Computer Incident Response Center Luxembourg logo
Organization

Computer Incident Response Center Luxembourg

Governmental, Luxembourg

GNA-1

The Computer Incident Response Center Luxembourg (CIRCL) is a government-driven initiative designed to provide a systematic response facility to computer security threats and incidents.


Website
https://www.circl.lu
Email
info@circl.lu
GNA identifier
GNA-1 GCVE registry Recent publications

Recent vulnerabilities

355 GCVE records assigned by this organization as GNA-1

GCVE-1-2026-20292 (CVE-2026-100190)

Vulnerability from gna-1 – Published: 2026-09-25 13:49 – Updated: 2026-09-25 13:52
VLAI
Title
Stored Cross-Site Scripting (XSS) via Crawler Capture Import in AIL Framework showDomain Page
Summary
The AIL Framework crawler splash domain page (showDomain.html) is vulnerable to stored cross-site script injection (XSS). User-supplied data originating from imported crawler captures—specifically item IDs, URLs, and screenshot file paths—was interpolated directly into inline JavaScript contexts within the HTML template. This included an onclick attribute that embedded raw screenshot and URL values into a JavaScript function call, and an inline script block that assigned a screenshot value to a JavaScript variable without escaping. An attacker with a user-role API client could craft a malicious crawler capture import containing JavaScript payloads in these fields. When any user (including privileged users) subsequently viewed the affected domain page, the injected script would execute in the victim's browser context, potentially allowing session hijacking, data exfiltration, or unauthorized actions on behalf of the victim. The vulnerability is stored in the application's data layer and triggered upon page rendering, requiring no further interaction beyond loading the domain view.
CWE
  • CWE-79 - Improper Neutralization of Input in Web Page ('Cross-site Scripting')
Assigner
CIRCL GNA GNA-1
GNA scorecard A 91/100 over 279 records in the last 180 days details
References
Impacted products
Vendor Product Version
ail project ail framework Affected: unspecified , < 7.1 (semver)
Create a notification for this product.
GCVE extensions
bcp-05-x-01
AI-assisted vulnerability information annotation
GCVE-BCP-05-X-01
Whole record AI-generated Human-reviewed GNA-1

Draft vulnerability metadata was generated from a git-format patch using an Ollama-hosted language model.

ai-computer-assisted:llm-generatedai-computer-assisted:classification
Model Source Identifier
qwen3.8:27b ollama qwen3.8:27b
bcp-05-x-02
Patch-to-vulnerability generation provenance
GCVE-BCP-05-X-02
Generator
patch2vuln.py on 2026-09-25 13:45
Model
qwen3.8:27b
Input
https://github.com/ail-project/ail-framework/commit/31376ee3d04519c898bb9b3671453a41c4117340.patch c110d16dd8ee…
Confidence
medium
Commit Subject Patch SHA-256
31376ee3d045 chg: [security] prevent stored XSS through imported capture c110d16dd8ee…
Fix summary

The fix removes all capture-derived values (item IDs, URLs, screenshot paths) from inline JavaScript contexts (onclick attributes and inline script variable assignments). These values are now stored in HTML data attributes (data-screenshot, data-url, data-item-id), which are properly HTML-escaped by the Jinja2 template engine. Screenshot click handlers are bound programmatically via addEventListener on elements with a .reload-image class, and the reload_image function reads values from the DOM element's dataset rather than receiving raw string parameters. Additionally, item IDs are URL-encoded with encodeURIComponent before being appended to link hrefs, preventing injection through the ID field.

Patch summary

In showDomain.html: (1) Replaced inline onclick="reload_image('...')" with a data-attribute-based approach using data-screenshot, data-url, and data-item-id attributes on the table cell element, plus a .reload-image CSS class. (2) Removed the inline script block that assigned screenshot, URL, and ID values directly into JavaScript variables via Jinja2 interpolation; replaced with dataset reads from the domain_screenshot img element. (3) Changed reload_image() to accept a DOM element and read values from element.dataset instead of raw string arguments. (4) Added a programmatic event binding loop using document.querySelectorAll('.reload-image').forEach() with addEventListener('click'). (5) Applied encodeURIComponent() to item IDs when constructing screenshot link hrefs. (6) Changed the href URL construction to use url_for with keyword arguments instead of string interpolation.

CVSS rationale

AV:N: Exploitable over the network via the web application. AC:L: No special race conditions or complex setup; a crafted capture import suffices. AT:N: No manipulation of the attack target required. PR:L: Requires a user-role API client to import a malicious crawler capture (authenticated, low privilege). UI:P: The stored XSS in the inline script block executes passively when the victim loads the domain page; no click or active interaction is needed for the inline script vector. VC:N/VI:N/VA:N: The vulnerable component (server-side application) is not directly compromised in confidentiality, integrity, or availability. SC:H: Scope changes to the victim's browser session, enabling theft of session tokens, cookies, and sensitive data. SI:H: Scope integrity is impacted as the attacker can perform arbitrary actions in the victim's authenticated session. SA:N: No meaningful impact on the availability of the attacker's scope.

Weakness rationale
  • CWE-79 The patch explicitly addresses stored XSS by removing user-controlled values (screenshot paths, URLs, item IDs from crawler capture imports) from inline JavaScript contexts and replacing them with properly escaped data attributes and programmatic event binding. The commit message confirms the intent: 'prevent stored XSS through imported capture metadata in showDomain.'
Attack pattern rationale
  • CAPEC-1 The patch directly remediates a stored XSS vector where attacker-controlled data from a crawler capture import is reflected into inline JavaScript without neutralization. CAPEC-1 is the canonical attack pattern for XSS. The specific sub-variant is stored (persistent) XSS via a data import mechanism, but CAPEC-1 is the closest and most appropriate mapping available in the CAPEC catalog.
Assumptions to verify
  • The UI:P (Passive) rating assumes the inline script vector (screenshot variable assignment) executes on page load without requiring a user click. The onclick vector would require UI:A, but the inline script vector is sufficient for exploitation.
  • PR:L assumes the attacker needs a user-role API client to import a crafted crawler capture, as stated in the commit message. The exact privilege level and authentication mechanism are not further specified in the patch.
  • The version boundary 'less_than 7.1' is inferred from the tag_version_boundary metadata showing 21 commits after the fix before the v7.1 tag, suggesting the fix is included in v7.1. Exact affected version ranges are not confirmed.
  • CAPEC-1 is selected as the closest available attack pattern for stored XSS; no more specific CAPEC for stored XSS via data import exists in the catalog.
  • The patch does not include test cases or additional context about whether other template files have similar inline JavaScript interpolation patterns; this advisory covers only the single file in the patch.
Model comparison

Selected qwen3.8:27b by deterministic-consensus-v1
The selected result is closest to model consensus; this heuristic does not establish factual correctness and human review remains required.

Model Score Agreement Confidence Assumptions
qwen3.8:27b 5 9 medium 5

{
  "containers": {
    "cna": {
      "affected": [
        {
          "defaultStatus": "unaffected",
          "modules": [
            "crawler/crawler_splash"
          ],
          "product": "ail framework",
          "programFiles": [
            "var/www/templates/crawler/crawler_splash/showDomain.html"
          ],
          "repo": "https://github.com/ail-project/ail-framework",
          "vendor": "ail project",
          "versions": [
            {
              "lessThan": "7.1",
              "status": "affected",
              "version": "unspecified",
              "versionType": "semver"
            }
          ]
        }
      ],
      "credits": [
        {
          "lang": "en",
          "type": "remediation developer",
          "value": "Aurelien Thirion"
        },
        {
          "lang": "en",
          "type": "finder",
          "value": "Jeroen Pinoy"
        }
      ],
      "descriptions": [
        {
          "lang": "en",
          "supportingMedia": [
            {
              "base64": false,
              "type": "text/html",
              "value": "\u003cp\u003eThe AIL Framework crawler splash domain page (showDomain.html) is vulnerable to stored cross-site script injection (XSS). User-supplied data originating from imported crawler captures\u2014specifically item IDs, URLs, and screenshot file paths\u2014was interpolated directly into inline JavaScript contexts within the HTML template. This included an onclick attribute that embedded raw screenshot and URL values into a JavaScript function call, and an inline script block that assigned a screenshot value to a JavaScript variable without escaping. An attacker with a user-role API client could craft a malicious crawler capture import containing JavaScript payloads in these fields. When any user (including privileged users) subsequently viewed the affected domain page, the injected script would execute in the victim\u0027s browser context, potentially allowing session hijacking, data exfiltration, or unauthorized actions on behalf of the victim. The vulnerability is stored in the application\u0027s data layer and triggered upon page rendering, requiring no further interaction beyond loading the domain view.\u003c/p\u003e"
            }
          ],
          "value": "The AIL Framework crawler splash domain page (showDomain.html) is vulnerable to stored cross-site script injection (XSS). User-supplied data originating from imported crawler captures\u2014specifically item IDs, URLs, and screenshot file paths\u2014was interpolated directly into inline JavaScript contexts within the HTML template. This included an onclick attribute that embedded raw screenshot and URL values into a JavaScript function call, and an inline script block that assigned a screenshot value to a JavaScript variable without escaping. An attacker with a user-role API client could craft a malicious crawler capture import containing JavaScript payloads in these fields. When any user (including privileged users) subsequently viewed the affected domain page, the injected script would execute in the victim\u0027s browser context, potentially allowing session hijacking, data exfiltration, or unauthorized actions on behalf of the victim. The vulnerability is stored in the application\u0027s data layer and triggered upon page rendering, requiring no further interaction beyond loading the domain view."
        }
      ],
      "impacts": [
        {
          "capecId": "CAPEC-1",
          "descriptions": [
            {
              "lang": "en",
              "value": "CAPEC-1 Cross Site Scripting (XSS)"
            }
          ]
        }
      ],
      "metrics": [
        {
          "cvssV4_0": {
            "Automatable": "NOT_DEFINED",
            "Recovery": "NOT_DEFINED",
            "Safety": "NOT_DEFINED",
            "attackComplexity": "LOW",
            "attackRequirements": "NONE",
            "attackVector": "NETWORK",
            "baseScore": 6.3,
            "baseSeverity": "MEDIUM",
            "privilegesRequired": "LOW",
            "providerUrgency": "NOT_DEFINED",
            "subAvailabilityImpact": "NONE",
            "subConfidentialityImpact": "HIGH",
            "subIntegrityImpact": "HIGH",
            "userInteraction": "PASSIVE",
            "valueDensity": "NOT_DEFINED",
            "vectorString": "CVSS:4.0/AV:N/AC:L/AT:N/PR:L/UI:P/VC:N/VI:N/VA:N/SC:H/SI:H/SA:N",
            "version": "4.0",
            "vulnAvailabilityImpact": "NONE",
            "vulnConfidentialityImpact": "NONE",
            "vulnIntegrityImpact": "NONE",
            "vulnerabilityResponseEffort": "NOT_DEFINED"
          },
          "format": "CVSS",
          "scenarios": [
            {
              "lang": "en",
              "value": "GENERAL"
            }
          ]
        }
      ],
      "problemTypes": [
        {
          "descriptions": [
            {
              "cweId": "CWE-79",
              "description": "CWE-79 Improper Neutralization of Input in Web Page (\u0027Cross-site Scripting\u0027)",
              "lang": "en",
              "type": "CWE"
            }
          ]
        }
      ],
      "providerMetadata": {
        "orgId": "00000000-0000-4000-9000-000000000000"
      },
      "references": [
        {
          "name": "Security patch",
          "tags": [
            "patch"
          ],
          "url": "https://github.com/ail-project/ail-framework/commit/31376ee3d04519c898bb9b3671453a41c4117340"
        }
      ],
      "solutions": [
        {
          "lang": "en",
          "supportingMedia": [
            {
              "base64": false,
              "type": "text/html",
              "value": "\u003cp\u003eThe fix removes all capture-derived values (item IDs, URLs, screenshot paths) from inline JavaScript contexts (onclick attributes and inline script variable assignments). These values are now stored in HTML data attributes (data-screenshot, data-url, data-item-id), which are properly HTML-escaped by the Jinja2 template engine. Screenshot click handlers are bound programmatically via addEventListener on elements with a .reload-image class, and the reload_image function reads values from the DOM element\u0027s dataset rather than receiving raw string parameters. Additionally, item IDs are URL-encoded with encodeURIComponent before being appended to link hrefs, preventing injection through the ID field.\u003c/p\u003e"
            }
          ],
          "value": "The fix removes all capture-derived values (item IDs, URLs, screenshot paths) from inline JavaScript contexts (onclick attributes and inline script variable assignments). These values are now stored in HTML data attributes (data-screenshot, data-url, data-item-id), which are properly HTML-escaped by the Jinja2 template engine. Screenshot click handlers are bound programmatically via addEventListener on elements with a .reload-image class, and the reload_image function reads values from the DOM element\u0027s dataset rather than receiving raw string parameters. Additionally, item IDs are URL-encoded with encodeURIComponent before being appended to link hrefs, preventing injection through the ID field."
        }
      ],
      "source": {
        "discovery": "UNKNOWN"
      },
      "title": "Stored Cross-Site Scripting (XSS) via Crawler Capture Import in AIL Framework showDomain Page",
      "x_gcve": [
        {
          "extensions": {
            "bcp-05-x-01": {
              "ai_annotations": [
                {
                  "ai_level": "generated",
                  "description": "Draft vulnerability metadata was generated from a git-format patch using an Ollama-hosted language model.",
                  "gna_source": 1,
                  "models": [
                    {
                      "gna_source": 1,
                      "identifier": "qwen3.8:27b",
                      "name": "qwen3.8:27b",
                      "source": "ollama"
                    }
                  ],
                  "review_status": "full",
                  "scope": "record",
                  "tags": [
                    "ai-computer-assisted:llm-generated",
                    "ai-computer-assisted:classification"
                  ]
                }
              ]
            },
            "bcp-05-x-02": {
              "x_patch2vuln": {
                "assumptions": [
                  "The UI:P (Passive) rating assumes the inline script vector (screenshot variable assignment) executes on page load without requiring a user click. The onclick vector would require UI:A, but the inline script vector is sufficient for exploitation.",
                  "PR:L assumes the attacker needs a user-role API client to import a crafted crawler capture, as stated in the commit message. The exact privilege level and authentication mechanism are not further specified in the patch.",
                  "The version boundary \u0027less_than 7.1\u0027 is inferred from the tag_version_boundary metadata showing 21 commits after the fix before the v7.1 tag, suggesting the fix is included in v7.1. Exact affected version ranges are not confirmed.",
                  "CAPEC-1 is selected as the closest available attack pattern for stored XSS; no more specific CAPEC for stored XSS via data import exists in the catalog.",
                  "The patch does not include test cases or additional context about whether other template files have similar inline JavaScript interpolation patterns; this advisory covers only the single file in the patch."
                ],
                "capecRationale": [
                  {
                    "capecId": "CAPEC-1",
                    "rationale": "The patch directly remediates a stored XSS vector where attacker-controlled data from a crawler capture import is reflected into inline JavaScript without neutralization. CAPEC-1 is the canonical attack pattern for XSS. The specific sub-variant is stored (persistent) XSS via a data import mechanism, but CAPEC-1 is the closest and most appropriate mapping available in the CAPEC catalog."
                  }
                ],
                "commit": "31376ee3d04519c898bb9b3671453a41c4117340",
                "confidence": "medium",
                "credits": [
                  {
                    "lang": "en",
                    "type": "remediation developer",
                    "value": "terrtia"
                  }
                ],
                "cvssRationale": "AV:N: Exploitable over the network via the web application. AC:L: No special race conditions or complex setup; a crafted capture import suffices. AT:N: No manipulation of the attack target required. PR:L: Requires a user-role API client to import a malicious crawler capture (authenticated, low privilege). UI:P: The stored XSS in the inline script block executes passively when the victim loads the domain page; no click or active interaction is needed for the inline script vector. VC:N/VI:N/VA:N: The vulnerable component (server-side application) is not directly compromised in confidentiality, integrity, or availability. SC:H: Scope changes to the victim\u0027s browser session, enabling theft of session tokens, cookies, and sensitive data. SI:H: Scope integrity is impacted as the attacker can perform arbitrary actions in the victim\u0027s authenticated session. SA:N: No meaningful impact on the availability of the attacker\u0027s scope.",
                "fixSummary": "The fix removes all capture-derived values (item IDs, URLs, screenshot paths) from inline JavaScript contexts (onclick attributes and inline script variable assignments). These values are now stored in HTML data attributes (data-screenshot, data-url, data-item-id), which are properly HTML-escaped by the Jinja2 template engine. Screenshot click handlers are bound programmatically via addEventListener on elements with a .reload-image class, and the reload_image function reads values from the DOM element\u0027s dataset rather than receiving raw string parameters. Additionally, item IDs are URL-encoded with encodeURIComponent before being appended to link hrefs, preventing injection through the ID field.",
                "generatedAt": "2026-09-25T13:45:46.199650Z",
                "generator": "patch2vuln.py",
                "model": "qwen3.8:27b",
                "modelComparison": {
                  "rankings": [
                    {
                      "agreementScore": 9,
                      "assumptionCount": 5,
                      "confidence": "medium",
                      "model": "qwen3.8:27b",
                      "score": 5
                    }
                  ],
                  "selectedModel": "qwen3.8:27b",
                  "selectionMethod": "deterministic-consensus-v1",
                  "selectionNotice": "The selected result is closest to model consensus; this heuristic does not establish factual correctness and human review remains required."
                },
                "patchSha256": "c110d16dd8eee489472853615089aeb6b6e034457dcbaec9b038bf85ad96b920",
                "patchSummary": "In showDomain.html: (1) Replaced inline onclick=\"reload_image(\u0027...\u0027)\" with a data-attribute-based approach using data-screenshot, data-url, and data-item-id attributes on the table cell element, plus a .reload-image CSS class. (2) Removed the inline script block that assigned screenshot, URL, and ID values directly into JavaScript variables via Jinja2 interpolation; replaced with dataset reads from the domain_screenshot img element. (3) Changed reload_image() to accept a DOM element and read values from element.dataset instead of raw string arguments. (4) Added a programmatic event binding loop using document.querySelectorAll(\u0027.reload-image\u0027).forEach() with addEventListener(\u0027click\u0027). (5) Applied encodeURIComponent() to item IDs when constructing screenshot link hrefs. (6) Changed the href URL construction to use url_for with keyword arguments instead of string interpolation.",
                "patchTruncated": false,
                "patches": [
                  {
                    "commit": "31376ee3d04519c898bb9b3671453a41c4117340",
                    "patchSha256": "c110d16dd8eee489472853615089aeb6b6e034457dcbaec9b038bf85ad96b920",
                    "source": "https://github.com/ail-project/ail-framework/commit/31376ee3d04519c898bb9b3671453a41c4117340.patch",
                    "sourceUrl": "https://github.com/ail-project/ail-framework/commit/31376ee3d04519c898bb9b3671453a41c4117340.patch",
                    "subject": "chg: [security] prevent stored XSS through imported capture"
                  }
                ],
                "source": "https://github.com/ail-project/ail-framework/commit/31376ee3d04519c898bb9b3671453a41c4117340.patch",
                "subject": "chg: [security] prevent stored XSS through imported capture",
                "tagVersionBoundary": {
                  "commits_after_fix": 21,
                  "repository": "https://github.com/ail-project/ail-framework",
                  "tag": "v7.1",
                  "version": "7.1",
                  "version_type": "custom"
                },
                "weaknessRationale": [
                  {
                    "cweId": "CWE-79",
                    "rationale": "The patch explicitly addresses stored XSS by removing user-controlled values (screenshot paths, URLs, item IDs from crawler capture imports) from inline JavaScript contexts and replacing them with properly escaped data attributes and programmatic event binding. The commit message confirms the intent: \u0027prevent stored XSS through imported capture metadata in showDomain.\u0027"
                  }
                ]
              }
            }
          },
          "recordType": "advisory",
          "vulnId": "GCVE-1-2026-20292"
        }
      ],
      "x_generator": {
        "engine": "Vulnogram 0.2.0"
      }
    }
  },
  "cveMetadata": {
    "assignerOrgId": "00000000-0000-4000-9000-000000000000",
    "cveId": "CVE-2026-100190",
    "datePublished": "2026-09-25T13:49:00.000Z",
    "dateReserved": "2026-09-25T13:52:13.769Z",
    "dateUpdated": "2026-09-25T13:52:14.626493Z",
    "requesterUserId": "00000000-0000-4000-9000-000000000000",
    "serial": 1,
    "state": "PUBLISHED",
    "vulnId": "GCVE-1-2026-20292"
  },
  "dataType": "CVE_RECORD",
  "dataVersion": "5.1"
}

GCVE-1-2026-20108 (CVE-2026-97898)

Vulnerability from gna-1 – Published: 2026-09-25 09:44 – Updated: 2026-09-25 10:07 Exclusively Hosted Service
VLAI
Title
Broken authorization in Akia keyless entry lets an authenticated guest unlock other rooms
Summary
Insecure Direct Object Reference / missing object-level authorization in the Akia keyless entry cloud service. The unlock action is relying on a client-supplied room/door identifier that is not properly authorized server-side against the authenticated guest's booking. An authenticated guest could unlock rooms other than their own, resulting in unauthorized physical access to guest rooms at an affected property. As of 19th September 2026 the service is no more vulnerable to this attack (feedback received by the reporter).  The attack is remote but the effect is local to an affected property.
CWE
  • CWE-862 - Missing Authorization
  • CWE-639 - Authorization Bypass Through User-Controlled Key
Assigner
CIRCL GNA GNA-1
GNA scorecard A 91/100 over 279 records in the last 180 days details
References
Impacted products
Vendor Product Version
akia akia Affected: 0 , < 2026.10.19 (semver)
Create a notification for this product.

{
  "containers": {
    "cna": {
      "affected": [
        {
          "defaultStatus": "unaffected",
          "product": "akia",
          "vendor": "akia",
          "versions": [
            {
              "lessThan": "2026.10.19",
              "status": "affected",
              "version": "0",
              "versionType": "semver"
            }
          ]
        }
      ],
      "credits": [
        {
          "lang": "en",
          "type": "finder",
          "value": "Philipp Br\u00fcgger"
        }
      ],
      "descriptions": [
        {
          "lang": "en",
          "supportingMedia": [
            {
              "base64": false,
              "type": "text/html",
              "value": "\u003cdiv\u003eInsecure Direct Object Reference / missing object-level authorization in the Akia keyless entry cloud service. The unlock action is relying on a client-supplied room/door identifier that is not properly authorized server-side against the authenticated guest\u0027s booking. An authenticated guest could unlock rooms other than their own, resulting in unauthorized physical access to guest rooms at an affected property.\u003c/div\u003e\u003cdiv\u003e\u003cbr\u003e\u003c/div\u003e\u003cdiv\u003eAs of 19th September 2026 the service is no more vulnerable to this attack (feedback received by the reporter).\u0026nbsp;\u003c/div\u003e\u003cdiv\u003e\u003cbr\u003e\u003c/div\u003e\u003cdiv\u003eThe attack is remote but the effect is local to an affected property.\u0026nbsp;\u003c/div\u003e\u003cbr\u003e"
            }
          ],
          "value": "Insecure Direct Object Reference / missing object-level authorization in the Akia keyless entry cloud service. The unlock action is relying on a client-supplied room/door identifier that is not properly authorized server-side against the authenticated guest\u0027s booking. An authenticated guest could unlock rooms other than their own, resulting in unauthorized physical access to guest rooms at an affected property.\n\n\n\n\nAs of 19th September 2026 the service is no more vulnerable to this attack (feedback received by the reporter).\u00a0\n\n\n\n\nThe attack is remote but the effect is local to an affected property."
        }
      ],
      "impacts": [
        {
          "capecId": "CAPEC-1",
          "descriptions": [
            {
              "lang": "en",
              "value": "CAPEC-1 Accessing Functionality Not Properly Constrained by ACLs"
            }
          ]
        }
      ],
      "metrics": [
        {
          "cvssV4_0": {
            "Automatable": "NOT_DEFINED",
            "Recovery": "NOT_DEFINED",
            "Safety": "NOT_DEFINED",
            "attackComplexity": "LOW",
            "attackRequirements": "NONE",
            "attackVector": "NETWORK",
            "baseScore": 8.4,
            "baseSeverity": "HIGH",
            "privilegesRequired": "NONE",
            "providerUrgency": "NOT_DEFINED",
            "subAvailabilityImpact": "NONE",
            "subConfidentialityImpact": "HIGH",
            "subIntegrityImpact": "HIGH",
            "userInteraction": "ACTIVE",
            "valueDensity": "NOT_DEFINED",
            "vectorString": "CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:A/VC:H/VI:L/VA:N/SC:H/SI:H/SA:N",
            "version": "4.0",
            "vulnAvailabilityImpact": "NONE",
            "vulnConfidentialityImpact": "HIGH",
            "vulnIntegrityImpact": "LOW",
            "vulnerabilityResponseEffort": "NOT_DEFINED"
          },
          "format": "CVSS",
          "scenarios": [
            {
              "lang": "en",
              "value": "GENERAL"
            }
          ]
        }
      ],
      "problemTypes": [
        {
          "descriptions": [
            {
              "cweId": "CWE-862",
              "description": "CWE-862 Missing Authorization",
              "lang": "en",
              "type": "CWE"
            }
          ]
        },
        {
          "descriptions": [
            {
              "cweId": "CWE-639",
              "description": "CWE-639 Authorization Bypass Through User-Controlled Key",
              "lang": "en",
              "type": "CWE"
            }
          ]
        }
      ],
      "providerMetadata": {
        "orgId": "00000000-0000-4000-9000-000000000000"
      },
      "references": [
        {
          "tags": [
            "product"
          ],
          "url": "https://www.akia.com/platform/keyless-entry"
        }
      ],
      "source": {
        "discovery": "UNKNOWN"
      },
      "tags": [
        "exclusively-hosted-service"
      ],
      "title": "Broken authorization in Akia keyless entry lets an authenticated guest unlock other rooms",
      "x_gcve": [
        {
          "recordType": "advisory",
          "vulnId": "gcve-1-2026-20108"
        }
      ],
      "x_generator": {
        "engine": "Vulnogram 0.2.0"
      }
    }
  },
  "cveMetadata": {
    "assignerOrgId": "00000000-0000-4000-9000-000000000000",
    "cveId": "CVE-2026-97898",
    "datePublished": "2026-09-25T09:44:00.000Z",
    "dateReserved": "2026-09-25T10:07:22.362Z",
    "dateUpdated": "2026-09-25T10:07:22.429982Z",
    "requesterUserId": "00000000-0000-4000-9000-000000000000",
    "serial": 1,
    "state": "PUBLISHED",
    "vulnId": "GCVE-1-2026-20108"
  },
  "dataType": "CVE_RECORD",
  "dataVersion": "5.1"
}

GCVE-1-2026-20190 (CVE-2026-97863)

Vulnerability from gna-1 – Published: 2026-09-25 08:02 – Updated: 2026-09-25 09:49
VLAI
Title
misp-modules: Shell Command Injection in MISP cisco_firesight_manager_ACL_rule_export Module via Unescaped Attribute Values
Summary
The cisco_firesight_manager_ACL_rule_export module in misp-modules generates a shell script (.sh) that authenticates to and calls the Cisco fireSIGHT Manager API. The module interpolates configuration values (IP address, login, password, domain ID, policy ID) and MISP attribute values (destination IPs, URLs, event info comments) directly into single-quoted shell string assignments without any escaping or sanitization. Because the values are placed inside single-quoted shell strings, any value containing a single-quote character (e.g., a crafted ip-dst or url attribute value submitted to MISP) breaks out of the quoting context, allowing an attacker to inject arbitrary shell commands into the exported script. A security analyst who subsequently executes the generated .sh file unmodified would run the injected commands with their own privileges, potentially exposing fireSIGHT Manager credentials, modifying ACL rules, or compromising the analyst workstation. Additionally, the module contained a secondary defect where the variable 'config' was only assigned inside a conditional block but referenced unconditionally afterward, causing a NameError (denial of service) when the request payload lacked a 'config' key. The vulnerability requires the attacker to have the ability to submit MISP events or attributes containing a single-quote character and the victim to execute the exported script. No authentication bypass is required beyond standard MISP event-submission privileges.
CWE
  • CWE-78 - Improper Neutralization of Special Elements used in a Command ("Command Injection")
Assigner
CIRCL GNA GNA-1
GNA scorecard A 91/100 over 279 records in the last 180 days details
References
Impacted products
Vendor Product Version
misp misp-modules Affected: 0 , ≤ 3.0.10 (semver)
Create a notification for this product.
GCVE extensions
bcp-05-x-01
AI-assisted vulnerability information annotation
GCVE-BCP-05-X-01
Whole record AI-generated Not human-reviewed GNA-1

Draft vulnerability metadata was generated from a git-format patch using an Ollama-hosted language model. Human validation is required before publication.

ai-computer-assisted:llm-generatedai-computer-assisted:classification
Model Source Identifier
qwen3.8:27b ollama qwen3.8:27b
bcp-05-x-02
Patch-to-vulnerability generation provenance
GCVE-BCP-05-X-02
Generator
patch2vuln.py on 2026-09-25 07:48
Model
qwen3.8:27b
Input
https://github.com/elhoim/misp-modules/commit/625b54908efbd6acc8343aa3370d401dd370e748.patch 9c34349775d0…
Confidence
high
Commit Subject Patch SHA-256
625b54908efb fix: [cisco_firesight_manager_ACL_rule_export] shell-escape 9c34349775d0…
Fix summary

All user-controlled and configuration values interpolated into the generated shell script are now passed through Python's shlex.quote() function, which produces a safely quoted string that cannot be broken out of by embedded shell metacharacters. For the JSON access-rule block that embeds multiple attribute values within a single shell assignment, the JSON content is first assembled as plain text and then the entire assembled string is shlex-quoted once, preventing stray quote characters from corrupting the outer quoting. The 'config' variable is now initialized to an empty dictionary before the conditional assignment, eliminating the NameError.

Patch summary

Added 'import shlex' at the top of the module. Replaced direct .format() interpolation of config values (fmc_ip_addr, fmc_login, fmc_pass, domain_id, acpolicy_id) with shlex.quote()-wrapped values. Refactored the BLOCK_JSON_TMPL to separate the JSON content template (BLOCK_JSON_CONTENT_TMPL) from the shell assignment template; the JSON content is formatted first with raw attribute values, then the resulting string is passed through shlex.quote() before being substituted into the shell assignment. Initialized 'config = {}' before the 'if "config" in request:' block to prevent NameError.

CVSS rationale

AV:N: The attacker submits crafted MISP attribute values over the network to a MISP instance. AC:L: Injecting a single-quote character into an attribute value is trivial and requires no race conditions or special timing. AT:N: No special data manipulation or prior access to the target system is needed beyond submitting an event. PR:L: The attacker needs only basic MISP user privileges to create or modify events/attributes. UI:A: The victim analyst must actively execute the generated .sh file for the injected commands to run. VC:N, VI:N, VA:N: The MISP instance itself is not directly compromised; the impact is on the downstream system where the script executes. SC:H: The injected commands can exfiltrate fireSIGHT Manager credentials and analyst system data. SI:H: Arbitrary command execution allows modification of ACL rules, system files, or other integrity-critical resources. SA:H: Injected commands can disrupt or destroy the analyst workstation or fireSIGHT Manager availability.

Weakness rationale
  • CWE-78 The module constructs a shell script by interpolating untrusted values (MISP attribute values, configuration fields) into single-quoted shell strings without escaping. A single-quote character in any interpolated value breaks the quoting and allows arbitrary shell command injection. This is a textbook command injection via insufficient output encoding in a generated shell script.
Attack pattern rationale
  • CAPEC-88 The attack pattern involves an attacker supplying input containing shell metacharacters (specifically a single quote) that is incorporated into a shell command or script without proper sanitization, resulting in execution of attacker-controlled commands. The patch confirms the vulnerable pattern: values are interpolated into single-quoted shell assignments, and the fix is to apply shlex.quote() to neutralize shell metacharacters. This is the closest and most specific CAPEC mapping for the observed vulnerability.
Assumptions to verify
  • The CVSS UI metric is set to A (Active) because the victim analyst must deliberately execute the generated .sh file; CVSS v4.0 does not use the v3.x 'R' value.
  • PR is set to L assuming the attacker holds a basic MISP user role sufficient to create or modify events and attributes; the patch does not specify exact role requirements.
  • SC, SI, SA are rated H based on the potential for arbitrary command execution on the analyst workstation and the fireSIGHT Manager; actual impact depends on the analyst's local privileges and the fireSIGHT Manager's exposure.
  • The CAPEC-88 mapping is the closest available pattern; the vulnerability is specifically an output-encoding failure in generated shell script rather than direct command-line injection, but CAPEC-88 is the narrowest defensible match.
  • No specific affected or fixed version numbers are available from the patch metadata; the commit date (2026-08-31) is the only temporal anchor.
  • The secondary NameError bug (config variable) is a minor denial-of-service issue included in the same patch but is not the primary security concern.
Model comparison

Selected qwen3.8:27b by deterministic-consensus-v1
The selected result is closest to model consensus; this heuristic does not establish factual correctness and human review remains required.

Model Score Agreement Confidence Assumptions
qwen3.8:27b 5 9 high 6

{
  "containers": {
    "cna": {
      "affected": [
        {
          "collectionURL": "https://github.com/MISP/misp-modules",
          "defaultStatus": "unaffected",
          "modules": [
            "cisco_firesight_manager_ACL_rule_export"
          ],
          "product": "misp-modules",
          "programFiles": [
            "misp_modules/modules/export_mod/cisco_firesight_manager_ACL_rule_export.py"
          ],
          "repo": "https://github.com/misp/misp-modules",
          "vendor": "misp",
          "versions": [
            {
              "lessThanOrEqual": "3.0.10",
              "status": "affected",
              "version": "0",
              "versionType": "semver"
            }
          ]
        }
      ],
      "credits": [
        {
          "lang": "en",
          "type": "remediation developer",
          "value": "elhoim"
        },
        {
          "lang": "en",
          "type": "remediation developer",
          "value": "Claude Opus 5"
        }
      ],
      "descriptions": [
        {
          "lang": "en",
          "supportingMedia": [
            {
              "base64": false,
              "type": "text/html",
              "value": "\u003cp\u003eThe cisco_firesight_manager_ACL_rule_export module in misp-modules generates a shell script (.sh) that authenticates to and calls the Cisco fireSIGHT Manager API. The module interpolates configuration values (IP address, login, password, domain ID, policy ID) and MISP attribute values (destination IPs, URLs, event info comments) directly into single-quoted shell string assignments without any escaping or sanitization. Because the values are placed inside single-quoted shell strings, any value containing a single-quote character (e.g., a crafted ip-dst or url attribute value submitted to MISP) breaks out of the quoting context, allowing an attacker to inject arbitrary shell commands into the exported script. A security analyst who subsequently executes the generated .sh file unmodified would run the injected commands with their own privileges, potentially exposing fireSIGHT Manager credentials, modifying ACL rules, or compromising the analyst workstation. Additionally, the module contained a secondary defect where the variable \u0027config\u0027 was only assigned inside a conditional block but referenced unconditionally afterward, causing a NameError (denial of service) when the request payload lacked a \u0027config\u0027 key. The vulnerability requires the attacker to have the ability to submit MISP events or attributes containing a single-quote character and the victim to execute the exported script. No authentication bypass is required beyond standard MISP event-submission privileges.\u003c/p\u003e"
            }
          ],
          "value": "The cisco_firesight_manager_ACL_rule_export module in misp-modules generates a shell script (.sh) that authenticates to and calls the Cisco fireSIGHT Manager API. The module interpolates configuration values (IP address, login, password, domain ID, policy ID) and MISP attribute values (destination IPs, URLs, event info comments) directly into single-quoted shell string assignments without any escaping or sanitization. Because the values are placed inside single-quoted shell strings, any value containing a single-quote character (e.g., a crafted ip-dst or url attribute value submitted to MISP) breaks out of the quoting context, allowing an attacker to inject arbitrary shell commands into the exported script. A security analyst who subsequently executes the generated .sh file unmodified would run the injected commands with their own privileges, potentially exposing fireSIGHT Manager credentials, modifying ACL rules, or compromising the analyst workstation. Additionally, the module contained a secondary defect where the variable \u0027config\u0027 was only assigned inside a conditional block but referenced unconditionally afterward, causing a NameError (denial of service) when the request payload lacked a \u0027config\u0027 key. The vulnerability requires the attacker to have the ability to submit MISP events or attributes containing a single-quote character and the victim to execute the exported script. No authentication bypass is required beyond standard MISP event-submission privileges."
        }
      ],
      "impacts": [
        {
          "capecId": "CAPEC-88",
          "descriptions": [
            {
              "lang": "en",
              "value": "CAPEC-88 Shell Command Injection"
            }
          ]
        }
      ],
      "metrics": [
        {
          "cvssV4_0": {
            "Automatable": "NOT_DEFINED",
            "Recovery": "NOT_DEFINED",
            "Safety": "NOT_DEFINED",
            "attackComplexity": "LOW",
            "attackRequirements": "NONE",
            "attackVector": "NETWORK",
            "baseScore": 6.3,
            "baseSeverity": "MEDIUM",
            "privilegesRequired": "LOW",
            "providerUrgency": "NOT_DEFINED",
            "subAvailabilityImpact": "HIGH",
            "subConfidentialityImpact": "HIGH",
            "subIntegrityImpact": "HIGH",
            "userInteraction": "ACTIVE",
            "valueDensity": "NOT_DEFINED",
            "vectorString": "CVSS:4.0/AV:N/AC:L/AT:N/PR:L/UI:A/VC:N/VI:N/VA:N/SC:H/SI:H/SA:H",
            "version": "4.0",
            "vulnAvailabilityImpact": "NONE",
            "vulnConfidentialityImpact": "NONE",
            "vulnIntegrityImpact": "NONE",
            "vulnerabilityResponseEffort": "NOT_DEFINED"
          },
          "format": "CVSS",
          "scenarios": [
            {
              "lang": "en",
              "value": "GENERAL"
            }
          ]
        }
      ],
      "problemTypes": [
        {
          "descriptions": [
            {
              "cweId": "CWE-78",
              "description": "CWE-78 Improper Neutralization of Special Elements used in a Command (\"Command Injection\")",
              "lang": "en",
              "type": "CWE"
            }
          ]
        }
      ],
      "providerMetadata": {
        "orgId": "00000000-0000-4000-9000-000000000000"
      },
      "references": [
        {
          "name": "Security patch",
          "tags": [
            "patch"
          ],
          "url": "https://github.com/misp/misp-modules/commit/625b54908efbd6acc8343aa3370d401dd370e748"
        }
      ],
      "solutions": [
        {
          "lang": "en",
          "supportingMedia": [
            {
              "base64": false,
              "type": "text/html",
              "value": "\u003cp\u003eAll user-controlled and configuration values interpolated into the generated shell script are now passed through Python\u0027s shlex.quote() function, which produces a safely quoted string that cannot be broken out of by embedded shell metacharacters. For the JSON access-rule block that embeds multiple attribute values within a single shell assignment, the JSON content is first assembled as plain text and then the entire assembled string is shlex-quoted once, preventing stray quote characters from corrupting the outer quoting. The \u0027config\u0027 variable is now initialized to an empty dictionary before the conditional assignment, eliminating the NameError.\u003c/p\u003e"
            }
          ],
          "value": "All user-controlled and configuration values interpolated into the generated shell script are now passed through Python\u0027s shlex.quote() function, which produces a safely quoted string that cannot be broken out of by embedded shell metacharacters. For the JSON access-rule block that embeds multiple attribute values within a single shell assignment, the JSON content is first assembled as plain text and then the entire assembled string is shlex-quoted once, preventing stray quote characters from corrupting the outer quoting. The \u0027config\u0027 variable is now initialized to an empty dictionary before the conditional assignment, eliminating the NameError."
        }
      ],
      "source": {
        "discovery": "UNKNOWN"
      },
      "title": "misp-modules: Shell Command Injection in MISP cisco_firesight_manager_ACL_rule_export Module via Unescaped Attribute Values",
      "x_gcve": [
        {
          "extensions": {
            "bcp-05-x-01": {
              "ai_annotations": [
                {
                  "ai_level": "generated",
                  "description": "Draft vulnerability metadata was generated from a git-format patch using an Ollama-hosted language model. Human validation is required before publication.",
                  "gna_source": 1,
                  "models": [
                    {
                      "gna_source": 1,
                      "identifier": "qwen3.8:27b",
                      "name": "qwen3.8:27b",
                      "source": "ollama"
                    }
                  ],
                  "review_status": "none",
                  "scope": "record",
                  "tags": [
                    "ai-computer-assisted:llm-generated",
                    "ai-computer-assisted:classification"
                  ]
                }
              ]
            },
            "bcp-05-x-02": {
              "x_patch2vuln": {
                "assumptions": [
                  "The CVSS UI metric is set to A (Active) because the victim analyst must deliberately execute the generated .sh file; CVSS v4.0 does not use the v3.x \u0027R\u0027 value.",
                  "PR is set to L assuming the attacker holds a basic MISP user role sufficient to create or modify events and attributes; the patch does not specify exact role requirements.",
                  "SC, SI, SA are rated H based on the potential for arbitrary command execution on the analyst workstation and the fireSIGHT Manager; actual impact depends on the analyst\u0027s local privileges and the fireSIGHT Manager\u0027s exposure.",
                  "The CAPEC-88 mapping is the closest available pattern; the vulnerability is specifically an output-encoding failure in generated shell script rather than direct command-line injection, but CAPEC-88 is the narrowest defensible match.",
                  "No specific affected or fixed version numbers are available from the patch metadata; the commit date (2026-08-31) is the only temporal anchor.",
                  "The secondary NameError bug (config variable) is a minor denial-of-service issue included in the same patch but is not the primary security concern."
                ],
                "capecRationale": [
                  {
                    "capecId": "CAPEC-88",
                    "rationale": "The attack pattern involves an attacker supplying input containing shell metacharacters (specifically a single quote) that is incorporated into a shell command or script without proper sanitization, resulting in execution of attacker-controlled commands. The patch confirms the vulnerable pattern: values are interpolated into single-quoted shell assignments, and the fix is to apply shlex.quote() to neutralize shell metacharacters. This is the closest and most specific CAPEC mapping for the observed vulnerability."
                  }
                ],
                "commit": "625b54908efbd6acc8343aa3370d401dd370e748",
                "confidence": "high",
                "credits": [
                  {
                    "lang": "en",
                    "type": "remediation developer",
                    "value": "elhoim"
                  },
                  {
                    "lang": "en",
                    "type": "remediation developer",
                    "value": "Claude Opus 5"
                  }
                ],
                "cvssRationale": "AV:N: The attacker submits crafted MISP attribute values over the network to a MISP instance. AC:L: Injecting a single-quote character into an attribute value is trivial and requires no race conditions or special timing. AT:N: No special data manipulation or prior access to the target system is needed beyond submitting an event. PR:L: The attacker needs only basic MISP user privileges to create or modify events/attributes. UI:A: The victim analyst must actively execute the generated .sh file for the injected commands to run. VC:N, VI:N, VA:N: The MISP instance itself is not directly compromised; the impact is on the downstream system where the script executes. SC:H: The injected commands can exfiltrate fireSIGHT Manager credentials and analyst system data. SI:H: Arbitrary command execution allows modification of ACL rules, system files, or other integrity-critical resources. SA:H: Injected commands can disrupt or destroy the analyst workstation or fireSIGHT Manager availability.",
                "draft": false,
                "fixSummary": "All user-controlled and configuration values interpolated into the generated shell script are now passed through Python\u0027s shlex.quote() function, which produces a safely quoted string that cannot be broken out of by embedded shell metacharacters. For the JSON access-rule block that embeds multiple attribute values within a single shell assignment, the JSON content is first assembled as plain text and then the entire assembled string is shlex-quoted once, preventing stray quote characters from corrupting the outer quoting. The \u0027config\u0027 variable is now initialized to an empty dictionary before the conditional assignment, eliminating the NameError.",
                "generatedAt": "2026-09-25T07:48:15.163752Z",
                "generator": "patch2vuln.py",
                "model": "qwen3.8:27b",
                "modelComparison": {
                  "rankings": [
                    {
                      "agreementScore": 9,
                      "assumptionCount": 6,
                      "confidence": "high",
                      "model": "qwen3.8:27b",
                      "score": 5
                    }
                  ],
                  "selectedModel": "qwen3.8:27b",
                  "selectionMethod": "deterministic-consensus-v1",
                  "selectionNotice": "The selected result is closest to model consensus; this heuristic does not establish factual correctness and human review remains required."
                },
                "patchSha256": "9c34349775d058188a7b9a4d44ff873fd285faf0ddda00b46aadbaa72f3b4e6e",
                "patchSummary": "Added \u0027import shlex\u0027 at the top of the module. Replaced direct .format() interpolation of config values (fmc_ip_addr, fmc_login, fmc_pass, domain_id, acpolicy_id) with shlex.quote()-wrapped values. Refactored the BLOCK_JSON_TMPL to separate the JSON content template (BLOCK_JSON_CONTENT_TMPL) from the shell assignment template; the JSON content is formatted first with raw attribute values, then the resulting string is passed through shlex.quote() before being substituted into the shell assignment. Initialized \u0027config = {}\u0027 before the \u0027if \"config\" in request:\u0027 block to prevent NameError.",
                "patchTruncated": false,
                "patches": [
                  {
                    "commit": "625b54908efbd6acc8343aa3370d401dd370e748",
                    "patchSha256": "9c34349775d058188a7b9a4d44ff873fd285faf0ddda00b46aadbaa72f3b4e6e",
                    "source": "https://github.com/elhoim/misp-modules/commit/625b54908efbd6acc8343aa3370d401dd370e748.patch",
                    "sourceUrl": "https://github.com/elhoim/misp-modules/commit/625b54908efbd6acc8343aa3370d401dd370e748.patch",
                    "subject": "fix: [cisco_firesight_manager_ACL_rule_export] shell-escape"
                  }
                ],
                "source": "https://github.com/elhoim/misp-modules/commit/625b54908efbd6acc8343aa3370d401dd370e748.patch",
                "subject": "fix: [cisco_firesight_manager_ACL_rule_export] shell-escape",
                "weaknessRationale": [
                  {
                    "cweId": "CWE-78",
                    "rationale": "The module constructs a shell script by interpolating untrusted values (MISP attribute values, configuration fields) into single-quoted shell strings without escaping. A single-quote character in any interpolated value breaks the quoting and allows arbitrary shell command injection. This is a textbook command injection via insufficient output encoding in a generated shell script."
                  }
                ]
              }
            }
          },
          "recordType": "advisory",
          "vulnId": "gcve-1-2026-20190"
        }
      ],
      "x_generator": {
        "engine": "Vulnogram 0.2.0"
      }
    }
  },
  "cveMetadata": {
    "assignerOrgId": "00000000-0000-4000-9000-000000000000",
    "cveId": "CVE-2026-97863",
    "datePublished": "2026-09-25T08:02:00.000Z",
    "dateReserved": "2026-09-25T08:02:00.000Z",
    "dateUpdated": "2026-09-25T09:49:45.859656Z",
    "requesterUserId": "00000000-0000-4000-9000-000000000000",
    "serial": 1,
    "state": "PUBLISHED",
    "vulnId": "gcve-1-2026-20190"
  },
  "dataType": "CVE_RECORD",
  "dataVersion": "5.1"
}

GCVE-1-2026-20263 (CVE-2026-95806)

Vulnerability from gna-1 – Published: 2026-09-22 15:08 – Updated: 2026-09-22 15:09
VLAI
Title
MISP: PHP phar stream wrapper enables deserialization and code execution via caller-influenced filesystem paths
Summary
MISP ships with PHP's phar stream wrapper registered in both its web entry point and its console entry point.  The phar stream wrapper causes PHP to treat a phar archive as a directory, which has two security consequences:    - any filesystem operation on a caller-influenced path that resolves to a phar archive triggers an implicit unserialize() call, creating a deserialization sink;  - a relocated application root can reach executable code inside an uploaded phar file, enabling arbitrary code execution as the web user. No component of MISP, the vendored CakePHP framework, or any runtime-loaded library reads or constructs phar archives. The wrapper therefore serves no legitimate purpose in the MISP runtime and exists solely as an available primitive for an attacker who can influence a filesystem path argument.
CWE
  • CWE-502 - Deserialization of Untrusted Data
  • CWE-74 - Improper Neutralization of Special Elements in Output Used by a Downstream Component
Assigner
CIRCL GNA GNA-1
GNA scorecard A 91/100 over 279 records in the last 180 days details
References
Impacted products
Vendor Product Version
MISP MISP Affected: 0 , < 2.5.47 (semver)
Create a notification for this product.
GCVE extensions
bcp-05-x-01
AI-assisted vulnerability information annotation
GCVE-BCP-05-X-01
Whole record AI-generated Human-reviewed GNA-1

Draft vulnerability metadata was generated from a git-format patch using an Ollama-hosted language model. Human validation is required before publication.

ai-computer-assisted:llm-generatedai-computer-assisted:classification
Model Source Identifier
qwen3.8:27b ollama qwen3.8:27b
bcp-05-x-02
Patch-to-vulnerability generation provenance
GCVE-BCP-05-X-02
Generator
patch2vuln.py on 2026-09-22 15:01
Model
qwen3.8:27b
Input
https://github.com/MISP/MISP/commit/08fa755b6.patch f4fc3223e642…
Confidence
medium
Commit Subject Patch SHA-256
08fa755b6363 chg: [internal] Unregister the phar stream wrapper in the f4fc3223e642…
Fix summary

The phar stream wrapper is unregistered via stream_wrapper_unregister('phar') at the top of both the web and console entry points, before any framework bootstrap or application code executes. Because no MISP component, CakePHP, or runtime library requires the phar wrapper, removing it eliminates the implicit unserialize() sink and the directory-like phar archive behavior without functional impact. This closes the deserialization and code-execution primitive application-wide, independent of whether individual callers validate their path arguments.

Patch summary

Two files are modified. In app/Console/cake.php and app/webroot/index.php, a 14-line block is inserted immediately after the file header comment and before any existing logic. The block checks whether 'phar' is present in stream_get_wrappers() and, if so, calls stream_wrapper_unregister('phar'). A docblock comment explains the security rationale: the wrapper turns filesystem calls on caller-influenced paths into unserialize() sinks and allows a relocated application root to reach code inside an uploaded file. No other code is changed; the total diff is 28 insertions across the two files.

CVSS rationale

AV:N: The vulnerability is reachable through the web entry point (app/webroot/index.php), making it network-accessible. AC:H: Exploitation requires crafting a valid phar archive with a malicious serialized payload and identifying a code path where a caller-influenced filesystem argument resolves to that archive; the commit describes this as an 'argument-injection exploit,' implying non-trivial path manipulation. AT:N: No specific attack-target conditions are indicated. PR:L: MISP is a threat-intelligence platform that typically requires authenticated access; the commit references 'the web user' context, suggesting the attacker operates within the application's privilege boundary. UI:N: No user interaction is required. VC/VI/VA:H: Successful exploitation results in arbitrary code execution as the web user, compromising confidentiality, integrity, and availability of the MISP instance. SC/SI/SA:N: No evidence of impact on subsequent or other components beyond the MISP instance itself.

Weakness rationale
  • CWE-502 The phar stream wrapper causes PHP to invoke unserialize() implicitly whenever a filesystem operation resolves to a phar archive. An attacker who can influence the path argument (e.g., via argument injection) can supply a crafted phar file, triggering deserialization of attacker-controlled data and leading to code execution. This is the primary and most specific weakness.
  • CWE-74 The phar stream wrapper is a special element of the PHP runtime that was not neutralized (unregistered) in the MISP entry points. Its presence allows downstream filesystem calls to be subverted into deserialization sinks. This is a secondary, broader characterization of the same issue.
Attack pattern rationale
  • CAPEC-570 The core attack mechanism is that the phar stream wrapper turns a filesystem call on a caller-influenced path into an implicit unserialize() invocation. An attacker crafts a phar archive containing a malicious serialized payload and causes the application to perform a filesystem operation on that path, triggering deserialization and code execution. CAPEC-570 is the closest match. Uncertainty: the exact injection vector (which specific MISP endpoint or console command accepts the path) is not detailed in the patch, but the deserialization sink is explicitly described in the commit message.
  • CAPEC-100 The phar stream wrapper is a trusted, built-in PHP component that the attacker leverages in the MISP runtime environment where it serves no legitimate purpose. The attacker does not need to exploit a flaw in the wrapper itself; rather, its mere presence in the runtime provides the primitive. This is a secondary mapping; CAPEC-570 is preferred as the primary because it more precisely describes the deserialization mechanism.
Assumptions to verify
  • The affected version range is inferred from the tag_version_boundary (v2.5.47, 42 commits after fix), suggesting the fix is included in v2.5.47 and earlier versions are affected. The explicit context lists affected_version and fixed_version as null, so the exact boundary is uncertain.
  • PR:L is assumed because MISP is an authenticated threat-intelligence platform; however, the patch does not explicitly state whether the vulnerable code path requires authentication. If the argument-injection vector is reachable unauthenticated, PR should be N.
  • AC:H is assumed because crafting a valid phar archive and identifying the correct code path for the filesystem call is non-trivial; the commit describes it as an 'argument-injection exploit,' implying specific conditions must be met.
  • The CAPEC-570 mapping is based on the commit message's explicit description of the phar wrapper as an 'unserialize() sink'; the exact MISP endpoint or console command that accepts the attacker-influenced path is not identified in the patch.
  • The commit message references 'the job-argument guard reverted,' implying a prior guard existed and was removed, making the phar wrapper the remaining defense. The exact prior guard is not described in this patch.
  • The Co-Authored-By line lists 'Claude Opus 5 (1M context)' as a co-author; this is recorded in the metadata as a remediation developer but is an AI assistant, not a human contributor. It is excluded from credits to avoid attributing a CVE credit to a non-human entity.
Model comparison

Selected qwen3.8:27b by deterministic-consensus-v1
The selected result is closest to model consensus; this heuristic does not establish factual correctness and human review remains required.

Model Score Agreement Confidence Assumptions
qwen3.8:27b 4 9 medium 6

{
  "containers": {
    "cna": {
      "affected": [
        {
          "modules": [
            "app/Console/cake.php",
            "app/webroot/index.php"
          ],
          "product": "MISP",
          "programFiles": [
            "app/Console/cake.php",
            "app/webroot/index.php"
          ],
          "repo": "https://github.com/MISP/MISP",
          "vendor": "MISP",
          "versions": [
            {
              "lessThan": "2.5.47",
              "status": "affected",
              "version": "0",
              "versionType": "semver"
            }
          ]
        }
      ],
      "credits": [
        {
          "lang": "en",
          "type": "remediation developer",
          "value": "iglocska"
        },
        {
          "lang": "en",
          "type": "remediation developer",
          "value": "Claude Opus 5 (1M context)"
        }
      ],
      "descriptions": [
        {
          "lang": "en",
          "supportingMedia": [
            {
              "base64": false,
              "type": "text/html",
              "value": "\u003cp\u003eMISP ships with PHP\u0027s phar stream wrapper registered in both its web entry point and its console entry point.\u0026nbsp;\u003c/p\u003e\u003cp\u003eThe phar stream wrapper causes PHP to treat a phar archive as a directory, which has two security consequences:\u0026nbsp;\u2003\u003c/p\u003e\u2003-\u0026nbsp;any filesystem operation on a caller-influenced path that resolves to a phar archive triggers an implicit unserialize() call, creating a deserialization sink;\u003cbr\u003e\u003cdiv\u003e\u2003-\u0026nbsp;a relocated application root can reach executable code inside an uploaded phar file, enabling arbitrary code execution as the web user.\u003c/div\u003e\u003cdiv\u003e\u003cbr\u003e\u003c/div\u003e\u003cp\u003eNo component of MISP, the vendored CakePHP framework, or any runtime-loaded library reads or constructs phar archives. The wrapper therefore serves no legitimate purpose in the MISP runtime and exists solely as an available primitive for an attacker who can influence a filesystem path argument.\u0026nbsp;\u003c/p\u003e"
            }
          ],
          "value": "MISP ships with PHP\u0027s phar stream wrapper registered in both its web entry point and its console entry point.\u00a0\n\nThe phar stream wrapper causes PHP to treat a phar archive as a directory, which has two security consequences:\u00a0\u2003\n\n\u2003-\u00a0any filesystem operation on a caller-influenced path that resolves to a phar archive triggers an implicit unserialize() call, creating a deserialization sink;\n\u2003-\u00a0a relocated application root can reach executable code inside an uploaded phar file, enabling arbitrary code execution as the web user.\n\n\n\n\nNo component of MISP, the vendored CakePHP framework, or any runtime-loaded library reads or constructs phar archives. The wrapper therefore serves no legitimate purpose in the MISP runtime and exists solely as an available primitive for an attacker who can influence a filesystem path argument."
        }
      ],
      "impacts": [
        {
          "capecId": "CAPEC-570",
          "descriptions": [
            {
              "lang": "en",
              "value": "CAPEC-570 Deserialization of Untrusted Data"
            }
          ]
        },
        {
          "capecId": "CAPEC-100",
          "descriptions": [
            {
              "lang": "en",
              "value": "CAPEC-100 Leveraging Trusted Components in an Untrusted Environment"
            }
          ]
        }
      ],
      "metrics": [
        {
          "cvssV4_0": {
            "Automatable": "NOT_DEFINED",
            "Recovery": "NOT_DEFINED",
            "Safety": "NOT_DEFINED",
            "attackComplexity": "HIGH",
            "attackRequirements": "NONE",
            "attackVector": "NETWORK",
            "baseScore": 7.7,
            "baseSeverity": "HIGH",
            "privilegesRequired": "LOW",
            "providerUrgency": "NOT_DEFINED",
            "subAvailabilityImpact": "NONE",
            "subConfidentialityImpact": "NONE",
            "subIntegrityImpact": "NONE",
            "userInteraction": "NONE",
            "valueDensity": "NOT_DEFINED",
            "vectorString": "CVSS:4.0/AV:N/AC:H/AT:N/PR:L/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N",
            "version": "4.0",
            "vulnAvailabilityImpact": "HIGH",
            "vulnConfidentialityImpact": "HIGH",
            "vulnIntegrityImpact": "HIGH",
            "vulnerabilityResponseEffort": "NOT_DEFINED"
          },
          "format": "CVSS",
          "scenarios": [
            {
              "lang": "en",
              "value": "GENERAL"
            }
          ]
        }
      ],
      "problemTypes": [
        {
          "descriptions": [
            {
              "cweId": "CWE-502",
              "description": "CWE-502 Deserialization of Untrusted Data",
              "lang": "en",
              "type": "CWE"
            }
          ]
        },
        {
          "descriptions": [
            {
              "cweId": "CWE-74",
              "description": "CWE-74 Improper Neutralization of Special Elements in Output Used by a Downstream Component",
              "lang": "en",
              "type": "CWE"
            }
          ]
        }
      ],
      "providerMetadata": {
        "orgId": "00000000-0000-4000-9000-000000000000"
      },
      "references": [
        {
          "name": "Security patch",
          "tags": [
            "patch"
          ],
          "url": "https://github.com/MISP/MISP/commit/08fa755b6"
        }
      ],
      "solutions": [
        {
          "lang": "en",
          "supportingMedia": [
            {
              "base64": false,
              "type": "text/html",
              "value": "\u003cp\u003eThe phar stream wrapper is unregistered via stream_wrapper_unregister(\u0027phar\u0027) at the top of both the web and console entry points, before any framework bootstrap or application code executes. Because no MISP component, CakePHP, or runtime library requires the phar wrapper, removing it eliminates the implicit unserialize() sink and the directory-like phar archive behavior without functional impact. This closes the deserialization and code-execution primitive application-wide, independent of whether individual callers validate their path arguments.\u003c/p\u003e"
            }
          ],
          "value": "The phar stream wrapper is unregistered via stream_wrapper_unregister(\u0027phar\u0027) at the top of both the web and console entry points, before any framework bootstrap or application code executes. Because no MISP component, CakePHP, or runtime library requires the phar wrapper, removing it eliminates the implicit unserialize() sink and the directory-like phar archive behavior without functional impact. This closes the deserialization and code-execution primitive application-wide, independent of whether individual callers validate their path arguments."
        }
      ],
      "title": "MISP: PHP phar stream wrapper enables deserialization and code execution via caller-influenced filesystem paths",
      "x_gcve": [
        {
          "extensions": {
            "bcp-05-x-01": {
              "ai_annotations": [
                {
                  "ai_level": "generated",
                  "description": "Draft vulnerability metadata was generated from a git-format patch using an Ollama-hosted language model. Human validation is required before publication.",
                  "gna_source": 1,
                  "models": [
                    {
                      "gna_source": 1,
                      "identifier": "qwen3.8:27b",
                      "name": "qwen3.8:27b",
                      "source": "ollama"
                    }
                  ],
                  "review_status": "full",
                  "scope": "record",
                  "tags": [
                    "ai-computer-assisted:llm-generated",
                    "ai-computer-assisted:classification"
                  ]
                }
              ]
            },
            "bcp-05-x-02": {
              "x_patch2vuln": {
                "assumptions": [
                  "The affected version range is inferred from the tag_version_boundary (v2.5.47, 42 commits after fix), suggesting the fix is included in v2.5.47 and earlier versions are affected. The explicit context lists affected_version and fixed_version as null, so the exact boundary is uncertain.",
                  "PR:L is assumed because MISP is an authenticated threat-intelligence platform; however, the patch does not explicitly state whether the vulnerable code path requires authentication. If the argument-injection vector is reachable unauthenticated, PR should be N.",
                  "AC:H is assumed because crafting a valid phar archive and identifying the correct code path for the filesystem call is non-trivial; the commit describes it as an \u0027argument-injection exploit,\u0027 implying specific conditions must be met.",
                  "The CAPEC-570 mapping is based on the commit message\u0027s explicit description of the phar wrapper as an \u0027unserialize() sink\u0027; the exact MISP endpoint or console command that accepts the attacker-influenced path is not identified in the patch.",
                  "The commit message references \u0027the job-argument guard reverted,\u0027 implying a prior guard existed and was removed, making the phar wrapper the remaining defense. The exact prior guard is not described in this patch.",
                  "The Co-Authored-By line lists \u0027Claude Opus 5 (1M context)\u0027 as a co-author; this is recorded in the metadata as a remediation developer but is an AI assistant, not a human contributor. It is excluded from credits to avoid attributing a CVE credit to a non-human entity."
                ],
                "capecRationale": [
                  {
                    "capecId": "CAPEC-570",
                    "rationale": "The core attack mechanism is that the phar stream wrapper turns a filesystem call on a caller-influenced path into an implicit unserialize() invocation. An attacker crafts a phar archive containing a malicious serialized payload and causes the application to perform a filesystem operation on that path, triggering deserialization and code execution. CAPEC-570 is the closest match. Uncertainty: the exact injection vector (which specific MISP endpoint or console command accepts the path) is not detailed in the patch, but the deserialization sink is explicitly described in the commit message."
                  },
                  {
                    "capecId": "CAPEC-100",
                    "rationale": "The phar stream wrapper is a trusted, built-in PHP component that the attacker leverages in the MISP runtime environment where it serves no legitimate purpose. The attacker does not need to exploit a flaw in the wrapper itself; rather, its mere presence in the runtime provides the primitive. This is a secondary mapping; CAPEC-570 is preferred as the primary because it more precisely describes the deserialization mechanism."
                  }
                ],
                "commit": "08fa755b6363ec0b7194d97ea36800ae8eb8f919",
                "confidence": "medium",
                "credits": [
                  {
                    "lang": "en",
                    "type": "remediation developer",
                    "value": "iglocska"
                  },
                  {
                    "lang": "en",
                    "type": "remediation developer",
                    "value": "Claude Opus 5 (1M context)"
                  }
                ],
                "cvssRationale": "AV:N: The vulnerability is reachable through the web entry point (app/webroot/index.php), making it network-accessible. AC:H: Exploitation requires crafting a valid phar archive with a malicious serialized payload and identifying a code path where a caller-influenced filesystem argument resolves to that archive; the commit describes this as an \u0027argument-injection exploit,\u0027 implying non-trivial path manipulation. AT:N: No specific attack-target conditions are indicated. PR:L: MISP is a threat-intelligence platform that typically requires authenticated access; the commit references \u0027the web user\u0027 context, suggesting the attacker operates within the application\u0027s privilege boundary. UI:N: No user interaction is required. VC/VI/VA:H: Successful exploitation results in arbitrary code execution as the web user, compromising confidentiality, integrity, and availability of the MISP instance. SC/SI/SA:N: No evidence of impact on subsequent or other components beyond the MISP instance itself.",
                "fixSummary": "The phar stream wrapper is unregistered via stream_wrapper_unregister(\u0027phar\u0027) at the top of both the web and console entry points, before any framework bootstrap or application code executes. Because no MISP component, CakePHP, or runtime library requires the phar wrapper, removing it eliminates the implicit unserialize() sink and the directory-like phar archive behavior without functional impact. This closes the deserialization and code-execution primitive application-wide, independent of whether individual callers validate their path arguments.",
                "generatedAt": "2026-09-22T15:01:21.211955Z",
                "generator": "patch2vuln.py",
                "model": "qwen3.8:27b",
                "modelComparison": {
                  "rankings": [
                    {
                      "agreementScore": 9,
                      "assumptionCount": 6,
                      "confidence": "medium",
                      "model": "qwen3.8:27b",
                      "score": 4
                    }
                  ],
                  "selectedModel": "qwen3.8:27b",
                  "selectionMethod": "deterministic-consensus-v1",
                  "selectionNotice": "The selected result is closest to model consensus; this heuristic does not establish factual correctness and human review remains required."
                },
                "patchSha256": "f4fc3223e6421557e1c03c2182cfaa2b0c17a1a20870c71214708034faf1350f",
                "patchSummary": "Two files are modified. In app/Console/cake.php and app/webroot/index.php, a 14-line block is inserted immediately after the file header comment and before any existing logic. The block checks whether \u0027phar\u0027 is present in stream_get_wrappers() and, if so, calls stream_wrapper_unregister(\u0027phar\u0027). A docblock comment explains the security rationale: the wrapper turns filesystem calls on caller-influenced paths into unserialize() sinks and allows a relocated application root to reach code inside an uploaded file. No other code is changed; the total diff is 28 insertions across the two files.",
                "patchTruncated": false,
                "patches": [
                  {
                    "commit": "08fa755b6363ec0b7194d97ea36800ae8eb8f919",
                    "patchSha256": "f4fc3223e6421557e1c03c2182cfaa2b0c17a1a20870c71214708034faf1350f",
                    "source": "https://github.com/MISP/MISP/commit/08fa755b6.patch",
                    "sourceUrl": "https://github.com/MISP/MISP/commit/08fa755b6.patch",
                    "subject": "chg: [internal] Unregister the phar stream wrapper in the"
                  }
                ],
                "source": "https://github.com/MISP/MISP/commit/08fa755b6.patch",
                "subject": "chg: [internal] Unregister the phar stream wrapper in the",
                "tagVersionBoundary": {
                  "commits_after_fix": 42,
                  "repository": "https://github.com/MISP/MISP",
                  "tag": "v2.5.47",
                  "version": "2.5.47",
                  "version_type": "semver"
                },
                "weaknessRationale": [
                  {
                    "cweId": "CWE-502",
                    "rationale": "The phar stream wrapper causes PHP to invoke unserialize() implicitly whenever a filesystem operation resolves to a phar archive. An attacker who can influence the path argument (e.g., via argument injection) can supply a crafted phar file, triggering deserialization of attacker-controlled data and leading to code execution. This is the primary and most specific weakness."
                  },
                  {
                    "cweId": "CWE-74",
                    "rationale": "The phar stream wrapper is a special element of the PHP runtime that was not neutralized (unregistered) in the MISP entry points. Its presence allows downstream filesystem calls to be subverted into deserialization sinks. This is a secondary, broader characterization of the same issue."
                  }
                ]
              }
            }
          },
          "recordType": "advisory",
          "vulnId": "GCVE-1-2026-20263"
        }
      ]
    }
  },
  "cveMetadata": {
    "assignerOrgId": "00000000-0000-4000-9000-000000000000",
    "cveId": "CVE-2026-95806",
    "datePublished": "2026-09-22T15:08:51.542843Z",
    "dateReserved": "2026-09-22T15:09:04.977Z",
    "dateUpdated": "2026-09-22T15:09:05.089365Z",
    "state": "PUBLISHED",
    "vulnId": "GCVE-1-2026-20263"
  },
  "dataType": "CVE_RECORD",
  "dataVersion": "5.1"
}

GCVE-1-2026-20240 (CVE-2026-95805)

Vulnerability from gna-1 – Published: 2026-09-22 14:59 – Updated: 2026-09-22 14:59
VLAI
Title
MISP ACLComponent: Typo in previewEventAttributes ACL key bypasses intended access restriction
Summary
A typo in the MISP ACLComponent access control configuration caused the ACL rule for the previewEventAttributes action to reference the permission string 'theming_enabled*' (with a trailing asterisk) instead of the correct 'theming_enabled'. In the MISP ACL system, the array values define which role or permission grants access to a given controller action. The adjacent entry previewEventObjects correctly uses ['theming_enabled'], confirming the intended restriction. The malformed key 'theming_enabled*' does not match any valid permission identifier, causing the access control check for previewEventAttributes to malfunction. Depending on the ACL evaluation logic, this could result in either unauthorized users gaining access to the previewEventAttributes endpoint (authorization bypass) or legitimate users being denied access (availability impact).  The previewEventAttributes endpoint exposes event attribute data within MISP so an authorization bypass could expose sensitive indicator and attribute data to users who should not have access.
CWE
Assigner
CIRCL GNA GNA-1
GNA scorecard A 91/100 over 279 records in the last 180 days details
References
Impacted products
Vendor Product Version
MISP MISP Affected: 0 , < 2.5.47 (semver)
Create a notification for this product.
GCVE extensions
bcp-05-x-01
AI-assisted vulnerability information annotation
GCVE-BCP-05-X-01
Whole record AI-generated Human-reviewed GNA-1

Draft vulnerability metadata was generated from a git-format patch using an Ollama-hosted language model. Human validation is required before publication.

ai-computer-assisted:llm-generatedai-computer-assisted:classification
Model Source Identifier
qwen3.8:27b ollama qwen3.8:27b
bcp-05-x-02
Patch-to-vulnerability generation provenance
GCVE-BCP-05-X-02
Generator
patch2vuln.py on 2026-09-22 14:55
Model
qwen3.8:27b
Input
https://github.com/MISP/MISP/commit/5ac8d1e4d.patch c102952ebfab…
Confidence
medium
Commit Subject Patch SHA-256
5ac8d1e4dea7 fix: [ACL] typo in previewEventAttributes key c102952ebfab…
Fix summary

The fix corrects the ACL permission key for the previewEventAttributes action from the malformed string 'theming_enabled*' to the correct 'theming_enabled', restoring the intended access control restriction so that only users holding the theming_enabled permission can invoke the endpoint, consistent with the adjacent previewEventObjects entry.

Patch summary

In app/Controller/Component/ACLComponent.php (line 473), the ACL array value for the 'previewEventAttributes' key was changed from ['theming_enabled*'] to ['theming_enabled'], removing the erroneous trailing asterisk that made the permission string non-matching. One line changed, one insertion, one deletion.

CVSS rationale

AV:N: MISP is a network-accessible web application. AC:L: The vulnerability is a static typo in configuration; no race or complex condition is needed. AT:N: No attack target manipulation required. PR:L: The attacker must be an authenticated MISP user to reach the endpoint. UI:N: No victim interaction needed. VC:L: If the ACL bypass is confirmed, an unauthorized user could view event attribute data (limited confidentiality impact). VI:N and VA:N: No integrity or availability impact is evident from the patch. SC/SI/SA:N: No secondary system impact. The overall severity is Low, reflecting a single-endpoint access control misconfiguration with limited data exposure.

Weakness rationale
  • CWE-285 The ACL rule for previewEventAttributes contained a typo ('theming_enabled*') that prevented the authorization check from matching the intended permission, causing the access control mechanism to not enforce the restriction as designed. This is a direct failure of the authorization logic due to a misconfigured rule.
Attack pattern rationale
  • CAPEC-126 The closest plausible attack pattern is Forced Browsing, where an attacker accesses a resource (the previewEventAttributes endpoint) that should be restricted by the ACL. The typo in the ACL key may cause the authorization check to fail to deny access, effectively allowing a user without the theming_enabled permission to reach the endpoint. Uncertainty: the exact ACL evaluation behavior on a non-matching rule (default-deny vs. default-allow) is not visible in the patch, so it is also possible the typo causes a denial rather than a bypass. CAPEC-126 is selected as the best available match for an access-control bypass via misconfiguration.
Assumptions to verify
  • The exact ACL evaluation behavior when a permission string does not match any valid role (default-deny vs. default-allow) is not visible in the patch; the CVSS assumes the more security-relevant interpretation (authorization bypass) but a denial-of-service interpretation is also possible.
  • The affected version range is inferred from the tag boundary v2.5.47 (227 commits after the fix); the exact first affected version is not stated in the patch.
  • The CAPEC-126 mapping is the closest available pattern; the actual attack is a configuration typo rather than an active browsing technique, so the mapping is approximate.
  • The security impact is assumed to be limited to the previewEventAttributes endpoint; no evidence in the patch suggests other endpoints or data stores are affected.
  • The CVSS assumes the endpoint exposes event attribute data viewable by unauthorized users; the actual data sensitivity depends on MISP deployment configuration.
Model comparison

Selected qwen3.8:27b by deterministic-consensus-v1
The selected result is closest to model consensus; this heuristic does not establish factual correctness and human review remains required.

Model Score Agreement Confidence Assumptions
qwen3.8:27b 5 9 medium 5

{
  "containers": {
    "cna": {
      "affected": [
        {
          "modules": [
            "ACLComponent"
          ],
          "product": "MISP",
          "programFiles": [
            "app/Controller/Component/ACLComponent.php"
          ],
          "repo": "https://github.com/MISP/MISP",
          "vendor": "MISP",
          "versions": [
            {
              "lessThan": "2.5.47",
              "status": "affected",
              "version": "0",
              "versionType": "semver"
            }
          ]
        }
      ],
      "credits": [
        {
          "lang": "en",
          "type": "remediation developer",
          "value": "Thomas Lacroix"
        }
      ],
      "descriptions": [
        {
          "lang": "en",
          "supportingMedia": [
            {
              "base64": false,
              "type": "text/html",
              "value": "\u003cp\u003eA typo in the MISP ACLComponent access control configuration caused the ACL rule for the previewEventAttributes action to reference the permission string \u0027theming_enabled*\u0027 (with a trailing asterisk) instead of the correct \u0027theming_enabled\u0027. In the MISP ACL system, the array values define which role or permission grants access to a given controller action. The adjacent entry previewEventObjects correctly uses [\u0027theming_enabled\u0027], confirming the intended restriction. The malformed key \u0027theming_enabled*\u0027 does not match any valid permission identifier, causing the access control check for previewEventAttributes to malfunction. Depending on the ACL evaluation logic, this could result in either unauthorized users gaining access to the previewEventAttributes endpoint (authorization bypass) or legitimate users being denied access (availability impact).\u0026nbsp;\u003c/p\u003e\u003cp\u003eThe previewEventAttributes endpoint exposes event attribute data within MISP so an authorization bypass could expose sensitive indicator and attribute data to users who should not have access.\u003c/p\u003e"
            }
          ],
          "value": "A typo in the MISP ACLComponent access control configuration caused the ACL rule for the previewEventAttributes action to reference the permission string \u0027theming_enabled*\u0027 (with a trailing asterisk) instead of the correct \u0027theming_enabled\u0027. In the MISP ACL system, the array values define which role or permission grants access to a given controller action. The adjacent entry previewEventObjects correctly uses [\u0027theming_enabled\u0027], confirming the intended restriction. The malformed key \u0027theming_enabled*\u0027 does not match any valid permission identifier, causing the access control check for previewEventAttributes to malfunction. Depending on the ACL evaluation logic, this could result in either unauthorized users gaining access to the previewEventAttributes endpoint (authorization bypass) or legitimate users being denied access (availability impact).\u00a0\n\nThe previewEventAttributes endpoint exposes event attribute data within MISP so an authorization bypass could expose sensitive indicator and attribute data to users who should not have access."
        }
      ],
      "impacts": [
        {
          "capecId": "CAPEC-126",
          "descriptions": [
            {
              "lang": "en",
              "value": "CAPEC-126 Forced Browsing"
            }
          ]
        }
      ],
      "metrics": [
        {
          "cvssV4_0": {
            "Automatable": "NOT_DEFINED",
            "Recovery": "NOT_DEFINED",
            "Safety": "NOT_DEFINED",
            "attackComplexity": "LOW",
            "attackRequirements": "NONE",
            "attackVector": "NETWORK",
            "baseScore": 5.3,
            "baseSeverity": "MEDIUM",
            "privilegesRequired": "LOW",
            "providerUrgency": "NOT_DEFINED",
            "subAvailabilityImpact": "NONE",
            "subConfidentialityImpact": "NONE",
            "subIntegrityImpact": "NONE",
            "userInteraction": "NONE",
            "valueDensity": "NOT_DEFINED",
            "vectorString": "CVSS:4.0/AV:N/AC:L/AT:N/PR:L/UI:N/VC:L/VI:N/VA:N/SC:N/SI:N/SA:N",
            "version": "4.0",
            "vulnAvailabilityImpact": "NONE",
            "vulnConfidentialityImpact": "LOW",
            "vulnIntegrityImpact": "NONE",
            "vulnerabilityResponseEffort": "NOT_DEFINED"
          },
          "format": "CVSS",
          "scenarios": [
            {
              "lang": "en",
              "value": "GENERAL"
            }
          ]
        }
      ],
      "problemTypes": [
        {
          "descriptions": [
            {
              "cweId": "CWE-285",
              "description": "CWE-285 Improper Authorization",
              "lang": "en",
              "type": "CWE"
            }
          ]
        }
      ],
      "providerMetadata": {
        "orgId": "00000000-0000-4000-9000-000000000000"
      },
      "references": [
        {
          "name": "Security patch",
          "tags": [
            "patch"
          ],
          "url": "https://github.com/MISP/MISP/commit/5ac8d1e4d"
        }
      ],
      "solutions": [
        {
          "lang": "en",
          "supportingMedia": [
            {
              "base64": false,
              "type": "text/html",
              "value": "\u003cp\u003eThe fix corrects the ACL permission key for the previewEventAttributes action from the malformed string \u0027theming_enabled*\u0027 to the correct \u0027theming_enabled\u0027, restoring the intended access control restriction so that only users holding the theming_enabled permission can invoke the endpoint, consistent with the adjacent previewEventObjects entry.\u003c/p\u003e"
            }
          ],
          "value": "The fix corrects the ACL permission key for the previewEventAttributes action from the malformed string \u0027theming_enabled*\u0027 to the correct \u0027theming_enabled\u0027, restoring the intended access control restriction so that only users holding the theming_enabled permission can invoke the endpoint, consistent with the adjacent previewEventObjects entry."
        }
      ],
      "title": "MISP ACLComponent: Typo in previewEventAttributes ACL key bypasses intended access restriction",
      "x_gcve": [
        {
          "extensions": {
            "bcp-05-x-01": {
              "ai_annotations": [
                {
                  "ai_level": "generated",
                  "description": "Draft vulnerability metadata was generated from a git-format patch using an Ollama-hosted language model. Human validation is required before publication.",
                  "gna_source": 1,
                  "models": [
                    {
                      "gna_source": 1,
                      "identifier": "qwen3.8:27b",
                      "name": "qwen3.8:27b",
                      "source": "ollama"
                    }
                  ],
                  "review_status": "full",
                  "scope": "record",
                  "tags": [
                    "ai-computer-assisted:llm-generated",
                    "ai-computer-assisted:classification"
                  ]
                }
              ]
            },
            "bcp-05-x-02": {
              "x_patch2vuln": {
                "assumptions": [
                  "The exact ACL evaluation behavior when a permission string does not match any valid role (default-deny vs. default-allow) is not visible in the patch; the CVSS assumes the more security-relevant interpretation (authorization bypass) but a denial-of-service interpretation is also possible.",
                  "The affected version range is inferred from the tag boundary v2.5.47 (227 commits after the fix); the exact first affected version is not stated in the patch.",
                  "The CAPEC-126 mapping is the closest available pattern; the actual attack is a configuration typo rather than an active browsing technique, so the mapping is approximate.",
                  "The security impact is assumed to be limited to the previewEventAttributes endpoint; no evidence in the patch suggests other endpoints or data stores are affected.",
                  "The CVSS assumes the endpoint exposes event attribute data viewable by unauthorized users; the actual data sensitivity depends on MISP deployment configuration."
                ],
                "capecRationale": [
                  {
                    "capecId": "CAPEC-126",
                    "rationale": "The closest plausible attack pattern is Forced Browsing, where an attacker accesses a resource (the previewEventAttributes endpoint) that should be restricted by the ACL. The typo in the ACL key may cause the authorization check to fail to deny access, effectively allowing a user without the theming_enabled permission to reach the endpoint. Uncertainty: the exact ACL evaluation behavior on a non-matching rule (default-deny vs. default-allow) is not visible in the patch, so it is also possible the typo causes a denial rather than a bypass. CAPEC-126 is selected as the best available match for an access-control bypass via misconfiguration."
                  }
                ],
                "commit": "5ac8d1e4dea72f71bec3789350b748f0ad21c42c",
                "confidence": "medium",
                "credits": [
                  {
                    "lang": "en",
                    "type": "remediation developer",
                    "value": "Thomas Lacroix"
                  }
                ],
                "cvssRationale": "AV:N: MISP is a network-accessible web application. AC:L: The vulnerability is a static typo in configuration; no race or complex condition is needed. AT:N: No attack target manipulation required. PR:L: The attacker must be an authenticated MISP user to reach the endpoint. UI:N: No victim interaction needed. VC:L: If the ACL bypass is confirmed, an unauthorized user could view event attribute data (limited confidentiality impact). VI:N and VA:N: No integrity or availability impact is evident from the patch. SC/SI/SA:N: No secondary system impact. The overall severity is Low, reflecting a single-endpoint access control misconfiguration with limited data exposure.",
                "fixSummary": "The fix corrects the ACL permission key for the previewEventAttributes action from the malformed string \u0027theming_enabled*\u0027 to the correct \u0027theming_enabled\u0027, restoring the intended access control restriction so that only users holding the theming_enabled permission can invoke the endpoint, consistent with the adjacent previewEventObjects entry.",
                "generatedAt": "2026-09-22T14:55:18.958107Z",
                "generator": "patch2vuln.py",
                "model": "qwen3.8:27b",
                "modelComparison": {
                  "rankings": [
                    {
                      "agreementScore": 9,
                      "assumptionCount": 5,
                      "confidence": "medium",
                      "model": "qwen3.8:27b",
                      "score": 5
                    }
                  ],
                  "selectedModel": "qwen3.8:27b",
                  "selectionMethod": "deterministic-consensus-v1",
                  "selectionNotice": "The selected result is closest to model consensus; this heuristic does not establish factual correctness and human review remains required."
                },
                "patchSha256": "c102952ebfaba16a0d028044d1bc0ab063381442b99de01bfc592144f2de9cc5",
                "patchSummary": "In app/Controller/Component/ACLComponent.php (line 473), the ACL array value for the \u0027previewEventAttributes\u0027 key was changed from [\u0027theming_enabled*\u0027] to [\u0027theming_enabled\u0027], removing the erroneous trailing asterisk that made the permission string non-matching. One line changed, one insertion, one deletion.",
                "patchTruncated": false,
                "patches": [
                  {
                    "commit": "5ac8d1e4dea72f71bec3789350b748f0ad21c42c",
                    "patchSha256": "c102952ebfaba16a0d028044d1bc0ab063381442b99de01bfc592144f2de9cc5",
                    "source": "https://github.com/MISP/MISP/commit/5ac8d1e4d.patch",
                    "sourceUrl": "https://github.com/MISP/MISP/commit/5ac8d1e4d.patch",
                    "subject": "fix: [ACL] typo in previewEventAttributes key"
                  }
                ],
                "source": "https://github.com/MISP/MISP/commit/5ac8d1e4d.patch",
                "subject": "fix: [ACL] typo in previewEventAttributes key",
                "tagVersionBoundary": {
                  "commits_after_fix": 227,
                  "repository": "https://github.com/MISP/MISP",
                  "tag": "v2.5.47",
                  "version": "2.5.47",
                  "version_type": "semver"
                },
                "weaknessRationale": [
                  {
                    "cweId": "CWE-285",
                    "rationale": "The ACL rule for previewEventAttributes contained a typo (\u0027theming_enabled*\u0027) that prevented the authorization check from matching the intended permission, causing the access control mechanism to not enforce the restriction as designed. This is a direct failure of the authorization logic due to a misconfigured rule."
                  }
                ]
              }
            }
          },
          "recordType": "advisory",
          "vulnId": "GCVE-1-2026-20240"
        }
      ]
    }
  },
  "cveMetadata": {
    "assignerOrgId": "00000000-0000-4000-9000-000000000000",
    "cveId": "CVE-2026-95805",
    "datePublished": "2026-09-22T14:59:06.674611Z",
    "dateReserved": "2026-09-22T14:59:19.797Z",
    "dateUpdated": "2026-09-22T14:59:19.861474Z",
    "state": "PUBLISHED",
    "vulnId": "GCVE-1-2026-20240"
  },
  "dataType": "CVE_RECORD",
  "dataVersion": "5.1"
}

GCVE-1-2026-20024 (CVE-2026-95754)

Vulnerability from gna-1 – Published: 2026-09-22 14:53 – Updated: 2026-09-22 14:53
VLAI
Title
MISP: Disabled-user check ineffective in pre-authentication TOTP login branch
Summary
In MISP's UsersController login() method, the pre-authentication database query used for the TOTP (two-factor authentication) verification branch did not include the User.disabled column in its SELECT fields list. The query selected only User.password, User.totp, and User.hotp_counter. When the TOTP branch subsequently accessed $unauth_user['User']['disabled'], the key was absent from the result set, producing a PHP 'Undefined array key' warning and causing the expression to evaluate as null (falsy). As a result, the disabled-user guard in the TOTP branch was effectively a no-op: a disabled, TOTP-enrolled user could proceed to the TOTP verification step rather than being rejected at that point.  The commit message explicitly states this was 'harmless in practice' because the subsequent identify() call re-validates the user and would still reject a disabled account.  The practical security impact is therefore minimal, limited to a very small information-disclosure difference in the login response (a TOTP prompt is presented instead of an immediate rejection) and a PHP warning in application logs.
CWE
  • CWE-285 - Improper Authentication
Assigner
CIRCL GNA GNA-1
GNA scorecard A 91/100 over 279 records in the last 180 days details
References
Impacted products
Vendor Product Version
MISP MISP Affected: 0 , < 2.5.47 (semver)
Create a notification for this product.
GCVE extensions
bcp-05-x-01
AI-assisted vulnerability information annotation
GCVE-BCP-05-X-01
Whole record AI-generated Human-reviewed GNA-1

Draft vulnerability metadata was generated from a git-format patch using an Ollama-hosted language model. Human validation is required before publication.

ai-computer-assisted:llm-generatedai-computer-assisted:classification
Model Source Identifier
qwen3.8:27b ollama qwen3.8:27b
bcp-05-x-02
Patch-to-vulnerability generation provenance
GCVE-BCP-05-X-02
Generator
patch2vuln.py on 2026-09-22 14:51
Model
qwen3.8:27b
Input
https://github.com/MISP/MISP/commit/3df982ab1.patch 8ddba4461463…
Confidence
medium
Commit Subject Patch SHA-256
3df982ab192e fix: [login] Fetch User.disabled in the pre-auth TOTP lookup 8ddba4461463…
Fix summary

The fix adds User.disabled to the fields array of the pre-authentication find() query so that the disabled-user check in the TOTP branch reads the actual column value and correctly rejects disabled users before they reach the TOTP verification step, restoring the intended guard behavior.

Patch summary

A single-line change in app/Controller/UsersController.php appends 'User.disabled' to the fields list of the pre-authentication User find() query in the login() method (line 1385), ensuring the disabled flag is present in the result set when the TOTP branch evaluates it.

CVSS rationale

The vulnerability is reachable over the network (AV:N) via the web login endpoint with low complexity (AC:L), no attack target (AT:N), no prior privileges (PR:N), and no user interaction (UI:N). The only potential impact is a very minor confidentiality difference: a disabled, TOTP-enrolled user receives a TOTP prompt instead of an immediate rejection, which could marginally aid account-status enumeration (VC:L). There is no integrity or availability impact (VI:N, VA:N) and no secondary impacts (SC:N, SI:N, SA:N) because the subsequent identify() call still enforces the disabled-user check. The commit message explicitly characterizes the issue as 'harmless in practice.'

Weakness rationale
  • CWE-285 The disabled-user check in the TOTP login branch was ineffective because the required column was not fetched, allowing a disabled user to pass that guard. The subsequent identify() call still enforced the check, limiting the impact. CWE-285 is the narrowest defensible mapping for an authentication-state check that does not function as intended.
Attack pattern rationale
  • CAPEC-1 The closest plausible CAPEC is Brute Force, as the scenario involves an attacker attempting to authenticate using known credentials of a disabled, TOTP-enrolled account. The disabled-user guard that should have blocked the attempt at the TOTP branch was ineffective, allowing the attempt to proceed one step further than intended. This mapping is uncertain because the vulnerability does not enable a full authentication bypass (identify() still rejects the user) and the primary effect is a minor information-disclosure difference in the login response rather than a successful brute-force attack.
Assumptions to verify
  • The commit message states the issue is 'harmless in practice' because identify() re-checks the user; this advisory treats the security impact as minimal (minor information disclosure only).
  • The TOTP feature was introduced in 2023 per the commit message; the exact affected version range is not specified in the patch. The tag boundary v2.5.47 is used as an upper bound for the affected range but is not confirmed as the exact fixed version.
  • CAPEC-1 (Brute Force) is the closest available attack pattern; the actual scenario is a minor authentication-guard bypass rather than a full brute-force attack, so the mapping is approximate.
  • CVSS VC:L is assigned for the marginal information-disclosure difference in the login response; if the CNA determines no meaningful confidentiality impact exists, VC:N would be more appropriate, resulting in a score of 0.0.
  • The PHP version behavior (Warning vs Notice vs Error for undefined array keys) may vary, but in all supported PHP versions the code continues execution with a null value, so the security analysis is unaffected.
Model comparison

Selected qwen3.8:27b by deterministic-consensus-v1
The selected result is closest to model consensus; this heuristic does not establish factual correctness and human review remains required.

Model Score Agreement Confidence Assumptions
qwen3.8:27b 5 9 medium 5

{
  "containers": {
    "cna": {
      "affected": [
        {
          "modules": [
            "UsersController (login method",
            "TOTP branch)"
          ],
          "product": "MISP",
          "programFiles": [
            "app/Controller/UsersController.php"
          ],
          "repo": "https://github.com/MISP/MISP",
          "vendor": "MISP",
          "versions": [
            {
              "lessThan": "2.5.47",
              "status": "affected",
              "version": "0",
              "versionType": "semver"
            }
          ]
        }
      ],
      "credits": [
        {
          "lang": "en",
          "type": "remediation developer",
          "value": "iglocska"
        },
        {
          "lang": "en",
          "type": "remediation developer",
          "value": "Claude Fable 5.1"
        }
      ],
      "descriptions": [
        {
          "lang": "en",
          "supportingMedia": [
            {
              "base64": false,
              "type": "text/html",
              "value": "\u003cp\u003eIn MISP\u0027s UsersController login() method, the pre-authentication database query used for the TOTP (two-factor authentication) verification branch did not include the User.disabled column in its SELECT fields list. The query selected only User.password, User.totp, and User.hotp_counter. When the TOTP branch subsequently accessed $unauth_user[\u0027User\u0027][\u0027disabled\u0027], the key was absent from the result set, producing a PHP \u0027Undefined array key\u0027 warning and causing the expression to evaluate as null (falsy). As a result, the disabled-user guard in the TOTP branch was effectively a no-op: a disabled, TOTP-enrolled user could proceed to the TOTP verification step rather than being rejected at that point.\u0026nbsp;\u003c/p\u003e\u003cp\u003eThe commit message explicitly states this was \u0027harmless in practice\u0027 because the subsequent identify() call re-validates the user and would still reject a disabled account.\u0026nbsp;\u003c/p\u003e\u003cp\u003eThe practical security impact is therefore minimal, limited to a very small information-disclosure difference in the login response (a TOTP prompt is presented instead of an immediate rejection) and a PHP warning in application logs.\u003c/p\u003e"
            }
          ],
          "value": "In MISP\u0027s UsersController login() method, the pre-authentication database query used for the TOTP (two-factor authentication) verification branch did not include the User.disabled column in its SELECT fields list. The query selected only User.password, User.totp, and User.hotp_counter. When the TOTP branch subsequently accessed $unauth_user[\u0027User\u0027][\u0027disabled\u0027], the key was absent from the result set, producing a PHP \u0027Undefined array key\u0027 warning and causing the expression to evaluate as null (falsy). As a result, the disabled-user guard in the TOTP branch was effectively a no-op: a disabled, TOTP-enrolled user could proceed to the TOTP verification step rather than being rejected at that point.\u00a0\n\nThe commit message explicitly states this was \u0027harmless in practice\u0027 because the subsequent identify() call re-validates the user and would still reject a disabled account.\u00a0\n\nThe practical security impact is therefore minimal, limited to a very small information-disclosure difference in the login response (a TOTP prompt is presented instead of an immediate rejection) and a PHP warning in application logs."
        }
      ],
      "impacts": [
        {
          "capecId": "CAPEC-1",
          "descriptions": [
            {
              "lang": "en",
              "value": "CAPEC-1 Brute Force"
            }
          ]
        }
      ],
      "metrics": [
        {
          "cvssV4_0": {
            "Automatable": "NOT_DEFINED",
            "Recovery": "NOT_DEFINED",
            "Safety": "NOT_DEFINED",
            "attackComplexity": "LOW",
            "attackRequirements": "NONE",
            "attackVector": "NETWORK",
            "baseScore": 6.9,
            "baseSeverity": "MEDIUM",
            "privilegesRequired": "NONE",
            "providerUrgency": "NOT_DEFINED",
            "subAvailabilityImpact": "NONE",
            "subConfidentialityImpact": "NONE",
            "subIntegrityImpact": "NONE",
            "userInteraction": "NONE",
            "valueDensity": "NOT_DEFINED",
            "vectorString": "CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:L/VI:N/VA:N/SC:N/SI:N/SA:N",
            "version": "4.0",
            "vulnAvailabilityImpact": "NONE",
            "vulnConfidentialityImpact": "LOW",
            "vulnIntegrityImpact": "NONE",
            "vulnerabilityResponseEffort": "NOT_DEFINED"
          },
          "format": "CVSS",
          "scenarios": [
            {
              "lang": "en",
              "value": "GENERAL"
            }
          ]
        }
      ],
      "problemTypes": [
        {
          "descriptions": [
            {
              "cweId": "CWE-285",
              "description": "CWE-285 Improper Authentication",
              "lang": "en",
              "type": "CWE"
            }
          ]
        }
      ],
      "providerMetadata": {
        "orgId": "00000000-0000-4000-9000-000000000000"
      },
      "references": [
        {
          "name": "Security patch",
          "tags": [
            "patch"
          ],
          "url": "https://github.com/MISP/MISP/commit/3df982ab1"
        }
      ],
      "solutions": [
        {
          "lang": "en",
          "supportingMedia": [
            {
              "base64": false,
              "type": "text/html",
              "value": "\u003cp\u003eThe fix adds User.disabled to the fields array of the pre-authentication find() query so that the disabled-user check in the TOTP branch reads the actual column value and correctly rejects disabled users before they reach the TOTP verification step, restoring the intended guard behavior.\u003c/p\u003e"
            }
          ],
          "value": "The fix adds User.disabled to the fields array of the pre-authentication find() query so that the disabled-user check in the TOTP branch reads the actual column value and correctly rejects disabled users before they reach the TOTP verification step, restoring the intended guard behavior."
        }
      ],
      "title": "MISP: Disabled-user check ineffective in pre-authentication TOTP login branch",
      "x_gcve": [
        {
          "extensions": {
            "bcp-05-x-01": {
              "ai_annotations": [
                {
                  "ai_level": "generated",
                  "description": "Draft vulnerability metadata was generated from a git-format patch using an Ollama-hosted language model. Human validation is required before publication.",
                  "gna_source": 1,
                  "models": [
                    {
                      "gna_source": 1,
                      "identifier": "qwen3.8:27b",
                      "name": "qwen3.8:27b",
                      "source": "ollama"
                    }
                  ],
                  "review_status": "full",
                  "scope": "record",
                  "tags": [
                    "ai-computer-assisted:llm-generated",
                    "ai-computer-assisted:classification"
                  ]
                }
              ]
            },
            "bcp-05-x-02": {
              "x_patch2vuln": {
                "assumptions": [
                  "The commit message states the issue is \u0027harmless in practice\u0027 because identify() re-checks the user; this advisory treats the security impact as minimal (minor information disclosure only).",
                  "The TOTP feature was introduced in 2023 per the commit message; the exact affected version range is not specified in the patch. The tag boundary v2.5.47 is used as an upper bound for the affected range but is not confirmed as the exact fixed version.",
                  "CAPEC-1 (Brute Force) is the closest available attack pattern; the actual scenario is a minor authentication-guard bypass rather than a full brute-force attack, so the mapping is approximate.",
                  "CVSS VC:L is assigned for the marginal information-disclosure difference in the login response; if the CNA determines no meaningful confidentiality impact exists, VC:N would be more appropriate, resulting in a score of 0.0.",
                  "The PHP version behavior (Warning vs Notice vs Error for undefined array keys) may vary, but in all supported PHP versions the code continues execution with a null value, so the security analysis is unaffected."
                ],
                "capecRationale": [
                  {
                    "capecId": "CAPEC-1",
                    "rationale": "The closest plausible CAPEC is Brute Force, as the scenario involves an attacker attempting to authenticate using known credentials of a disabled, TOTP-enrolled account. The disabled-user guard that should have blocked the attempt at the TOTP branch was ineffective, allowing the attempt to proceed one step further than intended. This mapping is uncertain because the vulnerability does not enable a full authentication bypass (identify() still rejects the user) and the primary effect is a minor information-disclosure difference in the login response rather than a successful brute-force attack."
                  }
                ],
                "commit": "3df982ab192e05f5369d62af5674c7a874b32756",
                "confidence": "medium",
                "credits": [
                  {
                    "lang": "en",
                    "type": "remediation developer",
                    "value": "iglocska"
                  },
                  {
                    "lang": "en",
                    "type": "remediation developer",
                    "value": "Claude Fable 5.1"
                  }
                ],
                "cvssRationale": "The vulnerability is reachable over the network (AV:N) via the web login endpoint with low complexity (AC:L), no attack target (AT:N), no prior privileges (PR:N), and no user interaction (UI:N). The only potential impact is a very minor confidentiality difference: a disabled, TOTP-enrolled user receives a TOTP prompt instead of an immediate rejection, which could marginally aid account-status enumeration (VC:L). There is no integrity or availability impact (VI:N, VA:N) and no secondary impacts (SC:N, SI:N, SA:N) because the subsequent identify() call still enforces the disabled-user check. The commit message explicitly characterizes the issue as \u0027harmless in practice.\u0027",
                "fixSummary": "The fix adds User.disabled to the fields array of the pre-authentication find() query so that the disabled-user check in the TOTP branch reads the actual column value and correctly rejects disabled users before they reach the TOTP verification step, restoring the intended guard behavior.",
                "generatedAt": "2026-09-22T14:51:29.308929Z",
                "generator": "patch2vuln.py",
                "model": "qwen3.8:27b",
                "modelComparison": {
                  "rankings": [
                    {
                      "agreementScore": 9,
                      "assumptionCount": 5,
                      "confidence": "medium",
                      "model": "qwen3.8:27b",
                      "score": 5
                    }
                  ],
                  "selectedModel": "qwen3.8:27b",
                  "selectionMethod": "deterministic-consensus-v1",
                  "selectionNotice": "The selected result is closest to model consensus; this heuristic does not establish factual correctness and human review remains required."
                },
                "patchSha256": "8ddba4461463bf2e5b115b4bdd168f91b246f1fcd796dc4da8873ed8da2397c0",
                "patchSummary": "A single-line change in app/Controller/UsersController.php appends \u0027User.disabled\u0027 to the fields list of the pre-authentication User find() query in the login() method (line 1385), ensuring the disabled flag is present in the result set when the TOTP branch evaluates it.",
                "patchTruncated": false,
                "patches": [
                  {
                    "commit": "3df982ab192e05f5369d62af5674c7a874b32756",
                    "patchSha256": "8ddba4461463bf2e5b115b4bdd168f91b246f1fcd796dc4da8873ed8da2397c0",
                    "source": "https://github.com/MISP/MISP/commit/3df982ab1.patch",
                    "sourceUrl": "https://github.com/MISP/MISP/commit/3df982ab1.patch",
                    "subject": "fix: [login] Fetch User.disabled in the pre-auth TOTP lookup"
                  }
                ],
                "source": "https://github.com/MISP/MISP/commit/3df982ab1.patch",
                "subject": "fix: [login] Fetch User.disabled in the pre-auth TOTP lookup",
                "tagVersionBoundary": {
                  "commits_after_fix": 271,
                  "repository": "https://github.com/MISP/MISP",
                  "tag": "v2.5.47",
                  "version": "2.5.47",
                  "version_type": "semver"
                },
                "weaknessRationale": [
                  {
                    "cweId": "CWE-285",
                    "rationale": "The disabled-user check in the TOTP login branch was ineffective because the required column was not fetched, allowing a disabled user to pass that guard. The subsequent identify() call still enforced the check, limiting the impact. CWE-285 is the narrowest defensible mapping for an authentication-state check that does not function as intended."
                  }
                ]
              }
            }
          },
          "recordType": "advisory",
          "vulnId": "GCVE-1-2026-20024"
        }
      ]
    }
  },
  "cveMetadata": {
    "assignerOrgId": "00000000-0000-4000-9000-000000000000",
    "cveId": "CVE-2026-95754",
    "datePublished": "2026-09-22T14:53:44.090577Z",
    "dateReserved": "2026-09-22T14:53:53.341Z",
    "dateUpdated": "2026-09-22T14:53:53.418288Z",
    "state": "PUBLISHED",
    "vulnId": "GCVE-1-2026-20024"
  },
  "dataType": "CVE_RECORD",
  "dataVersion": "5.1"
}

GCVE-1-2026-20270 (CVE-2026-95703)

Vulnerability from gna-1 – Published: 2026-09-22 14:49 – Updated: 2026-09-22 14:49
VLAI
Title
MISP OrganisationsController File Existence and Image-Type Oracle via Forged Upload tmp_name
Summary
In MISP, the OrganisationsController::__uploadLogo method processed a caller-supplied tmp_name value with filesystem probes (file_exists, MIME type detection, EXIF reading) before verifying that the value corresponded to a genuine PHP file upload via is_uploaded_file. An authenticated site-admin user could supply an arbitrary server file path as the tmp_name parameter. The application would then probe that path and return distinct validation error messages depending on whether the file existed and what its image type was, effectively creating a file-existence and image-type oracle against the server filesystem. The vulnerability requires site-admin privileges and does not allow arbitrary file read, code execution, or modification; the impact is limited to disclosure of whether a given path exists on the server and, for image files, their type.
CWE
  • CWE-200 - Exposure of Sensitive Information to an Unauthorized Actor
  • CWE-20 - Improper Input Validation
Assigner
CIRCL GNA GNA-1
GNA scorecard A 91/100 over 279 records in the last 180 days details
References
Impacted products
Vendor Product Version
MISP MISP Affected: 0 , < 2.5.47 (semver)
Create a notification for this product.
GCVE extensions
bcp-05-x-01
AI-assisted vulnerability information annotation
GCVE-BCP-05-X-01
Whole record AI-generated Human-reviewed GNA-1

Draft vulnerability metadata was generated from a git-format patch using an Ollama-hosted language model. Human validation is required before publication.

ai-computer-assisted:llm-generatedai-computer-assisted:classification
Model Source Identifier
qwen3.8:27b ollama qwen3.8:27b
bcp-05-x-02
Patch-to-vulnerability generation provenance
GCVE-BCP-05-X-02
Generator
patch2vuln.py on 2026-09-22 14:45
Model
qwen3.8:27b
Input
https://github.com/MISP/MISP/commit/12eaadc9e.patch 175abb34a19c…
Confidence
medium
Commit Subject Patch SHA-256
12eaadc9e28b fix: [organisation] Reject a forged upload path in the 175abb34a19c…
Fix summary

The is_uploaded_file() guard is hoisted to execute immediately after the basic size/error check and before any filesystem probe (file_exists, MIME detection, EXIF reading). If the tmp_name is empty or does not correspond to a genuine PHP upload, the method returns false early, preventing any filesystem interaction with an attacker-controlled path and eliminating the information oracle.

Patch summary

In app/Controller/OrganisationsController.php, six lines are inserted inside __uploadLogo() after the existing size/error check. The added block checks whether $logo['tmp_name'] is empty or fails is_uploaded_file(), and if so, returns false immediately. A comment explains that only a genuine PHP upload may reach the subsequent filesystem probes. No other logic is modified.

CVSS rationale

AV:N: exploited over the network via the MISP web interface. AC:L: the attack is a simple parameter substitution with no race or timing requirement. AT:N: no attack-target manipulation is needed. PR:H: the commit message explicitly states this is a site-admin-only issue. UI:N: no victim interaction beyond the admin's own request. VC:L: limited confidentiality impact — disclosure of file existence and image type on the server, but no file content is read. VI:N, VA:N: no integrity or availability impact. SC:N, SI:N, SA:N: no secondary-component impact.

Weakness rationale
  • CWE-200 The primary security impact is the disclosure of file existence and image type on the server through differential error messages, which constitutes sensitive information exposure to an authenticated (but not fully privileged) actor.
  • CWE-20 The root cause is that the caller-supplied tmp_name value was used in filesystem operations without first validating that it originated from a genuine PHP file upload (is_uploaded_file check was performed too late or not at all before the probes).
Attack pattern rationale
  • CAPEC-126 The attacker manipulates the tmp_name parameter of the upload form to point to an arbitrary server file path, causing the application to probe that path and leak information through its responses. This is the closest CAPEC pattern to the observed attack: tampering with a request parameter to trigger unintended server-side behavior. Uncertainty: no CAPEC pattern specifically covers file-existence oracles via forged upload paths, so CAPEC-126 is the best available match.
Assumptions to verify
  • The commit message states the issue is 'site-admin-only'; this is taken as the required privilege level (PR:H) without independent verification of MISP's role model.
  • The tag_version_boundary indicates v2.5.47 with 48 commits after the fix, suggesting the fix landed shortly after v2.5.47, but no explicit fixed version tag is provided; the affected range is therefore marked as less_than 2.5.47 with low confidence.
  • The CAPEC-126 mapping is the closest available pattern; no CAPEC specifically describes a file-existence oracle via forged upload parameters, so the mapping is approximate.
  • The commit message references a similar pattern in the event-report picture upload; whether that path was also fixed in this or a separate commit is not determined by this patch alone.
  • The 'found during the internal review' statement in the commit message is treated as a generic internal process note; no specific finder is credited because the metadata explicitly lists finders as empty.
Model comparison

Selected qwen3.8:27b by deterministic-consensus-v1
The selected result is closest to model consensus; this heuristic does not establish factual correctness and human review remains required.

Model Score Agreement Confidence Assumptions
qwen3.8:27b 5 9 medium 5

{
  "containers": {
    "cna": {
      "affected": [
        {
          "modules": [
            "OrganisationsController"
          ],
          "product": "MISP",
          "programFiles": [
            "app/Controller/OrganisationsController.php"
          ],
          "repo": "https://github.com/MISP/MISP",
          "vendor": "MISP",
          "versions": [
            {
              "lessThan": "2.5.47",
              "status": "affected",
              "version": "0",
              "versionType": "semver"
            }
          ]
        }
      ],
      "credits": [
        {
          "lang": "en",
          "type": "remediation developer",
          "value": "iglocska"
        },
        {
          "lang": "en",
          "type": "remediation developer",
          "value": "Claude Opus 4.8"
        }
      ],
      "descriptions": [
        {
          "lang": "en",
          "supportingMedia": [
            {
              "base64": false,
              "type": "text/html",
              "value": "\u003cp\u003eIn MISP, the OrganisationsController::__uploadLogo method processed a caller-supplied tmp_name value with filesystem probes (file_exists, MIME type detection, EXIF reading) before verifying that the value corresponded to a genuine PHP file upload via is_uploaded_file. An authenticated site-admin user could supply an arbitrary server file path as the tmp_name parameter. The application would then probe that path and return distinct validation error messages depending on whether the file existed and what its image type was, effectively creating a file-existence and image-type oracle against the server filesystem.\u003c/p\u003e\u003cp\u003eThe vulnerability requires site-admin privileges and does not allow arbitrary file read, code execution, or modification; the impact is limited to disclosure of whether a given path exists on the server and, for image files, their type.\u0026nbsp;\u003c/p\u003e"
            }
          ],
          "value": "In MISP, the OrganisationsController::__uploadLogo method processed a caller-supplied tmp_name value with filesystem probes (file_exists, MIME type detection, EXIF reading) before verifying that the value corresponded to a genuine PHP file upload via is_uploaded_file. An authenticated site-admin user could supply an arbitrary server file path as the tmp_name parameter. The application would then probe that path and return distinct validation error messages depending on whether the file existed and what its image type was, effectively creating a file-existence and image-type oracle against the server filesystem.\n\nThe vulnerability requires site-admin privileges and does not allow arbitrary file read, code execution, or modification; the impact is limited to disclosure of whether a given path exists on the server and, for image files, their type."
        }
      ],
      "impacts": [
        {
          "capecId": "CAPEC-126",
          "descriptions": [
            {
              "lang": "en",
              "value": "CAPEC-126 Parameter Tampering"
            }
          ]
        }
      ],
      "metrics": [
        {
          "cvssV4_0": {
            "Automatable": "NOT_DEFINED",
            "Recovery": "NOT_DEFINED",
            "Safety": "NOT_DEFINED",
            "attackComplexity": "LOW",
            "attackRequirements": "NONE",
            "attackVector": "NETWORK",
            "baseScore": 5.1,
            "baseSeverity": "MEDIUM",
            "privilegesRequired": "HIGH",
            "providerUrgency": "NOT_DEFINED",
            "subAvailabilityImpact": "NONE",
            "subConfidentialityImpact": "NONE",
            "subIntegrityImpact": "NONE",
            "userInteraction": "NONE",
            "valueDensity": "NOT_DEFINED",
            "vectorString": "CVSS:4.0/AV:N/AC:L/AT:N/PR:H/UI:N/VC:L/VI:N/VA:N/SC:N/SI:N/SA:N",
            "version": "4.0",
            "vulnAvailabilityImpact": "NONE",
            "vulnConfidentialityImpact": "LOW",
            "vulnIntegrityImpact": "NONE",
            "vulnerabilityResponseEffort": "NOT_DEFINED"
          },
          "format": "CVSS",
          "scenarios": [
            {
              "lang": "en",
              "value": "GENERAL"
            }
          ]
        }
      ],
      "problemTypes": [
        {
          "descriptions": [
            {
              "cweId": "CWE-200",
              "description": "CWE-200 Exposure of Sensitive Information to an Unauthorized Actor",
              "lang": "en",
              "type": "CWE"
            }
          ]
        },
        {
          "descriptions": [
            {
              "cweId": "CWE-20",
              "description": "CWE-20 Improper Input Validation",
              "lang": "en",
              "type": "CWE"
            }
          ]
        }
      ],
      "providerMetadata": {
        "orgId": "00000000-0000-4000-9000-000000000000"
      },
      "references": [
        {
          "name": "Security patch",
          "tags": [
            "patch"
          ],
          "url": "https://github.com/MISP/MISP/commit/12eaadc9e"
        }
      ],
      "solutions": [
        {
          "lang": "en",
          "supportingMedia": [
            {
              "base64": false,
              "type": "text/html",
              "value": "\u003cp\u003eThe is_uploaded_file() guard is hoisted to execute immediately after the basic size/error check and before any filesystem probe (file_exists, MIME detection, EXIF reading). If the tmp_name is empty or does not correspond to a genuine PHP upload, the method returns false early, preventing any filesystem interaction with an attacker-controlled path and eliminating the information oracle.\u003c/p\u003e"
            }
          ],
          "value": "The is_uploaded_file() guard is hoisted to execute immediately after the basic size/error check and before any filesystem probe (file_exists, MIME detection, EXIF reading). If the tmp_name is empty or does not correspond to a genuine PHP upload, the method returns false early, preventing any filesystem interaction with an attacker-controlled path and eliminating the information oracle."
        }
      ],
      "title": "MISP OrganisationsController File Existence and Image-Type Oracle via Forged Upload tmp_name",
      "x_gcve": [
        {
          "extensions": {
            "bcp-05-x-01": {
              "ai_annotations": [
                {
                  "ai_level": "generated",
                  "description": "Draft vulnerability metadata was generated from a git-format patch using an Ollama-hosted language model. Human validation is required before publication.",
                  "gna_source": 1,
                  "models": [
                    {
                      "gna_source": 1,
                      "identifier": "qwen3.8:27b",
                      "name": "qwen3.8:27b",
                      "source": "ollama"
                    }
                  ],
                  "review_status": "full",
                  "scope": "record",
                  "tags": [
                    "ai-computer-assisted:llm-generated",
                    "ai-computer-assisted:classification"
                  ]
                }
              ]
            },
            "bcp-05-x-02": {
              "x_patch2vuln": {
                "assumptions": [
                  "The commit message states the issue is \u0027site-admin-only\u0027; this is taken as the required privilege level (PR:H) without independent verification of MISP\u0027s role model.",
                  "The tag_version_boundary indicates v2.5.47 with 48 commits after the fix, suggesting the fix landed shortly after v2.5.47, but no explicit fixed version tag is provided; the affected range is therefore marked as less_than 2.5.47 with low confidence.",
                  "The CAPEC-126 mapping is the closest available pattern; no CAPEC specifically describes a file-existence oracle via forged upload parameters, so the mapping is approximate.",
                  "The commit message references a similar pattern in the event-report picture upload; whether that path was also fixed in this or a separate commit is not determined by this patch alone.",
                  "The \u0027found during the internal review\u0027 statement in the commit message is treated as a generic internal process note; no specific finder is credited because the metadata explicitly lists finders as empty."
                ],
                "capecRationale": [
                  {
                    "capecId": "CAPEC-126",
                    "rationale": "The attacker manipulates the tmp_name parameter of the upload form to point to an arbitrary server file path, causing the application to probe that path and leak information through its responses. This is the closest CAPEC pattern to the observed attack: tampering with a request parameter to trigger unintended server-side behavior. Uncertainty: no CAPEC pattern specifically covers file-existence oracles via forged upload paths, so CAPEC-126 is the best available match."
                  }
                ],
                "commit": "12eaadc9e28bdb7fc723faa2f006eaaa13c4ffdb",
                "confidence": "medium",
                "credits": [
                  {
                    "lang": "en",
                    "type": "remediation developer",
                    "value": "iglocska"
                  },
                  {
                    "lang": "en",
                    "type": "remediation developer",
                    "value": "Claude Opus 4.8"
                  }
                ],
                "cvssRationale": "AV:N: exploited over the network via the MISP web interface. AC:L: the attack is a simple parameter substitution with no race or timing requirement. AT:N: no attack-target manipulation is needed. PR:H: the commit message explicitly states this is a site-admin-only issue. UI:N: no victim interaction beyond the admin\u0027s own request. VC:L: limited confidentiality impact \u2014 disclosure of file existence and image type on the server, but no file content is read. VI:N, VA:N: no integrity or availability impact. SC:N, SI:N, SA:N: no secondary-component impact.",
                "fixSummary": "The is_uploaded_file() guard is hoisted to execute immediately after the basic size/error check and before any filesystem probe (file_exists, MIME detection, EXIF reading). If the tmp_name is empty or does not correspond to a genuine PHP upload, the method returns false early, preventing any filesystem interaction with an attacker-controlled path and eliminating the information oracle.",
                "generatedAt": "2026-09-22T14:45:44.315009Z",
                "generator": "patch2vuln.py",
                "model": "qwen3.8:27b",
                "modelComparison": {
                  "rankings": [
                    {
                      "agreementScore": 9,
                      "assumptionCount": 5,
                      "confidence": "medium",
                      "model": "qwen3.8:27b",
                      "score": 5
                    }
                  ],
                  "selectedModel": "qwen3.8:27b",
                  "selectionMethod": "deterministic-consensus-v1",
                  "selectionNotice": "The selected result is closest to model consensus; this heuristic does not establish factual correctness and human review remains required."
                },
                "patchSha256": "175abb34a19cd998d66ca5363124749b640d7b922547e5d3aff8bd4e4239dfa7",
                "patchSummary": "In app/Controller/OrganisationsController.php, six lines are inserted inside __uploadLogo() after the existing size/error check. The added block checks whether $logo[\u0027tmp_name\u0027] is empty or fails is_uploaded_file(), and if so, returns false immediately. A comment explains that only a genuine PHP upload may reach the subsequent filesystem probes. No other logic is modified.",
                "patchTruncated": false,
                "patches": [
                  {
                    "commit": "12eaadc9e28bdb7fc723faa2f006eaaa13c4ffdb",
                    "patchSha256": "175abb34a19cd998d66ca5363124749b640d7b922547e5d3aff8bd4e4239dfa7",
                    "source": "https://github.com/MISP/MISP/commit/12eaadc9e.patch",
                    "sourceUrl": "https://github.com/MISP/MISP/commit/12eaadc9e.patch",
                    "subject": "fix: [organisation] Reject a forged upload path in the"
                  }
                ],
                "source": "https://github.com/MISP/MISP/commit/12eaadc9e.patch",
                "subject": "fix: [organisation] Reject a forged upload path in the",
                "tagVersionBoundary": {
                  "commits_after_fix": 48,
                  "repository": "https://github.com/MISP/MISP",
                  "tag": "v2.5.47",
                  "version": "2.5.47",
                  "version_type": "semver"
                },
                "weaknessRationale": [
                  {
                    "cweId": "CWE-200",
                    "rationale": "The primary security impact is the disclosure of file existence and image type on the server through differential error messages, which constitutes sensitive information exposure to an authenticated (but not fully privileged) actor."
                  },
                  {
                    "cweId": "CWE-20",
                    "rationale": "The root cause is that the caller-supplied tmp_name value was used in filesystem operations without first validating that it originated from a genuine PHP file upload (is_uploaded_file check was performed too late or not at all before the probes)."
                  }
                ]
              }
            }
          },
          "recordType": "advisory",
          "vulnId": "GCVE-1-2026-20270"
        }
      ]
    }
  },
  "cveMetadata": {
    "assignerOrgId": "00000000-0000-4000-9000-000000000000",
    "cveId": "CVE-2026-95703",
    "datePublished": "2026-09-22T14:49:29.214426Z",
    "dateReserved": "2026-09-22T14:49:37.648Z",
    "dateUpdated": "2026-09-22T14:49:37.760620Z",
    "state": "PUBLISHED",
    "vulnId": "GCVE-1-2026-20270"
  },
  "dataType": "CVE_RECORD",
  "dataVersion": "5.1"
}

GCVE-1-2026-20065 (CVE-2026-95701)

Vulnerability from gna-1 – Published: 2026-09-22 14:44 – Updated: 2026-09-22 14:44
VLAI
Title
MISP Path Traversal via Organization Name in Org-Statistics Logo Check
Summary
In MISP, the __statisticsOrgs method in UsersController.php used the organization name directly as a file-system path component when checking for the existence of an organization logo image. The original code called file_exists() with a path constructed as APP . 'webroot' . DS . 'img' . DS . 'orgs' . DS . $k . '.png', where $k is the organization name. Because the referenced directory (app/webroot/img/orgs) no longer exists in current MISP deployments (org logos were relocated to files/img/orgs), the check was functionally dead and never triggered. However, the underlying pattern—concatenating an attacker-influenced organization name into a file path without sanitization—constitutes a path traversal weakness. An organization name containing directory traversal sequences (e.g., '../../../../etc/passwd') would, if the target directory existed, allow an authenticated user with the ability to create or rename an organization to probe for the existence of arbitrary files on the server.
CWE
  • CWE-22 - Improper Limitation of a Pathname to a Restricted Directory ('Path Traversal')
Assigner
CIRCL GNA GNA-1
GNA scorecard A 91/100 over 279 records in the last 180 days details
References
Impacted products
Vendor Product Version
MISP MISP Affected: 0 , < 2.5.47 (semver)
Create a notification for this product.
GCVE extensions
bcp-05-x-01
AI-assisted vulnerability information annotation
GCVE-BCP-05-X-01
Whole record AI-generated Human-reviewed GNA-1

Draft vulnerability metadata was generated from a git-format patch using an Ollama-hosted language model. Human validation is required before publication.

ai-computer-assisted:llm-generatedai-computer-assisted:classification
Model Source Identifier
qwen3.8:27b ollama qwen3.8:27b
bcp-05-x-02
Patch-to-vulnerability generation provenance
GCVE-BCP-05-X-02
Generator
patch2vuln.py on 2026-09-22 14:38
Model
qwen3.8:27b
Input
https://github.com/MISP/MISP/commit/a2f7cba6e.patch 79f69959d0bc…
Confidence
medium
Commit Subject Patch SHA-256
a2f7cba6e74b fix: [ui] Point the org-statistics logo flag at the live 79f69959d0bc…
Fix summary

The fix corrects the logo directory path from the obsolete app/webroot/img/orgs to the current files/img/orgs location, expands the lookup to cover id, name, and uuid fields across png and svg extensions (mirroring the getOrgLogo helper), and adds a security guard: the candidate path is resolved with realpath() and verified via str_starts_with() against the resolved base directory, so any organization name containing traversal sequences (../) that would resolve outside files/img/orgs is rejected.

Patch summary

In app/Controller/UsersController.php __statisticsOrgs(): (1) added 'uuid' to the Organisation query fields; (2) replaced the single file_exists() call against APP/webroot/img/orgs/$k.png with a nested loop over fields [id, name, uuid] and extensions [png, svg]; (3) for each candidate, called realpath() on the full path and compared it with str_starts_with() against the realpath of the base directory (APP/files/img/orgs/), setting the logo flag only when the resolved path is confirmed to be inside that directory; (4) added a break 2 to exit both loops once a match is found.

CVSS rationale

AV:N – MISP is a network-accessible web application. AC:L – the traversal sequence in an org name is straightforward to construct. AT:N – no special target-side conditions beyond the org name being stored. PR:H – the attacker must have sufficient privileges to create or rename an organization (typically admin or org-admin role). UI:N – no user interaction required; the check fires server-side during statistics rendering. VC:L – the impact is limited to file-existence disclosure (boolean oracle); file contents are not read. VI:N, VA:N – no integrity or availability impact. SC/SI/SA:N – no secondary-system impact. Note: the vulnerability was latent in practice because the referenced directory did not exist, but the code pattern was exploitable if the directory were present or restored.

Weakness rationale
  • CWE-22 The organization name (an attacker-influenced string) was concatenated directly into a file-system path for a file_exists() check without any sanitization or directory-confinement validation. The fix explicitly guards against '../' sequences escaping the intended directory, confirming the weakness is path traversal.
Attack pattern rationale
  • CAPEC-1 The attack pattern involves manipulating a string value (the organization name) to inject path traversal sequences (../) that cause the application to reference files outside the intended directory. CAPEC-1 is the closest match because the core technique is crafting a string input to alter the application's file-path resolution. Uncertainty: CAPEC-126 (Leveraging Unintended Functionality) could also apply since the logo-existence check was an unintended side channel, but CAPEC-1 more directly describes the string-manipulation mechanism.
Assumptions to verify
  • The affected version range is inferred from the tag_version_boundary (v2.5.47, 50 commits after fix); the exact fixed release tag is not stated in the patch metadata, so 'less_than 2.5.47' is an approximation and may need CNA confirmation.
  • The vulnerability was latent in deployed instances because the referenced directory (app/webroot/img/orgs) did not exist; the CVSS reflects the code-level weakness rather than a confirmed active exploit.
  • PR:H assumes that creating or renaming an organization requires elevated privileges (admin or org-admin); if MISP allows lower-privileged users to set org names, PR could be lower.
  • CAPEC-1 (String Manipulation) is the closest available pattern; the exact CAPEC for path traversal via a stored string field is not explicitly enumerated in the CAPEC catalog, so this is a best-effort mapping.
  • The Co-Authored-By line references an AI assistant (Claude Opus 4.8); it is credited as a tool rather than a human remediation developer.
Model comparison

Selected qwen3.8:27b by deterministic-consensus-v1
The selected result is closest to model consensus; this heuristic does not establish factual correctness and human review remains required.

Model Score Agreement Confidence Assumptions
qwen3.8:27b 5 9 medium 5

{
  "containers": {
    "cna": {
      "affected": [
        {
          "modules": [
            "UsersController::__statisticsOrgs"
          ],
          "product": "MISP",
          "programFiles": [
            "app/Controller/UsersController.php"
          ],
          "repo": "https://github.com/MISP/MISP",
          "vendor": "MISP",
          "versions": [
            {
              "lessThan": "2.5.47",
              "status": "affected",
              "version": "0",
              "versionType": "semver"
            }
          ]
        }
      ],
      "credits": [
        {
          "lang": "en",
          "type": "reporter",
          "value": "Jeroen Pinoy"
        },
        {
          "lang": "en",
          "type": "remediation developer",
          "value": "iglocska"
        },
        {
          "lang": "en",
          "type": "remediation developer",
          "value": "Claude Opus 4.8"
        }
      ],
      "descriptions": [
        {
          "lang": "en",
          "supportingMedia": [
            {
              "base64": false,
              "type": "text/html",
              "value": "\u003cp\u003eIn MISP, the __statisticsOrgs method in UsersController.php used the organization name directly as a file-system path component when checking for the existence of an organization logo image. The original code called file_exists() with a path constructed as APP . \u0027webroot\u0027 . DS . \u0027img\u0027 . DS . \u0027orgs\u0027 . DS . $k . \u0027.png\u0027, where $k is the organization name. Because the referenced directory (app/webroot/img/orgs) no longer exists in current MISP deployments (org logos were relocated to files/img/orgs), the check was functionally dead and never triggered. However, the underlying pattern\u2014concatenating an attacker-influenced organization name into a file path without sanitization\u2014constitutes a path traversal weakness. An organization name containing directory traversal sequences (e.g., \u0027../../../../etc/passwd\u0027) would, if the target directory existed, allow an authenticated user with the ability to create or rename an organization to probe for the existence of arbitrary files on the server.\u003c/p\u003e"
            }
          ],
          "value": "In MISP, the __statisticsOrgs method in UsersController.php used the organization name directly as a file-system path component when checking for the existence of an organization logo image. The original code called file_exists() with a path constructed as APP . \u0027webroot\u0027 . DS . \u0027img\u0027 . DS . \u0027orgs\u0027 . DS . $k . \u0027.png\u0027, where $k is the organization name. Because the referenced directory (app/webroot/img/orgs) no longer exists in current MISP deployments (org logos were relocated to files/img/orgs), the check was functionally dead and never triggered. However, the underlying pattern\u2014concatenating an attacker-influenced organization name into a file path without sanitization\u2014constitutes a path traversal weakness. An organization name containing directory traversal sequences (e.g., \u0027../../../../etc/passwd\u0027) would, if the target directory existed, allow an authenticated user with the ability to create or rename an organization to probe for the existence of arbitrary files on the server."
        }
      ],
      "impacts": [
        {
          "capecId": "CAPEC-1",
          "descriptions": [
            {
              "lang": "en",
              "value": "CAPEC-1 String Manipulation"
            }
          ]
        }
      ],
      "metrics": [
        {
          "cvssV4_0": {
            "Automatable": "NOT_DEFINED",
            "Recovery": "NOT_DEFINED",
            "Safety": "NOT_DEFINED",
            "attackComplexity": "LOW",
            "attackRequirements": "NONE",
            "attackVector": "NETWORK",
            "baseScore": 5.1,
            "baseSeverity": "MEDIUM",
            "privilegesRequired": "HIGH",
            "providerUrgency": "NOT_DEFINED",
            "subAvailabilityImpact": "NONE",
            "subConfidentialityImpact": "NONE",
            "subIntegrityImpact": "NONE",
            "userInteraction": "NONE",
            "valueDensity": "NOT_DEFINED",
            "vectorString": "CVSS:4.0/AV:N/AC:L/AT:N/PR:H/UI:N/VC:L/VI:N/VA:N/SC:N/SI:N/SA:N",
            "version": "4.0",
            "vulnAvailabilityImpact": "NONE",
            "vulnConfidentialityImpact": "LOW",
            "vulnIntegrityImpact": "NONE",
            "vulnerabilityResponseEffort": "NOT_DEFINED"
          },
          "format": "CVSS",
          "scenarios": [
            {
              "lang": "en",
              "value": "GENERAL"
            }
          ]
        }
      ],
      "problemTypes": [
        {
          "descriptions": [
            {
              "cweId": "CWE-22",
              "description": "CWE-22 Improper Limitation of a Pathname to a Restricted Directory (\u0027Path Traversal\u0027)",
              "lang": "en",
              "type": "CWE"
            }
          ]
        }
      ],
      "providerMetadata": {
        "orgId": "00000000-0000-4000-9000-000000000000"
      },
      "references": [
        {
          "name": "Security patch",
          "tags": [
            "patch"
          ],
          "url": "https://github.com/MISP/MISP/commit/a2f7cba6e"
        }
      ],
      "solutions": [
        {
          "lang": "en",
          "supportingMedia": [
            {
              "base64": false,
              "type": "text/html",
              "value": "\u003cp\u003eThe fix corrects the logo directory path from the obsolete app/webroot/img/orgs to the current files/img/orgs location, expands the lookup to cover id, name, and uuid fields across png and svg extensions (mirroring the getOrgLogo helper), and adds a security guard: the candidate path is resolved with realpath() and verified via str_starts_with() against the resolved base directory, so any organization name containing traversal sequences (../) that would resolve outside files/img/orgs is rejected.\u003c/p\u003e"
            }
          ],
          "value": "The fix corrects the logo directory path from the obsolete app/webroot/img/orgs to the current files/img/orgs location, expands the lookup to cover id, name, and uuid fields across png and svg extensions (mirroring the getOrgLogo helper), and adds a security guard: the candidate path is resolved with realpath() and verified via str_starts_with() against the resolved base directory, so any organization name containing traversal sequences (../) that would resolve outside files/img/orgs is rejected."
        }
      ],
      "title": "MISP Path Traversal via Organization Name in Org-Statistics Logo Check",
      "x_gcve": [
        {
          "extensions": {
            "bcp-05-x-01": {
              "ai_annotations": [
                {
                  "ai_level": "generated",
                  "description": "Draft vulnerability metadata was generated from a git-format patch using an Ollama-hosted language model. Human validation is required before publication.",
                  "gna_source": 1,
                  "models": [
                    {
                      "gna_source": 1,
                      "identifier": "qwen3.8:27b",
                      "name": "qwen3.8:27b",
                      "source": "ollama"
                    }
                  ],
                  "review_status": "full",
                  "scope": "record",
                  "tags": [
                    "ai-computer-assisted:llm-generated",
                    "ai-computer-assisted:classification"
                  ]
                }
              ]
            },
            "bcp-05-x-02": {
              "x_patch2vuln": {
                "assumptions": [
                  "The affected version range is inferred from the tag_version_boundary (v2.5.47, 50 commits after fix); the exact fixed release tag is not stated in the patch metadata, so \u0027less_than 2.5.47\u0027 is an approximation and may need CNA confirmation.",
                  "The vulnerability was latent in deployed instances because the referenced directory (app/webroot/img/orgs) did not exist; the CVSS reflects the code-level weakness rather than a confirmed active exploit.",
                  "PR:H assumes that creating or renaming an organization requires elevated privileges (admin or org-admin); if MISP allows lower-privileged users to set org names, PR could be lower.",
                  "CAPEC-1 (String Manipulation) is the closest available pattern; the exact CAPEC for path traversal via a stored string field is not explicitly enumerated in the CAPEC catalog, so this is a best-effort mapping.",
                  "The Co-Authored-By line references an AI assistant (Claude Opus 4.8); it is credited as a tool rather than a human remediation developer."
                ],
                "capecRationale": [
                  {
                    "capecId": "CAPEC-1",
                    "rationale": "The attack pattern involves manipulating a string value (the organization name) to inject path traversal sequences (../) that cause the application to reference files outside the intended directory. CAPEC-1 is the closest match because the core technique is crafting a string input to alter the application\u0027s file-path resolution. Uncertainty: CAPEC-126 (Leveraging Unintended Functionality) could also apply since the logo-existence check was an unintended side channel, but CAPEC-1 more directly describes the string-manipulation mechanism."
                  }
                ],
                "commit": "a2f7cba6e74b791e30013f10720dadbc4117d989",
                "confidence": "medium",
                "credits": [
                  {
                    "lang": "en",
                    "type": "reporter",
                    "value": "Jeroen Pinoy"
                  },
                  {
                    "lang": "en",
                    "type": "remediation developer",
                    "value": "iglocska"
                  },
                  {
                    "lang": "en",
                    "type": "remediation developer",
                    "value": "Claude Opus 4.8"
                  }
                ],
                "cvssRationale": "AV:N \u2013 MISP is a network-accessible web application. AC:L \u2013 the traversal sequence in an org name is straightforward to construct. AT:N \u2013 no special target-side conditions beyond the org name being stored. PR:H \u2013 the attacker must have sufficient privileges to create or rename an organization (typically admin or org-admin role). UI:N \u2013 no user interaction required; the check fires server-side during statistics rendering. VC:L \u2013 the impact is limited to file-existence disclosure (boolean oracle); file contents are not read. VI:N, VA:N \u2013 no integrity or availability impact. SC/SI/SA:N \u2013 no secondary-system impact. Note: the vulnerability was latent in practice because the referenced directory did not exist, but the code pattern was exploitable if the directory were present or restored.",
                "fixSummary": "The fix corrects the logo directory path from the obsolete app/webroot/img/orgs to the current files/img/orgs location, expands the lookup to cover id, name, and uuid fields across png and svg extensions (mirroring the getOrgLogo helper), and adds a security guard: the candidate path is resolved with realpath() and verified via str_starts_with() against the resolved base directory, so any organization name containing traversal sequences (../) that would resolve outside files/img/orgs is rejected.",
                "generatedAt": "2026-09-22T14:38:57.879793Z",
                "generator": "patch2vuln.py",
                "model": "qwen3.8:27b",
                "modelComparison": {
                  "rankings": [
                    {
                      "agreementScore": 9,
                      "assumptionCount": 5,
                      "confidence": "medium",
                      "model": "qwen3.8:27b",
                      "score": 5
                    }
                  ],
                  "selectedModel": "qwen3.8:27b",
                  "selectionMethod": "deterministic-consensus-v1",
                  "selectionNotice": "The selected result is closest to model consensus; this heuristic does not establish factual correctness and human review remains required."
                },
                "patchSha256": "79f69959d0bcac7da9eea5e1dcd7edc108868ea8382a8e477d3c72e749f06a37",
                "patchSummary": "In app/Controller/UsersController.php __statisticsOrgs(): (1) added \u0027uuid\u0027 to the Organisation query fields; (2) replaced the single file_exists() call against APP/webroot/img/orgs/$k.png with a nested loop over fields [id, name, uuid] and extensions [png, svg]; (3) for each candidate, called realpath() on the full path and compared it with str_starts_with() against the realpath of the base directory (APP/files/img/orgs/), setting the logo flag only when the resolved path is confirmed to be inside that directory; (4) added a break 2 to exit both loops once a match is found.",
                "patchTruncated": false,
                "patches": [
                  {
                    "commit": "a2f7cba6e74b791e30013f10720dadbc4117d989",
                    "patchSha256": "79f69959d0bcac7da9eea5e1dcd7edc108868ea8382a8e477d3c72e749f06a37",
                    "source": "https://github.com/MISP/MISP/commit/a2f7cba6e.patch",
                    "sourceUrl": "https://github.com/MISP/MISP/commit/a2f7cba6e.patch",
                    "subject": "fix: [ui] Point the org-statistics logo flag at the live"
                  }
                ],
                "source": "https://github.com/MISP/MISP/commit/a2f7cba6e.patch",
                "subject": "fix: [ui] Point the org-statistics logo flag at the live",
                "tagVersionBoundary": {
                  "commits_after_fix": 50,
                  "repository": "https://github.com/MISP/MISP",
                  "tag": "v2.5.47",
                  "version": "2.5.47",
                  "version_type": "semver"
                },
                "weaknessRationale": [
                  {
                    "cweId": "CWE-22",
                    "rationale": "The organization name (an attacker-influenced string) was concatenated directly into a file-system path for a file_exists() check without any sanitization or directory-confinement validation. The fix explicitly guards against \u0027../\u0027 sequences escaping the intended directory, confirming the weakness is path traversal."
                  }
                ]
              }
            }
          },
          "recordType": "advisory",
          "vulnId": "GCVE-1-2026-20065"
        }
      ]
    }
  },
  "cveMetadata": {
    "assignerOrgId": "00000000-0000-4000-9000-000000000000",
    "cveId": "CVE-2026-95701",
    "datePublished": "2026-09-22T14:44:11.001111Z",
    "dateReserved": "2026-09-22T14:44:19.514Z",
    "dateUpdated": "2026-09-22T14:44:19.846758Z",
    "state": "PUBLISHED",
    "vulnId": "GCVE-1-2026-20065"
  },
  "dataType": "CVE_RECORD",
  "dataVersion": "5.1"
}

GCVE-1-2026-20266 (CVE-2026-95698)

Vulnerability from gna-1 – Published: 2026-09-22 14:36 – Updated: 2026-09-22 14:36
VLAI
Title
MISP Path Traversal in OrgImgHelper findOrgImage via Crafted Organization Name
Summary
The findOrgImage method in MISP's OrgImgHelper constructs a filesystem path by concatenating a user-supplied organization identifier with a fixed image directory and a file extension, then calls file_exists() on the resulting path. The organization name field is attacker-controllable through event import, which sets Org.name. Because no validation was performed on the field value before path construction, an organization name containing directory traversal sequences (e.g., ../../etc/passwd) would cause the path to resolve outside the intended org-image directory.  This yields two security impacts:   - an arbitrary file existence oracle, since file_exists() returns a boolean for any path the process can access.  - an arbitrary file read limited to files with .png or .svg extensions, because the resolved filename is returned and subsequently rendered or served to the requesting user. The vulnerability requires an authenticated user with the ability to create or import events that set the organization name.
CWE
  • CWE-22 - Improper Limitation of a Pathname to a Restricted Directory ('Path Traversal')
Assigner
CIRCL GNA GNA-1
GNA scorecard A 91/100 over 279 records in the last 180 days details
References
Impacted products
Vendor Product Version
MISP MISP Affected: 0 , < 2.5.47 (semver)
Create a notification for this product.
GCVE extensions
bcp-05-x-01
AI-assisted vulnerability information annotation
GCVE-BCP-05-X-01
Whole record AI-generated Human-reviewed GNA-1

Draft vulnerability metadata was generated from a git-format patch using an Ollama-hosted language model. Human validation is required before publication.

ai-computer-assisted:llm-generatedai-computer-assisted:classification
Model Source Identifier
qwen3.8:27b ollama qwen3.8:27b
bcp-05-x-02
Patch-to-vulnerability generation provenance
GCVE-BCP-05-X-02
Generator
patch2vuln.py on 2026-09-22 14:32
Model
qwen3.8:27b
Input
https://github.com/MISP/MISP/commit/e00986075.patch 646ef716ac7a…
Confidence
high
Commit Subject Patch SHA-256
e00986075acf fix: [security] Reject path traversal in the org-logo 646ef716ac7a…
Fix summary

The fix adds input validation in findOrgImage so that each organization identifier value (id, name, uuid) is checked to be a non-empty plain filename component. Values containing path separators (detected via basename comparison) or dot-dot sequences are rejected, preventing the constructed path from escaping the designated org-image directory. The validated value is then used in the file_exists call and the returned image reference.

Patch summary

In app/View/Helper/OrgImgHelper.php, the findOrgImage method now casts each field value to a string and rejects it if it is empty, if basename($value) does not equal $value (indicating embedded path separators), or if it contains the substring '..'. The previously unvalidated $options[$field] is replaced with the validated $value in both the file_exists() call and the assignment to $image. Ten lines added, two lines removed in a single file.

CVSS rationale

AV:N: MISP is a network-accessible web application; the org name is set via API/event import over the network. AC:L: The attacker simply includes ../ in an org name during a normal import operation; no race or complex condition is needed. AT:N: No manipulation of the target beyond normal input is required. PR:L: The attacker must be an authenticated MISP user with permission to create or import events that set the organization name. UI:N: No victim interaction is required; the traversal occurs server-side during view rendering. VC:L: The attacker gains a file-existence oracle and can read .png/.svg files outside the intended directory, a limited but real confidentiality breach. VI:N, VA:N: No integrity or availability impact is evidenced. SC:N, SI:N, SA:N: No secondary impacts are indicated by the patch or commit message.

Weakness rationale
  • CWE-22 The vulnerability is a textbook path traversal: a user-controlled string (org name) is concatenated into a filesystem path without sanitization, allowing the resolved path to escape the intended directory. The fix explicitly rejects path separators and dot-dot sequences, confirming the root cause is missing path confinement.
Attack pattern rationale
  • CAPEC-126 CAPEC-126 describes an attacker manipulating path components in a request to access files or directories outside the intended scope. This matches the vulnerability exactly: the org name field is manipulated to include ../ sequences, causing the server to probe and potentially read files outside the org-image directory. The mapping is direct and unambiguous.
Assumptions to verify
  • The affected version boundary is inferred from the nearest git tag v2.5.47 with 51 commits after the fix; the exact last-affected and first-fixed release numbers are not explicitly stated in the patch metadata.
  • PR:L assumes the attacker needs an authenticated MISP account with event-import or org-creation privileges; the patch does not specify the exact permission level required.
  • The confidentiality impact is rated Low because the file read and existence oracle are limited to files with .png or .svg extensions; a broader arbitrary file read is not possible through this code path.
  • CAPEC-126 is selected as the closest match; no uncertainty is noted because the attack pattern (path traversal via crafted filename) maps directly to this CAPEC entry.
  • The commit date (2026-09-16) is taken at face value from the patch metadata; no independent verification of the timeline was performed.
Model comparison

Selected qwen3.8:27b by deterministic-consensus-v1
The selected result is closest to model consensus; this heuristic does not establish factual correctness and human review remains required.

Model Score Agreement Confidence Assumptions
qwen3.8:27b 6 9 high 5

{
  "containers": {
    "cna": {
      "affected": [
        {
          "modules": [
            "app/View/Helper/OrgImgHelper.php"
          ],
          "product": "MISP",
          "programFiles": [
            "app/View/Helper/OrgImgHelper.php"
          ],
          "repo": "https://github.com/MISP/MISP",
          "vendor": "MISP",
          "versions": [
            {
              "lessThan": "2.5.47",
              "status": "affected",
              "version": "0",
              "versionType": "semver"
            }
          ]
        }
      ],
      "credits": [
        {
          "lang": "en",
          "type": "reporter",
          "value": "Jeroen Pinoy"
        },
        {
          "lang": "en",
          "type": "remediation developer",
          "value": "iglocska"
        },
        {
          "lang": "en",
          "type": "remediation developer",
          "value": "Claude Opus 4.8"
        }
      ],
      "descriptions": [
        {
          "lang": "en",
          "supportingMedia": [
            {
              "base64": false,
              "type": "text/html",
              "value": "\u003cp\u003eThe findOrgImage method in MISP\u0027s OrgImgHelper constructs a filesystem path by concatenating a user-supplied organization identifier with a fixed image directory and a file extension, then calls file_exists() on the resulting path. The organization name field is attacker-controllable through event import, which sets Org.name. Because no validation was performed on the field value before path construction, an organization name containing directory traversal sequences (e.g., ../../etc/passwd) would cause the path to resolve outside the intended org-image directory.\u0026nbsp;\u003c/p\u003e\u003cp\u003eThis yields two security impacts:\u0026nbsp;\u003c/p\u003e\u2003- an arbitrary file existence oracle, since file_exists() returns a boolean for any path the process can access.\u003cbr\u003e\u003cdiv\u003e\u2003-\u0026nbsp;an arbitrary file read limited to files with .png or .svg extensions, because the resolved filename is returned and subsequently rendered or served to the requesting user.\u003c/div\u003e\u003cdiv\u003e\u003cbr\u003e\u003c/div\u003e\u003cp\u003eThe vulnerability requires an authenticated user with the ability to create or import events that set the organization name.\u003c/p\u003e"
            }
          ],
          "value": "The findOrgImage method in MISP\u0027s OrgImgHelper constructs a filesystem path by concatenating a user-supplied organization identifier with a fixed image directory and a file extension, then calls file_exists() on the resulting path. The organization name field is attacker-controllable through event import, which sets Org.name. Because no validation was performed on the field value before path construction, an organization name containing directory traversal sequences (e.g., ../../etc/passwd) would cause the path to resolve outside the intended org-image directory.\u00a0\n\nThis yields two security impacts:\u00a0\n\n\u2003- an arbitrary file existence oracle, since file_exists() returns a boolean for any path the process can access.\n\u2003-\u00a0an arbitrary file read limited to files with .png or .svg extensions, because the resolved filename is returned and subsequently rendered or served to the requesting user.\n\n\n\n\nThe vulnerability requires an authenticated user with the ability to create or import events that set the organization name."
        }
      ],
      "impacts": [
        {
          "capecId": "CAPEC-126",
          "descriptions": [
            {
              "lang": "en",
              "value": "CAPEC-126 Path Traversal"
            }
          ]
        }
      ],
      "metrics": [
        {
          "cvssV4_0": {
            "Automatable": "NOT_DEFINED",
            "Recovery": "NOT_DEFINED",
            "Safety": "NOT_DEFINED",
            "attackComplexity": "LOW",
            "attackRequirements": "NONE",
            "attackVector": "NETWORK",
            "baseScore": 5.3,
            "baseSeverity": "MEDIUM",
            "privilegesRequired": "LOW",
            "providerUrgency": "NOT_DEFINED",
            "subAvailabilityImpact": "NONE",
            "subConfidentialityImpact": "NONE",
            "subIntegrityImpact": "NONE",
            "userInteraction": "NONE",
            "valueDensity": "NOT_DEFINED",
            "vectorString": "CVSS:4.0/AV:N/AC:L/AT:N/PR:L/UI:N/VC:L/VI:N/VA:N/SC:N/SI:N/SA:N",
            "version": "4.0",
            "vulnAvailabilityImpact": "NONE",
            "vulnConfidentialityImpact": "LOW",
            "vulnIntegrityImpact": "NONE",
            "vulnerabilityResponseEffort": "NOT_DEFINED"
          },
          "format": "CVSS",
          "scenarios": [
            {
              "lang": "en",
              "value": "GENERAL"
            }
          ]
        }
      ],
      "problemTypes": [
        {
          "descriptions": [
            {
              "cweId": "CWE-22",
              "description": "CWE-22 Improper Limitation of a Pathname to a Restricted Directory (\u0027Path Traversal\u0027)",
              "lang": "en",
              "type": "CWE"
            }
          ]
        }
      ],
      "providerMetadata": {
        "orgId": "00000000-0000-4000-9000-000000000000"
      },
      "references": [
        {
          "name": "Security patch",
          "tags": [
            "patch"
          ],
          "url": "https://github.com/MISP/MISP/commit/e00986075"
        }
      ],
      "solutions": [
        {
          "lang": "en",
          "supportingMedia": [
            {
              "base64": false,
              "type": "text/html",
              "value": "\u003cp\u003eThe fix adds input validation in findOrgImage so that each organization identifier value (id, name, uuid) is checked to be a non-empty plain filename component. Values containing path separators (detected via basename comparison) or dot-dot sequences are rejected, preventing the constructed path from escaping the designated org-image directory. The validated value is then used in the file_exists call and the returned image reference.\u003c/p\u003e"
            }
          ],
          "value": "The fix adds input validation in findOrgImage so that each organization identifier value (id, name, uuid) is checked to be a non-empty plain filename component. Values containing path separators (detected via basename comparison) or dot-dot sequences are rejected, preventing the constructed path from escaping the designated org-image directory. The validated value is then used in the file_exists call and the returned image reference."
        }
      ],
      "title": "MISP Path Traversal in OrgImgHelper findOrgImage via Crafted Organization Name",
      "x_gcve": [
        {
          "extensions": {
            "bcp-05-x-01": {
              "ai_annotations": [
                {
                  "ai_level": "generated",
                  "description": "Draft vulnerability metadata was generated from a git-format patch using an Ollama-hosted language model. Human validation is required before publication.",
                  "gna_source": 1,
                  "models": [
                    {
                      "gna_source": 1,
                      "identifier": "qwen3.8:27b",
                      "name": "qwen3.8:27b",
                      "source": "ollama"
                    }
                  ],
                  "review_status": "full",
                  "scope": "record",
                  "tags": [
                    "ai-computer-assisted:llm-generated",
                    "ai-computer-assisted:classification"
                  ]
                }
              ]
            },
            "bcp-05-x-02": {
              "x_patch2vuln": {
                "assumptions": [
                  "The affected version boundary is inferred from the nearest git tag v2.5.47 with 51 commits after the fix; the exact last-affected and first-fixed release numbers are not explicitly stated in the patch metadata.",
                  "PR:L assumes the attacker needs an authenticated MISP account with event-import or org-creation privileges; the patch does not specify the exact permission level required.",
                  "The confidentiality impact is rated Low because the file read and existence oracle are limited to files with .png or .svg extensions; a broader arbitrary file read is not possible through this code path.",
                  "CAPEC-126 is selected as the closest match; no uncertainty is noted because the attack pattern (path traversal via crafted filename) maps directly to this CAPEC entry.",
                  "The commit date (2026-09-16) is taken at face value from the patch metadata; no independent verification of the timeline was performed."
                ],
                "capecRationale": [
                  {
                    "capecId": "CAPEC-126",
                    "rationale": "CAPEC-126 describes an attacker manipulating path components in a request to access files or directories outside the intended scope. This matches the vulnerability exactly: the org name field is manipulated to include ../ sequences, causing the server to probe and potentially read files outside the org-image directory. The mapping is direct and unambiguous."
                  }
                ],
                "commit": "e00986075acf47257030caa0298fad7b730a764c",
                "confidence": "high",
                "credits": [
                  {
                    "lang": "en",
                    "type": "reporter",
                    "value": "Jeroen Pinoy"
                  },
                  {
                    "lang": "en",
                    "type": "remediation developer",
                    "value": "iglocska"
                  },
                  {
                    "lang": "en",
                    "type": "remediation developer",
                    "value": "Claude Opus 4.8"
                  }
                ],
                "cvssRationale": "AV:N: MISP is a network-accessible web application; the org name is set via API/event import over the network. AC:L: The attacker simply includes ../ in an org name during a normal import operation; no race or complex condition is needed. AT:N: No manipulation of the target beyond normal input is required. PR:L: The attacker must be an authenticated MISP user with permission to create or import events that set the organization name. UI:N: No victim interaction is required; the traversal occurs server-side during view rendering. VC:L: The attacker gains a file-existence oracle and can read .png/.svg files outside the intended directory, a limited but real confidentiality breach. VI:N, VA:N: No integrity or availability impact is evidenced. SC:N, SI:N, SA:N: No secondary impacts are indicated by the patch or commit message.",
                "fixSummary": "The fix adds input validation in findOrgImage so that each organization identifier value (id, name, uuid) is checked to be a non-empty plain filename component. Values containing path separators (detected via basename comparison) or dot-dot sequences are rejected, preventing the constructed path from escaping the designated org-image directory. The validated value is then used in the file_exists call and the returned image reference.",
                "generatedAt": "2026-09-22T14:32:54.822056Z",
                "generator": "patch2vuln.py",
                "model": "qwen3.8:27b",
                "modelComparison": {
                  "rankings": [
                    {
                      "agreementScore": 9,
                      "assumptionCount": 5,
                      "confidence": "high",
                      "model": "qwen3.8:27b",
                      "score": 6
                    }
                  ],
                  "selectedModel": "qwen3.8:27b",
                  "selectionMethod": "deterministic-consensus-v1",
                  "selectionNotice": "The selected result is closest to model consensus; this heuristic does not establish factual correctness and human review remains required."
                },
                "patchSha256": "646ef716ac7a2ad0525154af9e0a6a7ffac9739a4f4d7114994521ae57d93fbd",
                "patchSummary": "In app/View/Helper/OrgImgHelper.php, the findOrgImage method now casts each field value to a string and rejects it if it is empty, if basename($value) does not equal $value (indicating embedded path separators), or if it contains the substring \u0027..\u0027. The previously unvalidated $options[$field] is replaced with the validated $value in both the file_exists() call and the assignment to $image. Ten lines added, two lines removed in a single file.",
                "patchTruncated": false,
                "patches": [
                  {
                    "commit": "e00986075acf47257030caa0298fad7b730a764c",
                    "patchSha256": "646ef716ac7a2ad0525154af9e0a6a7ffac9739a4f4d7114994521ae57d93fbd",
                    "source": "https://github.com/MISP/MISP/commit/e00986075.patch",
                    "sourceUrl": "https://github.com/MISP/MISP/commit/e00986075.patch",
                    "subject": "fix: [security] Reject path traversal in the org-logo"
                  }
                ],
                "source": "https://github.com/MISP/MISP/commit/e00986075.patch",
                "subject": "fix: [security] Reject path traversal in the org-logo",
                "tagVersionBoundary": {
                  "commits_after_fix": 51,
                  "repository": "https://github.com/MISP/MISP",
                  "tag": "v2.5.47",
                  "version": "2.5.47",
                  "version_type": "semver"
                },
                "weaknessRationale": [
                  {
                    "cweId": "CWE-22",
                    "rationale": "The vulnerability is a textbook path traversal: a user-controlled string (org name) is concatenated into a filesystem path without sanitization, allowing the resolved path to escape the intended directory. The fix explicitly rejects path separators and dot-dot sequences, confirming the root cause is missing path confinement."
                  }
                ]
              }
            }
          },
          "recordType": "advisory",
          "vulnId": "GCVE-1-2026-20266"
        }
      ]
    }
  },
  "cveMetadata": {
    "assignerOrgId": "00000000-0000-4000-9000-000000000000",
    "cveId": "CVE-2026-95698",
    "datePublished": "2026-09-22T14:36:25.776959Z",
    "dateReserved": "2026-09-22T14:36:33.476Z",
    "dateUpdated": "2026-09-22T14:36:33.546261Z",
    "state": "PUBLISHED",
    "vulnId": "GCVE-1-2026-20266"
  },
  "dataType": "CVE_RECORD",
  "dataVersion": "5.1"
}

GCVE-1-2026-20182 (CVE-2026-95697)

Vulnerability from gna-1 – Published: 2026-09-22 14:31 – Updated: 2026-09-22 14:31
VLAI
Title
MISP: Insufficient Authorization Allows Sharing Group Editors to Overwrite Organization Metadata
Summary
MISP contains an authorization flaw in the Organisation model's captureOrg method. When the $force parameter is set to true, the method unconditionally overwrites organization metadata fields without verifying that the invoking user holds sufficient privileges. A user with a sharing group (SG) editor role can trigger this code path, allowing them to modify organization metadata that should be restricted to site administrators or users with sync permissions. According to the commit message, this could lead to blueprint-based sharing group manipulation, meaning an attacker with SG editor access could alter organizational attributes in ways that influence how sharing groups and blueprints behave across the MISP instance.  The vulnerability requires an authenticated user with at least SG editor privileges and network access to the MISP web interface. The impact is primarily on the integrity of organization records and, potentially, on the integrity of sharing group configurations derived from those records.
CWE
Assigner
CIRCL GNA GNA-1
GNA scorecard A 91/100 over 279 records in the last 180 days details
References
Impacted products
Vendor Product Version
MISP MISP Affected: 0 , < 2.5.47 (semver)
Create a notification for this product.
GCVE extensions
bcp-05-x-01
AI-assisted vulnerability information annotation
GCVE-BCP-05-X-01
Whole record AI-generated Human-reviewed GNA-1

Draft vulnerability metadata was generated from a git-format patch using an Ollama-hosted language model. Human validation is required before publication.

ai-computer-assisted:llm-generatedai-computer-assisted:classification
Model Source Identifier
qwen3.8:27b ollama qwen3.8:27b
bcp-05-x-02
Patch-to-vulnerability generation provenance
GCVE-BCP-05-X-02
Generator
patch2vuln.py on 2026-09-22 14:23
Model
qwen3.8:27b
Input
https://github.com/MISP/MISP/commit/f3ec974ee.patch 63108ba85f58…
Confidence
medium
Commit Subject Patch SHA-256
f3ec974ee2d7 fix: [security] overwrite of org metadata by sg editors 63108ba85f58…
Fix summary

The fix adds an authorization check to the captureOrg method so that the forced overwrite of organization metadata fields is only permitted when the invoking user holds either the site_admin permission or the sync permission. This ensures that low-privilege roles such as sharing group editors can no longer trigger the metadata overwrite path, closing the authorization gap.

Patch summary

In app/Model/Organisation.php, the condition guarding the forced overwrite of organization metadata fields (type, date_created, date_modified, nationality, sector, contacts) was changed from a simple 'if ($force)' check to 'if ($force && (!empty($user['Role']['perm_site_admin']) || !empty($user['Role']['perm_sync'])))'. This one-line change adds a role-based authorization requirement so that only site administrators or sync-permitted users can execute the metadata overwrite when $force is true.

CVSS rationale

AV:N: MISP is a network-accessible web application. AC:L: The attack requires only calling the existing captureOrg function with $force=true; no race conditions or complex bypasses are needed. AT:N: No user interaction or attack tooling beyond normal API usage is required. PR:L: The attacker needs an authenticated account with at least SG editor privileges, which is a low-privilege role in MISP. UI:N: No victim interaction is required. VC:N: No confidentiality impact is evident from the patch. VI:H: Organization metadata (type, nationality, sector, contacts, dates) can be arbitrarily modified, representing high integrity impact on the vulnerable component. VA:N: No availability impact. SC:N: No direct confidentiality impact on other systems. SI:H: The commit message explicitly notes the issue 'could lead to blueprint based SG manipulation,' indicating integrity impact extends to sharing group configurations in the broader MISP ecosystem. SA:N: No availability impact on other systems.

Weakness rationale
  • CWE-862 The captureOrg method performed a privileged operation (overwriting organization metadata) based solely on the $force flag without verifying that the authenticated user held the necessary permission (site_admin or sync). The authorization check was entirely absent for this code path, which is the definition of a missing authorization vulnerability.
Attack pattern rationale
  • CAPEC-100 The closest plausible CAPEC is Parameter Tampering: an authenticated user with a lower-privilege role (SG editor) invokes the captureOrg function with the $force parameter set to true, triggering a code path that was intended only for higher-privilege users. The user manipulates a function parameter to cause unintended privileged behavior. This mapping is approximate because the core issue is a missing authorization check rather than classic parameter tampering, but no CAPEC entry more precisely describes an authenticated user exploiting a missing permission gate via a boolean flag.
Assumptions to verify
  • The affected version boundary is inferred from the tag_version_boundary metadata (v2.5.47, 139 commits after fix), suggesting the fix landed in or before v2.5.47. The exact first affected version is not specified in the patch.
  • The 'blueprint based SG manipulation' impact mentioned in the commit message is taken at face value for the SI:H rating; the patch itself only shows the org metadata overwrite fix and does not include code demonstrating the SG manipulation path.
  • CAPEC-100 (Parameter Tampering) is the closest available mapping; the vulnerability is more precisely a missing authorization check (CWE-862) than a classic parameter tampering scenario, but no CAPEC entry directly models 'authenticated user exploits missing permission gate via a boolean flag.'
  • The PR:L rating assumes that SG editor is a low-privilege role in MISP's permission hierarchy; the patch references perm_site_admin and perm_sync as the required higher-level permissions, implying SG editor is below that level.
  • No specific MISP version range is confirmed beyond the v2.5.47 tag boundary; earlier versions may or may not be affected depending on when the captureOrg method was introduced.
Model comparison

Selected qwen3.8:27b by deterministic-consensus-v1
The selected result is closest to model consensus; this heuristic does not establish factual correctness and human review remains required.

Model Score Agreement Confidence Assumptions
qwen3.8:27b 5 9 medium 5

{
  "containers": {
    "cna": {
      "affected": [
        {
          "modules": [
            "Organisation model (app/Model/Organisation.php)"
          ],
          "product": "MISP",
          "programFiles": [
            "app/Model/Organisation.php"
          ],
          "repo": "https://github.com/MISP/MISP",
          "vendor": "MISP",
          "versions": [
            {
              "lessThan": "2.5.47",
              "status": "affected",
              "version": "0",
              "versionType": "semver"
            }
          ]
        }
      ],
      "credits": [
        {
          "lang": "en",
          "type": "reporter",
          "value": "Jeroen Pinoy"
        },
        {
          "lang": "en",
          "type": "remediation developer",
          "value": "iglocska"
        }
      ],
      "descriptions": [
        {
          "lang": "en",
          "supportingMedia": [
            {
              "base64": false,
              "type": "text/html",
              "value": "\u003cp\u003eMISP contains an authorization flaw in the Organisation model\u0027s captureOrg method. When the $force parameter is set to true, the method unconditionally overwrites organization metadata fields without verifying that the invoking user holds sufficient privileges. A user with a sharing group (SG) editor role can trigger this code path, allowing them to modify organization metadata that should be restricted to site administrators or users with sync permissions.\u003c/p\u003e\u003cp\u003eAccording to the commit message, this could lead to blueprint-based sharing group manipulation, meaning an attacker with SG editor access could alter organizational attributes in ways that influence how sharing groups and blueprints behave across the MISP instance.\u0026nbsp;\u003c/p\u003e\u003cp\u003eThe vulnerability requires an authenticated user with at least SG editor privileges and network access to the MISP web interface. The impact is primarily on the integrity of organization records and, potentially, on the integrity of sharing group configurations derived from those records.\u003c/p\u003e"
            }
          ],
          "value": "MISP contains an authorization flaw in the Organisation model\u0027s captureOrg method. When the $force parameter is set to true, the method unconditionally overwrites organization metadata fields without verifying that the invoking user holds sufficient privileges. A user with a sharing group (SG) editor role can trigger this code path, allowing them to modify organization metadata that should be restricted to site administrators or users with sync permissions.\n\nAccording to the commit message, this could lead to blueprint-based sharing group manipulation, meaning an attacker with SG editor access could alter organizational attributes in ways that influence how sharing groups and blueprints behave across the MISP instance.\u00a0\n\nThe vulnerability requires an authenticated user with at least SG editor privileges and network access to the MISP web interface. The impact is primarily on the integrity of organization records and, potentially, on the integrity of sharing group configurations derived from those records."
        }
      ],
      "impacts": [
        {
          "capecId": "CAPEC-100",
          "descriptions": [
            {
              "lang": "en",
              "value": "CAPEC-100 Parameter Tampering"
            }
          ]
        }
      ],
      "metrics": [
        {
          "cvssV4_0": {
            "Automatable": "NOT_DEFINED",
            "Recovery": "NOT_DEFINED",
            "Safety": "NOT_DEFINED",
            "attackComplexity": "LOW",
            "attackRequirements": "NONE",
            "attackVector": "NETWORK",
            "baseScore": 5.3,
            "baseSeverity": "MEDIUM",
            "privilegesRequired": "LOW",
            "providerUrgency": "NOT_DEFINED",
            "subAvailabilityImpact": "NONE",
            "subConfidentialityImpact": "NONE",
            "subIntegrityImpact": "LOW",
            "userInteraction": "NONE",
            "valueDensity": "NOT_DEFINED",
            "vectorString": "CVSS:4.0/AV:N/AC:L/AT:N/PR:L/UI:N/VC:N/VI:L/VA:N/SC:N/SI:L/SA:N",
            "version": "4.0",
            "vulnAvailabilityImpact": "NONE",
            "vulnConfidentialityImpact": "NONE",
            "vulnIntegrityImpact": "LOW",
            "vulnerabilityResponseEffort": "NOT_DEFINED"
          },
          "format": "CVSS",
          "scenarios": [
            {
              "lang": "en",
              "value": "GENERAL"
            }
          ]
        }
      ],
      "problemTypes": [
        {
          "descriptions": [
            {
              "cweId": "CWE-862",
              "description": "CWE-862 Missing Authorization",
              "lang": "en",
              "type": "CWE"
            }
          ]
        }
      ],
      "providerMetadata": {
        "orgId": "00000000-0000-4000-9000-000000000000"
      },
      "references": [
        {
          "name": "Security patch",
          "tags": [
            "patch"
          ],
          "url": "https://github.com/MISP/MISP/commit/f3ec974ee"
        }
      ],
      "solutions": [
        {
          "lang": "en",
          "supportingMedia": [
            {
              "base64": false,
              "type": "text/html",
              "value": "\u003cp\u003eThe fix adds an authorization check to the captureOrg method so that the forced overwrite of organization metadata fields is only permitted when the invoking user holds either the site_admin permission or the sync permission. This ensures that low-privilege roles such as sharing group editors can no longer trigger the metadata overwrite path, closing the authorization gap.\u003c/p\u003e"
            }
          ],
          "value": "The fix adds an authorization check to the captureOrg method so that the forced overwrite of organization metadata fields is only permitted when the invoking user holds either the site_admin permission or the sync permission. This ensures that low-privilege roles such as sharing group editors can no longer trigger the metadata overwrite path, closing the authorization gap."
        }
      ],
      "title": "MISP: Insufficient Authorization Allows Sharing Group Editors to Overwrite Organization Metadata",
      "x_gcve": [
        {
          "extensions": {
            "bcp-05-x-01": {
              "ai_annotations": [
                {
                  "ai_level": "generated",
                  "description": "Draft vulnerability metadata was generated from a git-format patch using an Ollama-hosted language model. Human validation is required before publication.",
                  "gna_source": 1,
                  "models": [
                    {
                      "gna_source": 1,
                      "identifier": "qwen3.8:27b",
                      "name": "qwen3.8:27b",
                      "source": "ollama"
                    }
                  ],
                  "review_status": "full",
                  "scope": "record",
                  "tags": [
                    "ai-computer-assisted:llm-generated",
                    "ai-computer-assisted:classification"
                  ]
                }
              ]
            },
            "bcp-05-x-02": {
              "x_patch2vuln": {
                "assumptions": [
                  "The affected version boundary is inferred from the tag_version_boundary metadata (v2.5.47, 139 commits after fix), suggesting the fix landed in or before v2.5.47. The exact first affected version is not specified in the patch.",
                  "The \u0027blueprint based SG manipulation\u0027 impact mentioned in the commit message is taken at face value for the SI:H rating; the patch itself only shows the org metadata overwrite fix and does not include code demonstrating the SG manipulation path.",
                  "CAPEC-100 (Parameter Tampering) is the closest available mapping; the vulnerability is more precisely a missing authorization check (CWE-862) than a classic parameter tampering scenario, but no CAPEC entry directly models \u0027authenticated user exploits missing permission gate via a boolean flag.\u0027",
                  "The PR:L rating assumes that SG editor is a low-privilege role in MISP\u0027s permission hierarchy; the patch references perm_site_admin and perm_sync as the required higher-level permissions, implying SG editor is below that level.",
                  "No specific MISP version range is confirmed beyond the v2.5.47 tag boundary; earlier versions may or may not be affected depending on when the captureOrg method was introduced."
                ],
                "capecRationale": [
                  {
                    "capecId": "CAPEC-100",
                    "rationale": "The closest plausible CAPEC is Parameter Tampering: an authenticated user with a lower-privilege role (SG editor) invokes the captureOrg function with the $force parameter set to true, triggering a code path that was intended only for higher-privilege users. The user manipulates a function parameter to cause unintended privileged behavior. This mapping is approximate because the core issue is a missing authorization check rather than classic parameter tampering, but no CAPEC entry more precisely describes an authenticated user exploiting a missing permission gate via a boolean flag."
                  }
                ],
                "commit": "f3ec974ee2d72d77311dbb364c70f1aee83a58c2",
                "confidence": "medium",
                "credits": [
                  {
                    "lang": "en",
                    "type": "reporter",
                    "value": "Jeroen Pinoy"
                  },
                  {
                    "lang": "en",
                    "type": "remediation developer",
                    "value": "iglocska"
                  }
                ],
                "cvssRationale": "AV:N: MISP is a network-accessible web application. AC:L: The attack requires only calling the existing captureOrg function with $force=true; no race conditions or complex bypasses are needed. AT:N: No user interaction or attack tooling beyond normal API usage is required. PR:L: The attacker needs an authenticated account with at least SG editor privileges, which is a low-privilege role in MISP. UI:N: No victim interaction is required. VC:N: No confidentiality impact is evident from the patch. VI:H: Organization metadata (type, nationality, sector, contacts, dates) can be arbitrarily modified, representing high integrity impact on the vulnerable component. VA:N: No availability impact. SC:N: No direct confidentiality impact on other systems. SI:H: The commit message explicitly notes the issue \u0027could lead to blueprint based SG manipulation,\u0027 indicating integrity impact extends to sharing group configurations in the broader MISP ecosystem. SA:N: No availability impact on other systems.",
                "fixSummary": "The fix adds an authorization check to the captureOrg method so that the forced overwrite of organization metadata fields is only permitted when the invoking user holds either the site_admin permission or the sync permission. This ensures that low-privilege roles such as sharing group editors can no longer trigger the metadata overwrite path, closing the authorization gap.",
                "generatedAt": "2026-09-22T14:23:51.691782Z",
                "generator": "patch2vuln.py",
                "model": "qwen3.8:27b",
                "modelComparison": {
                  "rankings": [
                    {
                      "agreementScore": 9,
                      "assumptionCount": 5,
                      "confidence": "medium",
                      "model": "qwen3.8:27b",
                      "score": 5
                    }
                  ],
                  "selectedModel": "qwen3.8:27b",
                  "selectionMethod": "deterministic-consensus-v1",
                  "selectionNotice": "The selected result is closest to model consensus; this heuristic does not establish factual correctness and human review remains required."
                },
                "patchSha256": "63108ba85f58bd5ec8318ea052879e536aaa83464f8ff4f90ef1d386913dc835",
                "patchSummary": "In app/Model/Organisation.php, the condition guarding the forced overwrite of organization metadata fields (type, date_created, date_modified, nationality, sector, contacts) was changed from a simple \u0027if ($force)\u0027 check to \u0027if ($force \u0026\u0026 (!empty($user[\u0027Role\u0027][\u0027perm_site_admin\u0027]) || !empty($user[\u0027Role\u0027][\u0027perm_sync\u0027])))\u0027. This one-line change adds a role-based authorization requirement so that only site administrators or sync-permitted users can execute the metadata overwrite when $force is true.",
                "patchTruncated": false,
                "patches": [
                  {
                    "commit": "f3ec974ee2d72d77311dbb364c70f1aee83a58c2",
                    "patchSha256": "63108ba85f58bd5ec8318ea052879e536aaa83464f8ff4f90ef1d386913dc835",
                    "source": "https://github.com/MISP/MISP/commit/f3ec974ee.patch",
                    "sourceUrl": "https://github.com/MISP/MISP/commit/f3ec974ee.patch",
                    "subject": "fix: [security] overwrite of org metadata by sg editors"
                  }
                ],
                "source": "https://github.com/MISP/MISP/commit/f3ec974ee.patch",
                "subject": "fix: [security] overwrite of org metadata by sg editors",
                "tagVersionBoundary": {
                  "commits_after_fix": 139,
                  "repository": "https://github.com/MISP/MISP",
                  "tag": "v2.5.47",
                  "version": "2.5.47",
                  "version_type": "semver"
                },
                "weaknessRationale": [
                  {
                    "cweId": "CWE-862",
                    "rationale": "The captureOrg method performed a privileged operation (overwriting organization metadata) based solely on the $force flag without verifying that the authenticated user held the necessary permission (site_admin or sync). The authorization check was entirely absent for this code path, which is the definition of a missing authorization vulnerability."
                  }
                ]
              }
            }
          },
          "recordType": "advisory",
          "vulnId": "GCVE-1-2026-20182"
        }
      ]
    }
  },
  "cveMetadata": {
    "assignerOrgId": "00000000-0000-4000-9000-000000000000",
    "cveId": "CVE-2026-95697",
    "datePublished": "2026-09-22T14:31:15.792405Z",
    "dateReserved": "2026-09-22T14:31:23.351Z",
    "dateUpdated": "2026-09-22T14:31:23.423452Z",
    "state": "PUBLISHED",
    "vulnId": "GCVE-1-2026-20182"
  },
  "dataType": "CVE_RECORD",
  "dataVersion": "5.1"
}
displaying 81 - 90 publications in total 355