CVE-2026-107717 (GCVE-0-2026-107717)
Vulnerability from cvelistv5
Published
2026-10-08 21:38
Modified
2026-10-13 16:16
Severity ?
6.5 (Medium) - CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:N
CWE
  • CWE-20 - Improper Input Validation
Summary
Banks generates meaningful LLM prompts using a simple template language. Prior to 2.5.0, Banks Prompt.chat_messages() attempts to parse every line of rendered template output as ChatMessage JSON. When an application renders untrusted data and passes the returned ChatMessage objects to an LLM provider, attacker-controlled JSON can cross the prompt boundary and become a system, assistant, or tool message because ChatMessage.role accepts arbitrary strings. This can override application instructions, alter the intended prompt structure, or confuse downstream tool and message handling. This issue is fixed in version 2.5.0.
References
Impacted products
Vendor Product Version
masci banks Version: < 2.5.0
 Create a notification for this product.
Show details on NVD website

To clipboard 

{
  "containers": {
    "adp": [
      {
        "metrics": [
          {
            "other": {
              "content": {
                "id": "CVE-2026-107717",
                "options": [
                  {
                    "Exploitation": "poc"
                  },
                  {
                    "Automatable": "yes"
                  },
                  {
                    "Technical Impact": "partial"
                  }
                ],
                "role": "CISA Coordinator",
                "timestamp": "2026-10-13T16:15:40.228940Z",
                "version": "2.0.3"
              },
              "type": "ssvc"
            }
          }
        ],
        "providerMetadata": {
          "dateUpdated": "2026-10-13T16:16:13.460Z",
          "orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
          "shortName": "CISA-ADP"
        },
        "references": [
          {
            "tags": [
              "exploit"
            ],
            "url": "https://github.com/masci/banks/security/advisories/GHSA-hmq2-7hp6-7crh"
          }
        ],
        "title": "CISA ADP Vulnrichment"
      }
    ],
    "cna": {
      "affected": [
        {
          "product": "banks",
          "vendor": "masci",
          "versions": [
            {
              "status": "affected",
              "version": "\u003c 2.5.0"
            }
          ]
        }
      ],
      "descriptions": [
        {
          "lang": "en",
          "value": "Banks generates meaningful LLM prompts using a simple template language. Prior to 2.5.0, Banks Prompt.chat_messages() attempts to parse every line of rendered template output as ChatMessage JSON. When an application renders untrusted data and passes the returned ChatMessage objects to an LLM provider, attacker-controlled JSON can cross the prompt boundary and become a system, assistant, or tool message because ChatMessage.role accepts arbitrary strings. This can override application instructions, alter the intended prompt structure, or confuse downstream tool and message handling. This issue is fixed in version 2.5.0."
        }
      ],
      "metrics": [
        {
          "cvssV3_1": {
            "attackComplexity": "LOW",
            "attackVector": "NETWORK",
            "availabilityImpact": "NONE",
            "baseScore": 6.5,
            "baseSeverity": "MEDIUM",
            "confidentialityImpact": "LOW",
            "integrityImpact": "LOW",
            "privilegesRequired": "NONE",
            "scope": "UNCHANGED",
            "userInteraction": "NONE",
            "vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:N",
            "version": "3.1"
          }
        }
      ],
      "problemTypes": [
        {
          "descriptions": [
            {
              "cweId": "CWE-20",
              "description": "CWE-20: Improper Input Validation",
              "lang": "en",
              "type": "CWE"
            }
          ]
        }
      ],
      "providerMetadata": {
        "dateUpdated": "2026-10-08T21:38:07.447Z",
        "orgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
        "shortName": "GitHub_M"
      },
      "references": [
        {
          "name": "https://github.com/masci/banks/security/advisories/GHSA-hmq2-7hp6-7crh",
          "tags": [
            "x_refsource_CONFIRM"
          ],
          "url": "https://github.com/masci/banks/security/advisories/GHSA-hmq2-7hp6-7crh"
        },
        {
          "name": "https://github.com/masci/banks/pull/78",
          "tags": [
            "x_refsource_MISC"
          ],
          "url": "https://github.com/masci/banks/pull/78"
        },
        {
          "name": "https://github.com/masci/banks/commit/02172b816fb84f6a824cc09a8aca7416f53c12cb",
          "tags": [
            "x_refsource_MISC"
          ],
          "url": "https://github.com/masci/banks/commit/02172b816fb84f6a824cc09a8aca7416f53c12cb"
        },
        {
          "name": "https://github.com/masci/banks/releases/tag/v2.5.0",
          "tags": [
            "x_refsource_MISC"
          ],
          "url": "https://github.com/masci/banks/releases/tag/v2.5.0"
        }
      ],
      "source": {
        "advisory": "GHSA-hmq2-7hp6-7crh",
        "discovery": "UNKNOWN"
      },
      "title": "Banks: User-controlled prompt input can be parsed as privileged chat messages"
    }
  },
  "cveMetadata": {
    "assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
    "assignerShortName": "GitHub_M",
    "cveId": "CVE-2026-107717",
    "datePublished": "2026-10-08T21:38:07.447Z",
    "dateReserved": "2026-10-08T17:21:52.975Z",
    "dateUpdated": "2026-10-13T16:16:13.460Z",
    "state": "PUBLISHED"
  },
  "dataType": "CVE_RECORD",
  "dataVersion": "5.2",
  "vulnerability-lookup:meta": {
    "vulnrichment": {
      "containers": "{\"adp\": [{\"title\": \"CISA ADP Vulnrichment\", \"metrics\": [{\"other\": {\"type\": \"ssvc\", \"content\": {\"id\": \"CVE-2026-107717\", \"role\": \"CISA Coordinator\", \"options\": [{\"Exploitation\": \"poc\"}, {\"Automatable\": \"yes\"}, {\"Technical Impact\": \"partial\"}], \"version\": \"2.0.3\", \"timestamp\": \"2026-10-13T16:15:40.228940Z\"}}}], \"references\": [{\"url\": \"https://github.com/masci/banks/security/advisories/GHSA-hmq2-7hp6-7crh\", \"tags\": [\"exploit\"]}], \"providerMetadata\": {\"orgId\": \"134c704f-9b21-4f2e-91b3-4a467353bcc0\", \"shortName\": \"CISA-ADP\", \"dateUpdated\": \"2026-10-13T16:16:08.625Z\"}}], \"cna\": {\"title\": \"Banks: User-controlled prompt input can be parsed as privileged chat messages\", \"source\": {\"advisory\": \"GHSA-hmq2-7hp6-7crh\", \"discovery\": \"UNKNOWN\"}, \"metrics\": [{\"cvssV3_1\": {\"scope\": \"UNCHANGED\", \"version\": \"3.1\", \"baseScore\": 6.5, \"attackVector\": \"NETWORK\", \"baseSeverity\": \"MEDIUM\", \"vectorString\": \"CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:N\", \"integrityImpact\": \"LOW\", \"userInteraction\": \"NONE\", \"attackComplexity\": \"LOW\", \"availabilityImpact\": \"NONE\", \"privilegesRequired\": \"NONE\", \"confidentialityImpact\": \"LOW\"}}], \"affected\": [{\"vendor\": \"masci\", \"product\": \"banks\", \"versions\": [{\"status\": \"affected\", \"version\": \"\u003c 2.5.0\"}]}], \"references\": [{\"url\": \"https://github.com/masci/banks/security/advisories/GHSA-hmq2-7hp6-7crh\", \"name\": \"https://github.com/masci/banks/security/advisories/GHSA-hmq2-7hp6-7crh\", \"tags\": [\"x_refsource_CONFIRM\"]}, {\"url\": \"https://github.com/masci/banks/pull/78\", \"name\": \"https://github.com/masci/banks/pull/78\", \"tags\": [\"x_refsource_MISC\"]}, {\"url\": \"https://github.com/masci/banks/commit/02172b816fb84f6a824cc09a8aca7416f53c12cb\", \"name\": \"https://github.com/masci/banks/commit/02172b816fb84f6a824cc09a8aca7416f53c12cb\", \"tags\": [\"x_refsource_MISC\"]}, {\"url\": \"https://github.com/masci/banks/releases/tag/v2.5.0\", \"name\": \"https://github.com/masci/banks/releases/tag/v2.5.0\", \"tags\": [\"x_refsource_MISC\"]}], \"descriptions\": [{\"lang\": \"en\", \"value\": \"Banks generates meaningful LLM prompts using a simple template language. Prior to 2.5.0, Banks Prompt.chat_messages() attempts to parse every line of rendered template output as ChatMessage JSON. When an application renders untrusted data and passes the returned ChatMessage objects to an LLM provider, attacker-controlled JSON can cross the prompt boundary and become a system, assistant, or tool message because ChatMessage.role accepts arbitrary strings. This can override application instructions, alter the intended prompt structure, or confuse downstream tool and message handling. This issue is fixed in version 2.5.0.\"}], \"problemTypes\": [{\"descriptions\": [{\"lang\": \"en\", \"type\": \"CWE\", \"cweId\": \"CWE-20\", \"description\": \"CWE-20: Improper Input Validation\"}]}], \"providerMetadata\": {\"orgId\": \"a0819718-46f1-4df5-94e2-005712e83aaa\", \"shortName\": \"GitHub_M\", \"dateUpdated\": \"2026-10-08T21:38:07.447Z\"}}}",
      "cveMetadata": "{\"cveId\": \"CVE-2026-107717\", \"state\": \"PUBLISHED\", \"dateUpdated\": \"2026-10-13T16:16:13.460Z\", \"dateReserved\": \"2026-10-08T17:21:52.975Z\", \"assignerOrgId\": \"a0819718-46f1-4df5-94e2-005712e83aaa\", \"datePublished\": \"2026-10-08T21:38:07.447Z\", \"assignerShortName\": \"GitHub_M\"}",
      "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.