CVE-2026-111391
Code injection in SGLang 0.5.10‑0.5.22 via crafted filter bypassing eval sandbox.
- CVSS 9.2
- CWE-95
- Input Validation and Sanitization
- Remote
SGLang 0.5.10 through 0.5.22 contains a code injection vulnerability in _evaluate_filter() that allows remote attackers to execute Python code by escaping the eval() sandbox through re.search __globals__. Attackers can send a crafted filter expression to the unauthenticated /dumper/configure endpoint when DUMPER_SERVER_PORT is set, running code in the scheduler process on the next forward pass.
- CWE
- CWE-95
- CVSS base score
- 9.2
- Published
- 2026-10-16
- OWASP
- A03 Injection
- Orthogonal defect classification
- Checking
- Code defect classification
- Incorrect Check
- Category
- Input Validation and Sanitization
- Subcategory
- Command Injection
- Accessibility scope
- Remote
- Impact
- Arbitrary Code Execution
- Affected component
- SGLang
- Fixed by upgrading
- Yes
Solution
Upgrade SGLang to version 0.5.23 or newer, which patches the _evaluate_filter() code‑injection flaw.
Vulnerable code sample
import re
class Configurator:
def __init__(self):
self.allowed_keys = {"mode", "threshold"}
def _evaluate_filter(self, filter_expr: str) -> bool:
# VULNERABLE: eval with unrestricted globals allows __globals__ escape
return eval(filter_expr, {"re": re})
def configure(self, payload: dict):
filter_expr = payload.get("filter", "True")
if self._evaluate_filter(filter_expr):
for k, v in payload.get("settings", {}).items():
if k in self.allowed_keys:
setattr(self, k, v)Patched code sample
import re
import ast
class Configurator:
def __init__(self):
self.allowed_keys = {"mode", "threshold"}
def _evaluate_filter(self, filter_expr: str) -> bool:
# FIX: use ast.literal_eval to prevent code execution
try:
result = ast.literal_eval(filter_expr)
if isinstance(result, bool):
return result
except Exception:
pass
# fallback for simple regex checks without eval
return bool(re.search(filter_expr, ""))
def configure(self, payload: dict):
filter_expr = payload.get("filter", "True")
if self._evaluate_filter(filter_expr):
for k, v in payload.get("settings", {}).items():
if k in self.allowed_keys:
setattr(self, k, v)Cite this entry
@misc{vaitp:cve2026111391,
title = {{Code injection in SGLang 0.5.10‑0.5.22 via crafted filter bypassing eval sandbox.}},
author = {Bogaerts, Fr\'ed\'eric and Ivaki, Naghmeh and Fonseca, Jos\'e},
year = {2026},
note = {VAITP Python Vulnerability Dataset, entry CVE-2026-111391},
howpublished = {\url{https://netpack.pt/vaitp/vulnerability/CVE-2026-111391/}}
}
Introducing the "VAITP dataset": a specialized repository of Python vulnerabilities and patches, meticulously compiled for the use of the security research community. As Python's prominence grows, understanding and addressing potential security vulnerabilities become crucial. Crafted by and for the cybersecurity community, this dataset offers a valuable resource for researchers, analysts, and developers to analyze and mitigate the security risks associated with Python. Through the comprehensive exploration of vulnerabilities and corresponding patches, the VAITP dataset fosters a safer and more resilient Python ecosystem, encouraging collaborative advancements in programming security.
The supreme art of war is to subdue the enemy without fighting.
Sun Tzu – “The Art of War”
:: Shaping the future through research and ingenuity ::
