PatchSiren cyber security CVE debrief
CVE-2026-114612 BerriAI CVE debrief
AI-assisted PatchSiren debrief based on the supplied source corpus. The CVE record was published on 2026-10-22T18:01:52.768Z and has not been modified since then. The vulnerability in LiteLLM, a proxy server for LLM APIs, allows authenticated internal users to gain administrative access and execute arbitrary commands through crafted metadata containing a forged administrator credential. This issue affects versions 1.87.0 through 1.100.4, 1.101.3, 1.102.2, 1.103.1, and 1.104.0-rc.2. Defenders responsible for LiteLLM deployments should assess exposure and prioritize verification and upgrades to fixed versions.
- Vendor
- BerriAI
- Product
- litellm
- CVSS
- CRITICAL 9.9
- CISA KEV
- Not listed in stored evidence
- Original CVE published
- 2026-10-22
- Original CVE updated
- 2026-10-30
- Advisory published
- 2026-10-22
- Advisory updated
- 2026-10-30
Who should care
Defenders responsible for LiteLLM deployments, particularly those using versions 1.87.0 through 1.100.4, 1.101.3, 1.102.2, 1.103.1, and 1.104.0-rc.2, should assess exposure and prioritize verification and upgrades to fixed versions.
Why it matters
CVE-2026-114612 is a critical vulnerability in LiteLLM that allows authenticated internal users to gain administrative access and execute arbitrary commands. Defenders should prioritize verifying and upgrading to fixed versions, assessing exposure in their environments, and monitoring for potential exploitation attempts.
- Potential for authenticated internal users to gain administrative access
- Possible execution of arbitrary commands through the MCP stdio endpoint
- Need for verification of LiteLLM versions and upgrades to fixed versions
- Importance of monitoring for potential exploitation attempts
Technical summary
LiteLLM, a proxy server for LLM APIs, has a vulnerability in versions 1.87.0 through 1.100.4, 1.101.3, 1.102.2, 1.103.1, and 1.104.0-rc.2. The vulnerability allows an authenticated internal user to craft metadata containing a forged administrator credential, which can be replayed as a bearer token to gain administrative access and execute arbitrary commands through the MCP stdio endpoint. This issue arises from the proxy reusing one cryptographic context for at-rest secrets and session tokens. Defenders should prioritize verifying and upgrading to fixed versions of LiteLLM, specifically 1.100.4, 1.101.3, 1.102.2, 1.103.1, or 1.104.0-rc.2, and assess exposure in their environments.
Defensive priority
Defenders should prioritize verifying and upgrading to fixed versions of LiteLLM, specifically 1.100.4, 1.101.3, 1.102.2, 1.103.1, or 1.104.0-rc.2, and assess exposure in their environments.
Recommended defensive actions
- Verify and upgrade to fixed versions of LiteLLM (1.100.4, 1.101.3, 1.102.2, 1.103.1, or 1.104.0-rc.2)
- Assess exposure in environments using affected versions of LiteLLM
- Monitor for potential exploitation attempts
- Review compensating controls for exposed systems while remediation is scheduled and verified
- Check relevant monitoring, detection, and logs for exposed assets that need extra review
- Track exceptions, retest remediated assets, and close the item only after evidence is documented
- Confirm whether affected product deployments exist in managed environments and assign an owner for follow-up
Evidence notes
The CVE record and source item provide details on the vulnerability in LiteLLM, including affected versions and fixed versions. However, there is limited information on potential exploitation or victim impact.
Sources and references
Verified primary and authoritative sources
-
CVE-2026-114612 CVE Program record
Publisher, destination, and source semantics verified
URL: https://www.cve.org/CVERecord?id=CVE-2026-114612
CVE Program - Official CVE Program record with source-provided CVE metadata.
-
CVE-2026-114612 NVD vulnerability detail
Publisher, destination, and source semantics verified
URL: https://nvd.nist.gov/vuln/detail/CVE-2026-114612
NIST National Vulnerability Database - Official NIST NVD detail page and source-specific vulnerability assessment.
Supplemental references
-
LiteLLM: Privilege Escalation to Proxy Admin via Cross-Domain Reuse of the Salt Key
Unverified legacy reference
URL: https://raw.githubusercontent.com/CVEProject/cvelistV5/main/cves/2026/114xxx/CVE-2026-114612.json
cve_program_cvelist_v5
-
Source reference
Unverified legacy reference
URL: https://github.com/BerriAI/litellm/security/advisories/GHSA-7hp6-4w63-5g45
Supplemental source - x_refsource_CONFIRM
-
Source reference
Unverified legacy reference
URL: https://github.com/BerriAI/litellm/commit/1d5fdc87fed66b49e1df8cf7d708ad93249eb7d0
Supplemental source - x_refsource_MISC
-
Source reference
Unverified legacy reference
URL: https://github.com/BerriAI/litellm/commit/9161d7f9d6073603ba61ee4aa5e2c0ef5b37a81b
Supplemental source - x_refsource_MISC
-
Source reference
Unverified legacy reference
URL: https://github.com/BerriAI/litellm/commit/9fa1a641119dd0d4fe43e93622eae5f482ceb63f
Supplemental source - x_refsource_MISC
-
Source reference
Unverified legacy reference
URL: https://github.com/BerriAI/litellm/commit/a80eacacee118fbed54e3d98f239d17a63fac818
Supplemental source - x_refsource_MISC
-
Source reference
Unverified legacy reference
URL: https://github.com/BerriAI/litellm/commit/f7cd90f09e9716652b38daec9f0640cdf49d582b
Supplemental source - x_refsource_MISC
-
Source reference
Unverified legacy reference
URL: https://github.com/BerriAI/litellm/releases/tag/v1.100.4
Supplemental source - x_refsource_MISC
Methodology and review provenance
AI-assisted synthesis based on stored public vulnerability evidence. System validation, approval state, and publication status do not by themselves establish human review of this revision. PatchSiren helps prioritize defensive review and does not prove exposure or remediation on any system.