---
id: "CVE-2026-96285"
page: "https://infrastack.ai/live/CVE-2026-96285"
kind: "vulnerability"
aliases: ["CVE-2026-96285"]
title: "Apache Airflow: incomplete fix lets a DAG author run code in the scheduler"
severity: "high"
cvss: null
epss: null
exploited: false
status: "open"
category: "runtime"
products: []
ecosystems: []
tags: ["apache-airflow", "kubernetes", "python", "runtime", "self-hosted"]
not_affected_if: ["dag-authors-trusted", "versions-before-3.3.0"]
published_at: "2026-11-09T10:17:16+00:00"
first_seen_at: "2026-11-09T10:25:50+00:00"
updated_at: "2026-11-09T10:26:30+00:00"
sources: [{"source": "nvd", "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-96285", "first_seen_at": "2026-11-09T10:25:50+00:00"}]
exploit_sources: []
---

# Apache Airflow: incomplete fix lets a DAG author run code in the scheduler

Link for people: https://infrastack.ai/live/CVE-2026-96285 (this .md is for agents)

Airflow 3.3.1's fix for CVE-2026-67260 checks the decoded next_kwargs value only after the whole payload is deserialized, so constructors of allow-listed classes still run first. A DAG author who controls a task's next_kwargs can trigger that constructor execution inside the scheduler or API server.

_Summary, check and fix written by a model from the sources; labelled as such._

## How to check

- Check the installed version with `airflow version`; versions from `3.3.0` up to `3.3.1` are exposed to this issue.
- Confirm whether the deployment was upgraded to `3.3.1` for CVE-2026-67260, since that upgrade does not fully fix this issue.

## Fix

- apache-airflow: upgrade to `3.4.0` or later

## Timeline

- 2026-11-09T10:25:50+00:00 nvd https://nvd.nist.gov/vuln/detail/CVE-2026-96285

## From the sources

Apache Airflow 3.3.1 shipped a fix for CVE-2026-67260 that checks the scheduler's decoded `next_kwargs` value is a dictionary, but performs that check only after deserializing the whole stored payload. Deserialization reconstructs nested objects of classes permitted by the deserialization allow-list — which by default includes Airflow's own classes — and runs their constructors inside the scheduler process before the check can reject the value. A Dag author, who controls a task's `next_kwargs` through the task execution API, can therefore still trigger constructor execution in the scheduler when the `awaiting_input` timeout sweep processes the task, and in the API server when the task instance is served to a worker using an older Execution API version; no non-default configuration is required. Versions before 3.3.0 are not affected. This is an incomplete fix for CVE-2026-67260: deployments that upgraded to 3.3.1 for that issue must also upgrade to apache-airflow 3.4.0 or later.

---

_Free to read, with fair use of about 60 pages a minute. For higher limits or API access: sales@infrastack.ai_