Skip to content
Server Security Center
KEVcritical9.3

CVE-2026-33017: Langflow Code Injection

Langflow · listed in the CISA catalog of actively exploited vulnerabilities since 2026-03-25.

CVSS v4.0 (CVE record)
9.3
critical
EPSS (exploit probability, 30 days)
97%
higher than 99.8% of all CVEs
Added to CISA KEV
2026-03-25
US federal due date 2026-04-08
Automatable (CISA SSVC)
yes
technical impact: total

CVE-2026-33017 explained

CVE-2026-33017 is a code injection flaw in Langflow, a tool for building and deploying AI-powered agents and workflows. A public flow-building endpoint accepts attacker-supplied flow data and runs embedded Python code without a sandbox. This allows unauthenticated remote code execution, and CISA lists it as known to be exploited.

Impact

A remote attacker without authentication can run arbitrary Python code on the Langflow server. The endpoint for building public flows accepts attacker-controlled flow data, and code in node definitions is passed to exec() without sandboxing. CISA rates the flaw as automatable, meaning attackers can exploit it at scale, with total technical impact, meaning full control of the affected component. CISA's catalog confirms attackers are using it.

Who is affected

The CVE record lists langflow-ai langflow versions < 1.9.0 as affected. The record's structured version data names no fixed version. However, the description states that the issue has been fixed in version 1.9.0. Confirm the fixed release in the vendor advisory before upgrading. The flaw is distinct from CVE-2025-3248, which concerned a different endpoint.

Written with AI from the data on this page and checked against it automatically.

Official description

Langflow is a tool for building and deploying AI-powered agents and workflows. In versions prior to 1.9.0, the POST /api/v1/build_public_tmp/{flow_id}/flow endpoint allows building public flows without requiring authentication. When the optional data parameter is supplied, the endpoint uses attacker-controlled flow data (containing arbitrary Python code in node definitions) instead of the stored flow data from the database.

This code is passed to exec() with zero sandboxing, resulting in unauthenticated remote code execution. This is distinct from CVE-2025-3248, which fixed /api/v1/validate/code by adding authentication. The build_public_tmp endpoint is designed to be unauthenticated (for public flows) but incorrectly accepts attacker-supplied flow data containing arbitrary executable code.

This issue has been fixed in version 1.9.0.

CISA required action: Apply mitigations per vendor instructions, follow applicable BOD 22-01 guidance for cloud services, or discontinue use of the product if mitigations are unavailable.

CISA notes

Affected and fixed versions

From the CVE record published by the vendor or CVE numbering authority.

  • langflow-ai langflow
    Affected
    • < 1.9.0
    Fixed in
    not stated

How to fix it

  • The CVE record names no fixed version – check the vendor advisory and CISA's required action for updates and mitigations.
  • Vendor advisory: github.com/langflow-ai/langflow/security/advisories/GHSA-…
  • CISA's deadline for US federal agencies: 2026-04-08 (passed) – a useful benchmark for everyone else, too.

What to do

  1. Upgrade Langflow to 1.9.0, which the description names as the fixed version, after confirming it in the vendor advisory.
  2. Apply any mitigations in the vendor advisory, as CISA's required action directs.
  3. Do not expose Langflow instances to the internet. Restrict access to trusted networks and users.
  4. Check Langflow servers for signs of compromise, such as unexpected processes, files or outbound connections.
  5. If mitigations are unavailable, discontinue use of the product, as CISA's required action states.

The CVE record's structured data names no fixed version, but the description says the issue is fixed in 1.9.0. Confirm this in the vendor advisory. Until you can upgrade, follow CISA's required action: apply vendor mitigations, follow BOD 22-01 guidance for cloud services, or discontinue use if mitigations are unavailable. Keep Langflow off the internet in the meantime.

Written with AI from the data on this page and checked against it automatically.

CISA's assessment (SSVC)

Exploitation
Exploited in the wild
Automatable
Yes – attackers can automate it at scale
Technical impact
Total – full control of the component or total information disclosure

Weakness: CWE-94 Improper Control of Generation of Code ('Code Injection'), CWE-95 Improper Neutralization of Directives in Dynamically Evaluated Code ('Eval Injection'), CWE-306 Missing Authentication for Critical Function

Questions about CVE-2026-33017

Is CVE-2026-33017 being exploited?
Yes. CISA added it to its Known Exploited Vulnerabilities catalog on 2026-03-25, with a US federal due date of 2026-04-08. Whether it is used in ransomware campaigns is listed as unknown.
Does an attacker need credentials?
No. The affected endpoint is designed to work without authentication for public flows. It incorrectly accepts attacker-supplied flow data containing executable code, so no login is required.
Does the fix for CVE-2025-3248 protect against this?
No. The description states this is distinct from CVE-2025-3248, which added authentication to a different endpoint. Upgrade to 1.9.0, which the description names as the fix.

Written with AI from the data on this page and checked against it automatically.

References

Data: CISA Known Exploited Vulnerabilities (public domain) · CVE Program records incl. CISA ADP enrichment · OSV.dev (CC BY 4.0) incl. Ubuntu and Debian security data · FIRST EPSS · endoflife.date. Updated automatically – facts last changed 2026-11-21. Always confirm with the vendor's or distribution's advisory before patching.