PYSEC-2026-3714
Vulnerability from pysec - Published: 2026-08-12 11:17 - Updated: 2026-08-19 13:09
VLAI
Details
The Google Cloud Secret Manager secrets backend in Apache Airflow's Google provider never applied the team scope when resolving Connections and Variables: the caller's team_name was accepted by the backend but dropped at the internal call boundary, so every lookup resolved against the team-agnostic secret name. In a deployment running multi-team mode with this backend, a task or Dag belonging to one team resolved another team's Connection or Variable, obtaining its credentials in full. No unusual configuration is required beyond enabling multi-team mode and using this backend. Users are advised to upgrade to apache-airflow-providers-google 22.3.0 or later, which builds and applies the team-scoped secret name.
Severity
6.5 (Medium)
Impacted products
| Name | purl | apache-airflow-providers-google | pkg:pypi/apache-airflow-providers-google |
|---|
Aliases
{
"affected": [
{
"ecosystem_specific": {},
"package": {
"ecosystem": "PyPI",
"name": "apache-airflow-providers-google",
"purl": "pkg:pypi/apache-airflow-providers-google"
},
"ranges": [
{
"events": [
{
"introduced": "0"
},
{
"fixed": "22.3.0"
}
],
"type": "ECOSYSTEM"
}
],
"versions": [
"1.0.0",
"1.0.0b1",
"1.0.0b2",
"1.0.0rc1",
"10.0.0",
"10.0.0rc1",
"10.1.0",
"10.1.0rc1",
"10.1.0rc2",
"10.1.1",
"10.1.1rc1",
"10.10.0",
"10.10.0rc1",
"10.10.1",
"10.10.1rc1",
"10.11.0",
"10.11.0rc1",
"10.11.1",
"10.11.1rc1",
"10.12.0",
"10.12.0rc1",
"10.13.0",
"10.13.0rc1",
"10.13.0rc2",
"10.13.0rc3",
"10.13.1",
"10.13.1rc1",
"10.14.0",
"10.14.0rc1",
"10.14.0rc2",
"10.15.0",
"10.15.0rc1",
"10.16.0",
"10.16.0rc1",
"10.17.0",
"10.17.0rc1",
"10.18.0",
"10.18.0rc1",
"10.18.0rc2",
"10.19.0",
"10.19.0rc1",
"10.2.0",
"10.2.0rc1",
"10.20.0",
"10.20.0rc1",
"10.21.0",
"10.21.0rc1",
"10.21.1",
"10.21.1rc1",
"10.21.1rc2",
"10.22.0",
"10.22.0rc1",
"10.23.0",
"10.23.0rc1",
"10.24.0",
"10.24.0rc1",
"10.25.0",
"10.25.0rc1",
"10.26.0",
"10.26.0rc1",
"10.3.0",
"10.3.0rc1",
"10.3.0rc2",
"10.4.0",
"10.4.0rc1",
"10.5.0",
"10.5.0rc1",
"10.6.0",
"10.6.0rc1",
"10.6.0rc2",
"10.6.0rc3",
"10.7.0",
"10.7.0rc1",
"10.8.0",
"10.8.0rc1",
"10.9.0",
"10.9.0rc1",
"11.0.0",
"11.0.0rc1",
"12.0.0",
"12.0.0rc1",
"12.0.0rc2",
"13.0.0",
"14.0.0",
"14.0.0rc1",
"14.1.0",
"14.1.0rc1",
"15.0.0",
"15.0.0rc1",
"15.0.1",
"15.0.1rc1",
"15.1.0",
"15.1.0rc1",
"16.0.0",
"16.0.0rc1",
"16.1.0",
"16.1.0rc1",
"17.0.0",
"17.0.0rc1",
"17.1.0",
"17.1.0rc1",
"17.2.0",
"17.2.0rc1",
"18.0.0",
"18.0.0rc1",
"18.1.0",
"18.1.0rc1",
"19.0.0",
"19.0.0rc1",
"19.1.0",
"19.1.0rc1",
"19.1.0rc2",
"19.2.0",
"19.2.0rc1",
"19.3.0",
"19.3.0rc1",
"19.4.0",
"19.4.0rc1",
"19.4.0rc2",
"19.5.0",
"19.5.0rc1",
"2.0.0",
"2.0.0rc1",
"2.1.0",
"2.1.0rc1",
"2.2.0",
"2.2.0rc1",
"20.0.0",
"20.0.0rc1",
"20.0.0rc2",
"21.0.0",
"21.0.0rc1",
"21.0.0rc2",
"21.1.0",
"21.1.0rc1",
"21.2.0",
"21.2.0rc1",
"21.3.0",
"21.3.0rc1",
"22.0.0",
"22.0.0rc1",
"22.1.0",
"22.1.0rc1",
"22.2.0",
"22.2.0rc1",
"22.2.1",
"22.2.1rc1",
"22.2.2",
"22.2.2rc1",
"22.3.0rc1",
"22.3.0rc2",
"22.3.0rc3",
"22.3.0rc4",
"3.0.0",
"3.0.0rc1",
"4.0.0",
"4.0.0rc1",
"4.0.0rc2",
"4.1.0rc1",
"5.0.0",
"5.0.0rc2",
"5.1.0",
"5.1.0rc1",
"6.0.0",
"6.0.0rc1",
"6.1.0",
"6.1.0rc1",
"6.2.0",
"6.2.0rc1",
"6.3.0",
"6.3.0rc1",
"6.4.0",
"6.4.0rc1",
"6.4.0rc2",
"6.4.0rc3",
"6.5.0",
"6.5.0rc1",
"6.6.0",
"6.6.0rc1",
"6.7.0",
"6.7.0rc1",
"6.8.0",
"6.8.0rc1",
"7.0.0",
"7.0.0rc1",
"8.0.0",
"8.0.0rc1",
"8.0.0rc2",
"8.1.0",
"8.1.0rc1",
"8.10.0",
"8.10.0rc1",
"8.11.0",
"8.11.0rc1",
"8.12.0",
"8.12.0rc1",
"8.2.0",
"8.2.0rc1",
"8.3.0",
"8.3.0rc1",
"8.3.0rc2",
"8.3.0rc3",
"8.4.0",
"8.4.0rc1",
"8.4.0rc2",
"8.5.0",
"8.5.0rc1",
"8.6.0",
"8.6.0rc2",
"8.6.0rc3",
"8.7.0",
"8.7.0rc1",
"8.7.0rc2",
"8.8.0",
"8.8.0rc1",
"8.9.0",
"8.9.0rc1",
"9.0.0",
"9.0.0rc1",
"9.0.0rc2"
]
}
],
"aliases": [
"CVE-2026-68868"
],
"details": "The Google Cloud Secret Manager secrets backend in Apache Airflow\u0027s Google provider never applied the team scope when resolving Connections and Variables: the caller\u0027s `team_name` was accepted by the backend but dropped at the internal call boundary, so every lookup resolved against the team-agnostic secret name. In a deployment running multi-team mode with this backend, a task or Dag belonging to one team resolved another team\u0027s Connection or Variable, obtaining its credentials in full. No unusual configuration is required beyond enabling multi-team mode and using this backend. Users are advised to upgrade to apache-airflow-providers-google 22.3.0 or later, which builds and applies the team-scoped secret name.",
"id": "PYSEC-2026-3714",
"modified": "2026-08-19T13:09:02.511247Z",
"published": "2026-08-12T11:17:10.063Z",
"references": [
{
"type": "ADVISORY",
"url": "http://www.openwall.com/lists/oss-security/2026/08/12/3"
},
{
"type": "ADVISORY",
"url": "https://lists.apache.org/thread/03h5y0fmqlh0yf055zlocxh591ozx69x"
},
{
"type": "FIX",
"url": "https://github.com/apache/airflow/pull/70869"
}
],
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:N/A:N",
"type": "CVSS_V3"
}
]
}
Loading…
Loading…
Experimental. This forecast is provided for visualization only and may change without notice. Do not use it for operational decisions.
Forecast uses a logistic model when the trend is rising, or an exponential decay model when the trend is falling. Fitted via linearized least squares.
Sightings
| Author | Source | Type | Date | Other |
|---|
Nomenclature
- Seen: The vulnerability was mentioned, discussed, or observed by the user.
- Confirmed: The vulnerability has been validated from an analyst's perspective.
- Published Proof of Concept: A public proof of concept is available for this vulnerability.
- Exploited: The vulnerability was observed as exploited by the user who reported the sighting.
- Patched: The vulnerability was observed as successfully patched by the user who reported the sighting.
- Not exploited: The vulnerability was not observed as exploited by the user who reported the sighting.
- Not confirmed: The user expressed doubt about the validity of the vulnerability.
- Not patched: The vulnerability was not observed as successfully patched by the user who reported the sighting.
Loading…
The MITRE ATT&CK techniques below are AI-generated suggestions, inferred from the description of the
vulnerability by the CIRCL/vulnerability-attack-technique-classification-roberta-base
model, served locally by ML-Gateway.
They have not been verified by an analyst and are provided for guidance only.
The approach is described in our paper Mapping CVEs to MITRE ATT&CK Techniques: A Curated Gold-Set Classifier and the Limits of LLM-Assisted Label Expansion.
The approach is described in our paper Mapping CVEs to MITRE ATT&CK Techniques: A Curated Gold-Set Classifier and the Limits of LLM-Assisted Label Expansion.
Loading…
Loading…