CVE-2026-97674: Langflow: code injection allows authenticated OS command execution

HIGH

Published October 7, 2026

CISO Take

Langflow OSS versions 1.0.0 through 1.12.2 contain a code injection flaw (CWE-94) that lets any authenticated user run arbitrary OS commands on the server hosting the platform. The CVSS score is 8.1 (high), and the attack needs only low privileges, no user interaction and low complexity over the network. Exploitation pressure is currently modest: EPSS is 0.00294, the CVE is not in CISA KEV, and there is no public exploit or Nuclei template. Langflow has a record of code-execution bugs being weaponized quickly once details circulate, so the window is likely to be short. Upgrade to a release newer than 1.12.2 as described in the IBM advisory (https://www.ibm.com/support/pages/node/7290694), restrict Langflow account creation, and review who holds credentials on internet-exposed instances.

Sources:NVDEPSSIBM Support (ibm.com)ATLAS

What is the risk?

Risk is HIGH, not critical. The flaw is easy to exploit (network, low complexity, no user interaction) but requires a valid account, which lowers exposure on properly isolated deployments. The CVSS vector rates confidentiality and integrity impact as high and availability impact as none. Real-world impact can exceed that rating, because Langflow servers usually hold LLM provider API keys, vector DB credentials and database connection strings. Exploitation signals are weak today (EPSS 0.00294, no KEV listing, no public PoC, no scanner template). Treat any instance with open self-registration, shared or weak credentials, or internet exposure as higher risk.

How does the attack unfold?

Initial Access

Attacker authenticates to Langflow with a low-privileged account obtained via open registration or stolen credentials.

AML.T0012

Exploitation

Attacker submits crafted input that is executed as code by the server, exploiting the CWE-94 flaw.

AML.T0049

Execution

Injected code runs arbitrary OS commands under the Langflow service account.

AML.T0050

Impact

Attacker steals LLM API keys, database credentials and flow data, and pivots into internal systems.

Initial Access

Attacker authenticates to Langflow with a low-privileged account obtained via open registration or stolen credentials.

AML.T0012

Exploitation

Attacker submits crafted input that is executed as code by the server, exploiting the CWE-94 flaw.

AML.T0049

Execution

Injected code runs arbitrary OS commands under the Langflow service account.

AML.T0050

Impact

Attacker steals LLM API keys, database credentials and flow data, and pivots into internal systems.

What systems are affected?

Package	Ecosystem	Vulnerable Range	Patched
Langflow	pip	—	No patch
156.0KPushed 6d ago18% patched~66d to patchFull package profile →

Do you use Langflow? You're affected.

How severe is it?

CVSS 3.1

8.1 / 10

EPSS

0.3%

chance of exploitation in 30 days

Higher than 24% of all CVEs

Source: EPSS v3 — FIRST.org

Exploitation Status

No known exploitation

Sophistication

Trivial

What is the attack surface?

AVNetwork

ACLow

PRLow

UINone

SUnchanged

CHigh

IHigh

ANone

What should I do?

1 step

1) Upgrade Langflow beyond 1.12.2 following the IBM advisory (https://www.ibm.com/support/pages/node/7290694). 2) Until patched, disable open registration, remove unused accounts, enforce strong authentication and restrict network access to the Langflow UI and API (VPN, IP allowlist, reverse proxy with SSO). 3) Run Langflow in a container with a non-root user, a read-only filesystem where possible, minimal egress and no access to the host Docker socket. 4) Rotate any LLM API keys, database credentials and secrets stored in or reachable from Langflow if compromise is suspected. 5) Detection: alert on unexpected child processes spawned by the Langflow process (shell, curl, wget, python -c), anomalous outbound connections from the host, and unusual flow or component creation by low-privileged users. Review API logs for authenticated requests that submit custom component or code payloads.

What does CISA's SSVC say?

DecisionTrack

Exploitationnone

AutomatableNo

Technical Impacttotal

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

How is it classified?

Code ExecutionFrameworkAgentAML.T0012- Valid Accounts AML.T0049- Exploit Public-Facing Application AML.T0050- Command and Scripting Interpreter

Which compliance frameworks are affected?

This CVE is relevant to:

EU AI Act

Article 15-Accuracy, robustness and cybersecurity

ISO 42001

A.6.2.6-AI system operation and monitoring

NIST AI RMF

MEASURE 2.7-AI system security and resilience are evaluated and documented

OWASP LLM Top 10

LLM03-Supply Chain

Frequently Asked Questions

What is CVE-2026-97674?

Is CVE-2026-97674 actively exploited?

No confirmed active exploitation of CVE-2026-97674 has been reported, but organizations should still patch proactively.

How to fix CVE-2026-97674?

1) Upgrade Langflow beyond 1.12.2 following the IBM advisory (https://www.ibm.com/support/pages/node/7290694). 2) Until patched, disable open registration, remove unused accounts, enforce strong authentication and restrict network access to the Langflow UI and API (VPN, IP allowlist, reverse proxy with SSO). 3) Run Langflow in a container with a non-root user, a read-only filesystem where possible, minimal egress and no access to the host Docker socket. 4) Rotate any LLM API keys, database credentials and secrets stored in or reachable from Langflow if compromise is suspected. 5) Detection: alert on unexpected child processes spawned by the Langflow process (shell, curl, wget, python -c), anomalous outbound connections from the host, and unusual flow or component creation by low-privileged users. Review API logs for authenticated requests that submit custom component or code payloads.

What systems are affected by CVE-2026-97674?

This vulnerability affects the following AI/ML architecture patterns: LLM orchestration frameworks, RAG pipelines, agent frameworks, low-code AI workflow builders.

What is the CVSS score for CVE-2026-97674?

CVE-2026-97674 has a CVSS v3.1 base score of 8.1 (HIGH). The EPSS exploitation probability is 0.31%.

What is the AI security impact?

Affected AI Architectures

LLM orchestration frameworksRAG pipelinesagent frameworkslow-code AI workflow builders

MITRE ATLAS Techniques

AML.T0012Valid Accounts

AML.T0049Exploit Public-Facing Application

AML.T0050Command and Scripting Interpreter

Compliance Controls Affected

EU AI Act: Article 15

ISO 42001: A.6.2.6

NIST AI RMF: MEASURE 2.7

OWASP LLM Top 10: LLM03

What are the technical details?

Original Advisory

IBM Langflow OSS 1.0.0 through 1.12.2 could allow a remote authenticated attacker to execute arbitrary OS commands due to improper neutralization of special elements used in an OS command ('Code Injection'), aka improper control of code generation.

Exploitation Scenario

An attacker obtains a low-privileged Langflow account through open registration, a leaked credential or a phished employee. They authenticate and submit crafted input to a vulnerable code-generation or evaluation path, which executes injected code on the server. They then run OS commands to dump environment variables containing OpenAI, Anthropic and vector DB keys, read stored flows and data, and open a reverse shell to move laterally into the internal network or the data stores behind the RAG pipeline.

Weaknesses (CWE)

CWE-94Improper Control of Generation of Code ('Code Injection')Primary CWE-94Improper Control of Generation of Code ('Code Injection')

CWE-94 — Improper Control of Generation of Code ('Code Injection'): The product constructs all or part of a code segment using externally-influenced input from an upstream component, but it does not neutralize or incorrectly neutralizes special elements that could modify the syntax or behavior of the intended code segment.

[Architecture and Design] Refactor your program so that you do not have to dynamically generate code.
[Architecture and Design] Run your code in a "jail" or similar sandbox environment that enforces strict boundaries between the process and the operating system. This may effectively restrict which code can be executed by your product. Examples include the Unix chroot jail and AppArmor. In general, managed code may provide some protection. This may not be a feasible solution, and it only limits the impact to the operating system; the rest of your application may still be subject to compromise. Be careful to avoid CWE-243 and other weaknesses related to jails.

Source: MITRE CWE corpus.