CVE-2026-38743
Apache Airflow ≤ 3.2.1
Raw vector
CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:L/I:N/A:NSummary
CVE-2026-38743 is a medium-severity Insufficient Granularity of Access Control (CWE-1220) vulnerability in Apache Airflow. Its CVSS base score is 4.3 (Medium).
Operationally, exploitation aligns with the MITRE ATT&CK technique Data from Information Repositories (T1213); ranked at the 28th percentile by exploit likelihood (below the median); it is not currently listed in the CISA KEV catalog.
The strongest mitigations our analysis identified map to AC-3 (Access Enforcement) and AC-4 (Information Flow Enforcement) — see the control section below for these in your framework.
EU & UK References
- 🇪🇺 ENISA EUVD: EUVD-2026-25418
Vulnerability Data
The authenticated /ui/dags endpoint did not enforce per-DAG access control on embedded Human-in-the-Loop (HITL) and TaskInstance records: a logged-in Airflow user with read access to at least one DAG could retrieve HITL prompts (including their request parameters) and full TaskInstance…
more
details for DAGs outside their authorized scope. Because HITL prompts and TaskInstance fields routinely carry operator parameters and free-form context attached to a task, the leak widens visibility of DAG-run data beyond the intended per-DAG RBAC boundary for every authenticated user. Users are recommended to upgrade to version 3.2.1 , which fixes this issue.
- CWE(s)
Related Threats
MITRE ATT&CK Enterprise TechniquesAI
Why these techniques?
The vulnerability is an RBAC bypass on the /ui/dags endpoint that directly enables unauthorized retrieval of TaskInstance and HITL data stored in Airflow's workflow repositories.
CVEs Like This One
Affected Assets
Mitigating Controls
Control response
Mitigating Controls (NIST 800-53 r5) AI
Directly enforces the per-DAG access control policy on the /ui/dags endpoint and its embedded HITL/TaskInstance records so that read access granted for one DAG cannot expose data belonging to other DAGs.
Requires that each authenticated user be granted only the minimum privileges needed for the specific DAGs they are authorized to view, eliminating the broad read capability that enabled cross-DAG leakage.
Enforces information-flow rules that prevent TaskInstance and HITL data belonging to one DAG from flowing to a user whose authorization is limited to a different DAG.
Mitigating Controls (NIST CSF 2.0) AI
Derived directly from the weakness types (CWEs) cited in the NVD entry via our AI-authored CWE→CSF cross-walk (authority under review) — links open the control.
PR.IR-01's segmentation + zero-trust rules directly enforce minimum-necessary access at network boundaries, removing most instances of overly broad policy; CWE-1220 can still exist in application-layer or intra-segment controls that this single outcome does not address.
PR.PS-06's general SDLC practices can surface granularity issues during design/review (partial prevention) but do not specifically target access-control policy breadth, so they remove only part of CWE-1220's risk.
CWE-1220 permits unauthorized access to sensitive assets but neither prevents nor impairs the delivery of adverse-event information to authorized recipients.
Mitigating Controls (ISO/IEC 27001:2022 Annex A) AI
Derived directly from the weakness types (CWEs) cited in the NVD entry via our AI-authored CWE→ISO cross-walk (authority under review) — links open the control.
Directly requires information-access restriction at the level of individual assets and operations, addressing the granularity gap.
Establishes the overarching access-control policy whose granularity directly determines whether the weakness exists.
Defines the assignment and review of access rights; insufficient granularity in those rights is the root of CWE-1220.
Specifies management of privileged access rights, where overly broad privileges are a common manifestation of the weakness.
Limits use of privileged utilities; overly coarse access rules can still allow unauthorized utility access.