CVE-2026-79696: Remote Code Execution in Google ADK for Python via Incomplete Standard Library Denylist
A Code Injection vulnerability in adk web in Google Cloud Agent Development Kit (ADK) for Python versions 2.0.0 through 2.6.0 on Python (OSS), Cloud Run, and GKE environments where pytest is installed allows an unauthenticated remote attacker to execute arbitrary code using a crafted test session replay.
Security readout for executives and security teams
Plain-English summary
A remotely reachable ADK web deployment may let an unauthenticated attacker run arbitrary code when it uses Google ADK for Python 2.0.0–2.6.0 and pytest is installed. Successful exploitation could expose or alter data, disrupt services, and affect connected systems. The issue is rated CVSS 10.0.
Executive priority
Treat this as an immediate remediation priority for internet-facing or sensitive deployments. Identify exposed instances, restrict access, remove unnecessary pytest dependencies, and follow Google’s v2.7.0 guidance. Begin incident review where affected configurations were externally reachable, while avoiding claims of compromise without evidence.
Technical view
CVE-2026-79696 is a CWE-184 code-injection flaw caused by an incomplete standard-library denylist. A crafted test-session replay can trigger arbitrary code execution in affected ADK web deployments without authentication or user interaction. The stated scope covers Python OSS, Cloud Run, and GKE environments where pytest is installed.
Likely exposure
Highest exposure exists where ADK web versions 2.0.0–2.6.0 are remotely accessible and pytest is present in the runtime. Deployments without pytest, outside the stated version range, or without exposed ADK web functionality are not established as affected by the supplied evidence.
Exploitation context
The supplied record describes network-accessible, low-complexity, unauthenticated exploitation with complete confidentiality, integrity, and availability impact. It does not report known active exploitation, and the CVE is not listed as KEV in the bundle. No reliable exploitation prevalence is provided.
Researcher notes
The core condition combines an incomplete denylist with pytest availability and crafted test-session replay. The narrative states versions 2.0.0–2.6.0, while the structured affected entry lists only 2.0.0 with default status unaffected. Validate exact version applicability using the linked release and commit before final fleet scoping.
Mitigation direction
Inventory ADK web deployments and identify versions 2.0.0 through 2.6.0.
Remove pytest from production runtime images where it is unnecessary.
Restrict external access to ADK web and test-session functionality pending remediation.
Review Google ADK v2.7.0 guidance and upgrade according to vendor instructions.
Rotate exposed credentials if compromise indicators are discovered.
Validation and detection
Confirm the installed Google ADK for Python version in every environment.
Inspect runtime dependencies and container images for pytest.
Verify whether ADK web or test-session replay interfaces are externally reachable.
Review logs for unexpected test-session activity, process creation, or outbound connections.
Confirm remediation against the referenced v2.7.0 release and security commit.
Generated from the cited source records. This long-tail analysis has not been individually reviewed by a named human.
Potential ATT&CK relevance
Conservative CVE-to-ATT&CK context
These mappings and lookup hints may be relevant to the vulnerability behavior, CWE, affected product, or exposure path. Glexia-inferred context is not an official MITRE, ATT&CK, CWE, or CVE Program mapping.
ATT&CK lookup starting points
Use these exact CWE pages and searches to review the Glexia ATT&CK library from this CVE's weakness and description context.
cwe · low confidence lookup
CWE-184: Exact CWE lookup
Use the exact CWE identifier as the starting point before reviewing related ATT&CK behavior. Open the exact CWE lookup page first, then review the ATT&CK searches from that MITRE weakness context. This is a Glexia lookup hint, not an official ATT&CK mapping.
The CVE wording references code or command execution, so execution technique review may help defensive triage. This is a Glexia inferred lookup path, not an official MITRE, ATT&CK, or CVE Program mapping.
These fields come from the CVE record and ADP containers, not from Glexia's Take. They preserve time-varying source decisions such as CISA SSVC, KEV status, CVSS metrics, and provider references.
1CVSS vectors
3Timeline events
1ADP providers
3Source links
SSVC decision data
CISA-ADPCISA Coordinator
Timestamp
Version
2.0.3
Exploitation: noneAutomatable: yesTechnical Impact: total
CVSS vector scores
1 official score
We collect every scored CVSS vector available in the official CNA and ADP containers. When more than one version is present, the table keeps the source vectors side by side instead of collapsing them into the highest score.
CWE links open Glexia weakness intelligence pages with official CWE context, developer remediation guidance, and related CVE mappings.
CWE-184 · source CWE mapping
Incomplete List of Disallowed Inputs
Incomplete List of Disallowed Inputs represents a recurring weakness pattern that can create exploitable paths when design, validation, or implementation controls are missing.