Security readout for executives and security teams
Plain-English summary
LiteLLM versions through 1.82.4 reportedly expose a prompt-testing function that can let an unauthenticated remote attacker run operating-system commands. Successful exploitation could enable data theft, service disruption, credential exposure, or full compromise of the LiteLLM host.
Executive priority
Treat internet-reachable affected deployments as an immediate remediation priority. Restrict the endpoint now, investigate possible compromise, and track vendor guidance. Internal-only instances remain urgent because unauthenticated access may still be possible from compromised or untrusted internal systems.
Technical view
CVE-2026-37004 is a server-side template injection vulnerability in LiteLLM's /prompts/test endpoint. Crafted dotprompt_content is reportedly processed using an unsandboxed jinja2.Environment, enabling arbitrary OS command execution without authentication or user interaction. The supplied CVSS 3.1 score is 9.8.
Likely exposure
Exposure is highest where LiteLLM 1.82.4 or earlier is deployed and /prompts/test is reachable by untrusted networks. The bundle's structured affected-product metadata is incomplete, so organisations should confirm versions and endpoint availability directly in their environments.
Exploitation context
The supplied bundle does not establish active exploitation, and the CVE is not identified as being in KEV. However, the reported attack requires no privileges or user interaction and has low complexity. Public technical disclosure may increase exploitation interest.
Researcher notes
The central claim is an unauthenticated SSTI-to-RCE path through dotprompt_content and an unsandboxed Jinja2 environment. The pinned source reference supports code-level review, but the supplied record provides no confirmed fixed version, patch identifier, or evidence of exploitation in the wild.
Mitigation direction
- Check current LiteLLM vendor guidance and apply the vendor-supported remediation when available.
- Temporarily block or tightly restrict access to /prompts/test from untrusted networks.
- Remove unnecessary public exposure of LiteLLM administrative and testing interfaces.
- Rotate credentials accessible to exposed LiteLLM hosts if compromise is suspected.
Validation and detection
- Inventory LiteLLM deployments and identify versions at or below 1.82.4.
- Confirm whether /prompts/test is enabled and reachable without authentication.
- Review proxy and application logs for unusual requests containing dotprompt_content.
- Inspect exposed hosts for unexpected processes, file changes, or credential access.
- After remediation, verify the endpoint is restricted using a safe, non-executing request.
Public sources used
Generated from the cited source records. This long-tail analysis has not been individually reviewed by a named human.
Conservative CVE-to-ATT&CK context
These mappings and lookup hints may be relevant to the vulnerability behavior, CWE, affected product, or exposure path. Glexia-inferred context is not an official MITRE, ATT&CK, CWE, or CVE Program mapping.
ATT&CK lookup starting points
Use these exact CWE pages and searches to review the Glexia ATT&CK library from this CVE's weakness and description context.
CWE-1336: Exact CWE lookup
Use the exact CWE identifier as the starting point before reviewing related ATT&CK behavior. Open the exact CWE lookup page first, then review the ATT&CK searches from that MITRE weakness context. This is a Glexia lookup hint, not an official ATT&CK mapping.
Open ATT&CK lookupCVE-2026-37004 mapping review
Open the CVE-to-ATT&CK bridge for reviewed, inferred, or future official mappings tied to this CVE.
Open ATT&CK lookup- Severity
- Critical
- CVSS
- 9.8 (3.1)
- Known Exploited
- No
- Published
Vector: CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
CNA and ADP enrichment extracted from CVE v5
These fields come from the CVE record and ADP containers, not from Glexia's Take. They preserve time-varying source decisions such as CISA SSVC, KEV status, CVSS metrics, and provider references.
SSVC decision data
CVSS vector scores
1 official scoreWe collect every scored CVSS vector available in the official CNA and ADP containers. When more than one version is present, the table keeps the source vectors side by side instead of collapsing them into the highest score.
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H3.95.9CISA-ADPVulnerability scoring details
Base CVSS 3.1 score
9.8CriticalVector: CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
Vulnerability timeline
Timeline events are normalized from CVE metadata, CNA source timelines, ADP timelines, and KEV metadata when present.
- CVE reservedCVE Program
The CVE ID was reserved by the assigning CNA.
- CVE publishedCVE Program
The CVE record was published.
- CVE updatedCVE Program
The CVE record metadata indicates this as the latest update time.
ADP provider summaries
Source materials
- CVE List V5 sourceCVE List V5
- https://github.com/BerriAI/litellm/blob/244bdffd1bfe7bebdfdef516e1ebe426a898e2f0/litellm/proxy/prompts/prompt_endpoints.py#L1073CVE reference, mitre
- https://yerangamage.com/cves/detail/?slug=litellm-ssti-rceCVE reference, mitre
Products and packages named in the record
CWE details
CWE links open Glexia weakness intelligence pages with official CWE context, developer remediation guidance, and related CVE mappings.
Improper Neutralization of Special Elements Used in a Template Engine
Improper Neutralization of Special Elements Used in a Template Engine represents a recurring weakness pattern that can create exploitable paths when design, validation, or implementation controls are missing.