CVE-2026-59821LiteLLM Code Injection
Prior to 1.82.0-stable, LiteLLM's Custom Code Guardrails production create and update paths did not apply the same sandboxing and validation used by the test endpoint, allowing a privileged user with access to create or update guardrails to submit custom Python code that…
Published Jul 8, 2026 · Updated Oct 21, 2026 · Also known as EUVD-2026-42350, GHSA-72m8-9m7m-h278, PYSEC-2026-3478
Key Facts
Patch immediately- CVE-2026-59821 is a low-severity vulnerability (CVSS 2.1 v4) affecting LiteLLM, airflow-2 (minimos), airflow-3 (minimos) and 1 more product.
- It is exploited in the wild, listed in Industry KEV feed since Jul 26, 2026. No public exploit code is known.
- EPSS estimates a 0.9% probability of exploitation in the next 30 days.
- Fixed versions are available, including airflow-2 2.11.2-r0, airflow-3 3.1.8-r0, LiteLLM 1.80.11+sp999.
- Brandefense priority: patch immediately, because it is listed as exploited in the wild.
Scores
CVSS v4
2.1
CVSS:4.0/AV:N/AC:L/AT:P/PR:H/UI:N/VC:L/VI:L/VA:L/SC:N/SI:N/SA:N
EPSS (30-day exploit likelihood)
0.9%
Exploit maturity
Exploited (KEV)
ATLAS score
41/100
Publication score, see How it works
Exploitation Verdict
KEV listed
Jul 26, 2026
18 days after publication
Federal due date
None
No federal due date
Ransomware use
Not reported
Attributed actors
None named
Exploit & Defence Guidance
Exploit maturity
Exploited (KEV)
Public exploit code
0
Proof-of-concept links
In-the-wild evidence
1
Reports of active use
Advisories And Exploit References
Description
LiteLLM is a proxy server (AI Gateway) to call LLM APIs in OpenAI (or native) format. Prior to 1.82.0-stable, LiteLLM's Custom Code Guardrails production create and update paths did not apply the same sandboxing and validation used by the test endpoint, allowing a privileged user with access to create or update guardrails to submit custom Python code that executed in the LiteLLM proxy environment and could expose secrets available to the process.
This issue is fixed in version 1.82.0-stable.
Known-Exploited Catalog Listings
| Catalog | Added | Remediation due | Ransomware use |
|---|---|---|---|
| Industry KEV feed | Jul 26, 2026 | n/a | Not reported |
Affected Products
9 affected entries in total| Product | Affected versions | Fixed in |
|---|---|---|
| LiteLLMLitellm | 1.82.0-nightly, < 1.82.0-stable | 1.82.0 |
| airflow-2minimos | See advisory | 2.11.2-r0 |
| airflow-3minimos | See advisory | 3.1.8-r0 |
| seal-litellmseal | See advisory | 1.80.11+sp999 |
Patched Releases
airflow-2
2.11.2-r0
airflow-3
3.1.8-r0
LiteLLM
1.80.11+sp999, 1.82.0, 1.82.0-stable
seal-litellm
1.80.11+sp999
Timeline
Publicly disclosed
Jul 8, 2026
Added to Industry KEV feed
Jul 26, 2026
Last updated
Oct 21, 2026
Weakness Type
- CWE-94 Improper Control of Generation of Code ('Code Injection')
References
- advisorycve.org · CVERecord
- advisorynvd.nist.gov · CVE-2026-59821
- advisorygithub.com · advisories
- advisoryGHSA advisory
- advisoryUSOM advisory
FAQ
Common Questions.
LiteLLM is a proxy server (AI Gateway) to call LLM APIs in OpenAI (or native) format.
Affected: LiteLLM versions 1.82.0-nightly, < 1.82.0-stable; airflow-2 (minimos); airflow-3 (minimos), among others. The full list is in the CVE brief.
It is exploited in the wild, listed in Industry KEV feed since Jul 26, 2026.
No, it is not currently listed in the CISA Known Exploited Vulnerabilities catalog.
No public exploit code is known.
Fixed versions are available, including airflow-2 2.11.2-r0, airflow-3 3.1.8-r0, LiteLLM 1.80.11+sp999.
Brandefense priority: patch immediately, because it is listed as exploited in the wild.
Is CVE-2026-59821 On Your Attack Surface?
Brandefense matches it against the products discovered on your perimeter and opens a finding the day a CVE like this is listed.
Request A Demo