatable: no Technical Impact: total

CISA Coordinator · CISA-ADP (v2.0.3)

Decision recorded 2026-06-02 01:01 UTC

CWE

CWE-1336 - Improper Neutralization of Special Elements Used in a Template Engine

Assigner

GitHub_M mitre root

References

1 reference

URL	Tags
https://github.com/infiniflow/ragflow/security/ad…	x_refsource_CONFIRM

Impacted products

1 product

Vendor	Product	Version	CPE status
infiniflow	ragflow	Affected: <= 0.24.0	guessed

Show details on NVD website

JSON

To clipboard

{
  "containers": {
    "adp": [
      {
        "metrics": [
          {
            "other": {
              "content": {
                "id": "CVE-2026-45312",
                "options": [
                  {
                    "Exploitation": "poc"
                  },
                  {
                    "Automatable": "no"
                  },
                  {
                    "Technical Impact": "total"
                  }
                ],
                "role": "CISA Coordinator",
                "timestamp": "2026-06-02T01:01:26.376594Z",
                "version": "2.0.3"
              },
              "type": "ssvc"
            }
          }
        ],
        "providerMetadata": {
          "dateUpdated": "2026-06-02T01:02:42.683Z",
          "orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
          "shortName": "CISA-ADP"
        },
        "references": [
          {
            "tags": [
              "exploit"
            ],
            "url": "https://github.com/infiniflow/ragflow/security/advisories/GHSA-wpg4-h5g2-jxm6"
          }
        ],
        "title": "CISA ADP Vulnrichment"
      }
    ],
    "cna": {
      "affected": [
        {
          "product": "ragflow",
          "vendor": "infiniflow",
          "versions": [
            {
              "status": "affected",
              "version": "\u003c= 0.24.0"
            }
          ]
        }
      ],
      "descriptions": [
        {
          "lang": "en",
          "value": "RAGFlow is an open-source RAG (Retrieval-Augmented Generation) engine. In 0.24.0 and earlier, a Jinja2 template injection in the prompt generator (rag/prompts/generator.py) allows any authenticated user to execute arbitrary OS commands on the server. Any normal user can register, create a Canvas workflow with a DuckDuckGo + LLM component chain, and trigger the SSTI."
        }
      ],
      "metrics": [
        {
          "cvssV3_1": {
            "attackComplexity": "LOW",
            "attackVector": "NETWORK",
            "availabilityImpact": "HIGH",
            "baseScore": 9.9,
            "baseSeverity": "CRITICAL",
            "confidentialityImpact": "HIGH",
            "integrityImpact": "HIGH",
            "privilegesRequired": "LOW",
            "scope": "CHANGED",
            "userInteraction": "NONE",
            "vectorString": "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H",
            "version": "3.1"
          }
        }
      ],
      "problemTypes": [
        {
          "descriptions": [
            {
              "cweId": "CWE-1336",
              "description": "CWE-1336: Improper Neutralization of Special Elements Used in a Template Engine",
              "lang": "en",
              "type": "CWE"
            }
          ]
        }
      ],
      "providerMetadata": {
        "dateUpdated": "2026-05-29T12:24:07.996Z",
        "orgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
        "shortName": "GitHub_M"
      },
      "references": [
        {
          "name": "https://github.com/infiniflow/ragflow/security/advisories/GHSA-wpg4-h5g2-jxm6",
          "tags": [
            "x_refsource_CONFIRM"
          ],
          "url": "https://github.com/infiniflow/ragflow/security/advisories/GHSA-wpg4-h5g2-jxm6"
        }
      ],
      "source": {
        "advisory": "GHSA-wpg4-h5g2-jxm6",
        "discovery": "UNKNOWN"
      },
      "title": "RAGFlow: Server-Side Template Injection in Prompt Generator leads to Remote Code Execution"
    }
  },
  "cveMetadata": {
    "assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
    "assignerShortName": "GitHub_M",
    "cveId": "CVE-2026-45312",
    "datePublished": "2026-05-29T12:24:07.996Z",
    "dateReserved": "2026-05-11T20:50:30.538Z",
    "dateUpdated": "2026-06-02T01:02:42.683Z",
    "state": "PUBLISHED"
  },
  "dataType": "CVE_RECORD",
  "dataVersion": "5.2",
  "vulnerability-lookup:meta": {
    "epss": {
      "cve": "CVE-2026-45312",
      "date": "2026-09-17",
      "epss": "0.00447",
      "percentile": "0.38094"
    },
    "nvd": "{\"cve\":{\"id\":\"CVE-2026-45312\",\"sourceIdentifier\":\"security-advisories@github.com\",\"published\":\"2026-05-29T13:16:22.770\",\"lastModified\":\"2026-07-21T12:10:00.090\",\"vulnStatus\":\"Deferred\",\"cveTags\":[],\"descriptions\":[{\"lang\":\"en\",\"value\":\"RAGFlow is an open-source RAG (Retrieval-Augmented Generation) engine. In 0.24.0 and earlier, a Jinja2 template injection in the prompt generator (rag/prompts/generator.py) allows any authenticated user to execute arbitrary OS commands on the server. Any normal user can register, create a Canvas workflow with a DuckDuckGo + LLM component chain, and trigger the SSTI.\"},{\"lang\":\"es\",\"value\":\"RAGFlow es un motor RAG (Generaci\u00f3n Aumentada por Recuperaci\u00f3n) de c\u00f3digo abierto. En la versi\u00f3n 0.24.0 y anteriores, una inyecci\u00f3n de plantilla Jinja2 en el generador de prompts (rag/prompts/generator.py) permite a cualquier usuario autenticado ejecutar comandos arbitrarios del sistema operativo en el servidor. Cualquier usuario normal puede registrarse, crear un flujo de trabajo de Canvas con una cadena de componentes DuckDuckGo + LLM, y activar la SSTI.\"}],\"affected\":[{\"source\":\"security-advisories@github.com\",\"affectedData\":[{\"vendor\":\"infiniflow\",\"product\":\"ragflow\",\"versions\":[{\"version\":\"\u003c= 0.24.0\",\"status\":\"affected\"}]}]}],\"metrics\":{\"cvssMetricV31\":[{\"source\":\"security-advisories@github.com\",\"type\":\"Secondary\",\"cvssData\":{\"version\":\"3.1\",\"vectorString\":\"CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H\",\"baseScore\":9.9,\"baseSeverity\":\"CRITICAL\",\"attackVector\":\"NETWORK\",\"attackComplexity\":\"LOW\",\"privilegesRequired\":\"LOW\",\"userInteraction\":\"NONE\",\"scope\":\"CHANGED\",\"confidentialityImpact\":\"HIGH\",\"integrityImpact\":\"HIGH\",\"availabilityImpact\":\"HIGH\"},\"exploitabilityScore\":3.1,\"impactScore\":6.0}],\"ssvcV203\":[{\"source\":\"134c704f-9b21-4f2e-91b3-4a467353bcc0\",\"ssvcData\":{\"timestamp\":\"2026-06-02T01:01:26.376594Z\",\"id\":\"CVE-2026-45312\",\"options\":[{\"exploitation\":\"poc\"},{\"automatable\":\"no\"},{\"technicalImpact\":\"total\"}],\"role\":\"CISA Coordinator\",\"version\":\"2.0.3\"}}]},\"weaknesses\":[{\"source\":\"security-advisories@github.com\",\"type\":\"Secondary\",\"description\":[{\"lang\":\"en\",\"value\":\"CWE-1336\"}]}],\"references\":[{\"url\":\"https://github.com/infiniflow/ragflow/security/advisories/GHSA-wpg4-h5g2-jxm6\",\"source\":\"security-advisories@github.com\"},{\"url\":\"https://github.com/infiniflow/ragflow/security/advisories/GHSA-wpg4-h5g2-jxm6\",\"source\":\"134c704f-9b21-4f2e-91b3-4a467353bcc0\"}]}}",
    "redhat_vex": {
      "current_release_date": "2026-06-02T04:02:44+00:00",
      "cve": "CVE-2026-45312",
      "id": "CVE-2026-45312",
      "initial_release_date": "2026-01-01T00:00:00+00:00",
      "product_status:known_not_affected": "1",
      "source": "Red Hat CSAF VEX",
      "status": "final",
      "title": "RAGFlow: Server-Side Template Injection in Prompt Generator leads to Remote Code Execution",
      "url": "https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-45312.json",
      "version": "3"
    },
    "vulnrichment": {
      "containers": "{\"adp\": [{\"title\": \"CISA ADP Vulnrichment\", \"metrics\": [{\"other\": {\"type\": \"ssvc\", \"content\": {\"id\": \"CVE-2026-45312\", \"role\": \"CISA Coordinator\", \"options\": [{\"Exploitation\": \"poc\"}, {\"Automatable\": \"no\"}, {\"Technical Impact\": \"total\"}], \"version\": \"2.0.3\", \"timestamp\": \"2026-06-02T01:01:26.376594Z\"}}}], \"references\": [{\"url\": \"https://github.com/infiniflow/ragflow/security/advisories/GHSA-wpg4-h5g2-jxm6\", \"tags\": [\"exploit\"]}], \"providerMetadata\": {\"orgId\": \"134c704f-9b21-4f2e-91b3-4a467353bcc0\", \"shortName\": \"CISA-ADP\", \"dateUpdated\": \"2026-06-02T01:02:32.753Z\"}}], \"cna\": {\"title\": \"RAGFlow: Server-Side Template Injection in Prompt Generator leads to Remote Code Execution\", \"source\": {\"advisory\": \"GHSA-wpg4-h5g2-jxm6\", \"discovery\": \"UNKNOWN\"}, \"metrics\": [{\"cvssV3_1\": {\"scope\": \"CHANGED\", \"version\": \"3.1\", \"baseScore\": 9.9, \"attackVector\": \"NETWORK\", \"baseSeverity\": \"CRITICAL\", \"vectorString\": \"CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H\", \"integrityImpact\": \"HIGH\", \"userInteraction\": \"NONE\", \"attackComplexity\": \"LOW\", \"availabilityImpact\": \"HIGH\", \"privilegesRequired\": \"LOW\", \"confidentialityImpact\": \"HIGH\"}}], \"affected\": [{\"vendor\": \"infiniflow\", \"product\": \"ragflow\", \"versions\": [{\"status\": \"affected\", \"version\": \"\u003c= 0.24.0\"}]}], \"references\": [{\"url\": \"https://github.com/infiniflow/ragflow/security/advisories/GHSA-wpg4-h5g2-jxm6\", \"name\": \"https://github.com/infiniflow/ragflow/security/advisories/GHSA-wpg4-h5g2-jxm6\", \"tags\": [\"x_refsource_CONFIRM\"]}], \"descriptions\": [{\"lang\": \"en\", \"value\": \"RAGFlow is an open-source RAG (Retrieval-Augmented Generation) engine. In 0.24.0 and earlier, a Jinja2 template injection in the prompt generator (rag/prompts/generator.py) allows any authenticated user to execute arbitrary OS commands on the server. Any normal user can register, create a Canvas workflow with a DuckDuckGo + LLM component chain, and trigger the SSTI.\"}], \"problemTypes\": [{\"descriptions\": [{\"lang\": \"en\", \"type\": \"CWE\", \"cweId\": \"CWE-1336\", \"description\": \"CWE-1336: Improper Neutralization of Special Elements Used in a Template Engine\"}]}], \"providerMetadata\": {\"orgId\": \"a0819718-46f1-4df5-94e2-005712e83aaa\", \"shortName\": \"GitHub_M\", \"dateUpdated\": \"2026-05-29T12:24:07.996Z\"}}}",
      "cveMetadata": "{\"cveId\": \"CVE-2026-45312\", \"state\": \"PUBLISHED\", \"dateUpdated\": \"2026-06-02T01:02:42.683Z\", \"dateReserved\": \"2026-05-11T20:50:30.538Z\", \"assignerOrgId\": \"a0819718-46f1-4df5-94e2-005712e83aaa\", \"datePublished\": \"2026-05-29T12:24:07.996Z\", \"assignerShortName\": \"GitHub_M\"}",
      "dataType": "CVE_RECORD",
      "dataVersion": "5.2"
    }
  }
}

FKIE_CVE-2026-45312

Vulnerability from fkie_nvd - Published: 2026-05-29 13:16 - Updated: 2026-07-21 12:10

Severity

9.9 (Critical) - CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H

Summary

References

	URL	Tags
	security-advisories@github.com	https://github.com/infiniflow/ragflow/security/advisories/GHSA-wpg4-h5g2-jxm6
	134c704f-9b21-4f2e-91b3-4a467353bcc0	https://github.com/infiniflow/ragflow/security/advisories/GHSA-wpg4-h5g2-jxm6

Impacted products

	Vendor	Product	Version

JSON

To clipboard

{
  "affected": [
    {
      "affectedData": [
        {
          "product": "ragflow",
          "vendor": "infiniflow",
          "versions": [
            {
              "status": "affected",
              "version": "\u003c= 0.24.0"
            }
          ]
        }
      ],
      "source": "security-advisories@github.com"
    }
  ],
  "cveTags": [],
  "descriptions": [
    {
      "lang": "en",
      "value": "RAGFlow is an open-source RAG (Retrieval-Augmented Generation) engine. In 0.24.0 and earlier, a Jinja2 template injection in the prompt generator (rag/prompts/generator.py) allows any authenticated user to execute arbitrary OS commands on the server. Any normal user can register, create a Canvas workflow with a DuckDuckGo + LLM component chain, and trigger the SSTI."
    },
    {
      "lang": "es",
      "value": "RAGFlow es un motor RAG (Generaci\u00f3n Aumentada por Recuperaci\u00f3n) de c\u00f3digo abierto. En la versi\u00f3n 0.24.0 y anteriores, una inyecci\u00f3n de plantilla Jinja2 en el generador de prompts (rag/prompts/generator.py) permite a cualquier usuario autenticado ejecutar comandos arbitrarios del sistema operativo en el servidor. Cualquier usuario normal puede registrarse, crear un flujo de trabajo de Canvas con una cadena de componentes DuckDuckGo + LLM, y activar la SSTI."
    }
  ],
  "id": "CVE-2026-45312",
  "lastModified": "2026-07-21T12:10:00.090",
  "metrics": {
    "cvssMetricV31": [
      {
        "cvssData": {
          "attackComplexity": "LOW",
          "attackVector": "NETWORK",
          "availabilityImpact": "HIGH",
          "baseScore": 9.9,
          "baseSeverity": "CRITICAL",
          "confidentialityImpact": "HIGH",
          "integrityImpact": "HIGH",
          "privilegesRequired": "LOW",
          "scope": "CHANGED",
          "userInteraction": "NONE",
          "vectorString": "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H",
          "version": "3.1"
        },
        "exploitabilityScore": 3.1,
        "impactScore": 6.0,
        "source": "security-advisories@github.com",
        "type": "Secondary"
      }
    ],
    "ssvcV203": [
      {
        "source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
        "ssvcData": {
          "id": "CVE-2026-45312",
          "options": [
            {
              "exploitation": "poc"
            },
            {
              "automatable": "no"
            },
            {
              "technicalImpact": "total"
            }
          ],
          "role": "CISA Coordinator",
          "timestamp": "2026-06-02T01:01:26.376594Z",
          "version": "2.0.3"
        }
      }
    ]
  },
  "published": "2026-05-29T13:16:22.770",
  "references": [
    {
      "source": "security-advisories@github.com",
      "url": "https://github.com/infiniflow/ragflow/security/advisories/GHSA-wpg4-h5g2-jxm6"
    },
    {
      "source": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
      "url": "https://github.com/infiniflow/ragflow/security/advisories/GHSA-wpg4-h5g2-jxm6"
    }
  ],
  "sourceIdentifier": "security-advisories@github.com",
  "vulnStatus": "Deferred",
  "weaknesses": [
    {
      "description": [
        {
          "lang": "en",
          "value": "CWE-1336"
        }
      ],
      "source": "security-advisories@github.com",
      "type": "Secondary"
    }
  ]
}

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.

Detection rules are retrieved from Rulezet.

The MITRE ATT&CK techniques below are AI-generated suggestions, inferred from the description of the vulnerability by the CIRCL/vulnerability-attack-technique-classification-roberta-base model, served locally by ML-Gateway. They have not been verified by an analyst and are provided for guidance only.
The approach is described in our paper Mapping CVEs to MITRE ATT&CK Techniques: A Curated Gold-Set Classifier and the Limits of LLM-Assisted Label Expansion.
Browse all ATT&CK techniques and the vulnerabilities related to each.

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.