hulumi IAM Privilege Escalation via Weekly Integration Policy
First seen Sep 1, 2026 · Updated Sep 1, 2026 · CVSS 9.8
A critical privilege escalation vulnerability in hulumi (versions prior to v1.3.2) allows attackers with access to a documented IAM principal to abuse an overly permissive weekly integration policy. This enables creation of persistent, higher-privilege af-e2e-* roles in sandbox accounts, potentially leading to full account compromise.
Technical Analysis
CVE-2026-82857 stems from insufficient permission boundary enforcement on IAM role lifecycle operations scoped to af-e2e-* roles within hulumi's weekly integration workflow. An attacker holding the documented principal's credentials can create or modify roles with elevated trust policies, bypassing intended least-privilege boundaries and establishing persistent backdoor access in the sandbox account. The CVSS score of 9.8 reflects low attack complexity, no required user interaction, and high impact to confidentiality, integrity, and availability. Because hulumi is an infrastructure-as-code (Pulumi-adjacent) tool likely used in CI/CD and automated provisioning pipelines, exploitation could cascade into broader cloud environments if sandbox roles are trusted by production automation. Organizations running AI agents that orchestrate infrastructure provisioning, CI/CD pipelines, or cloud resource management via hulumi-integrated workflows are at risk of agents inadvertently executing privilege escalation paths or having their scoped credentials abused to mint persistent high-privilege roles, resulting in downstream compromise of agent-accessible cloud resources and API keys.
Affected Systems
hulumi versions prior to v1.3.2; IAM configurations using the weekly integration policy; af-e2e-* IAM roles in sandbox AWS/cloud accounts
Indicators of Compromise
- N/A - vulnerability disclosure without confirmed exploitation indicators; monitor for unexpected af-e2e-* role creation, unusual IAM PassRole/CreateRole API calls, and new trust policy modifications in sandbox accounts
Remediation Steps
- 1
Upgrade hulumi
Update hulumi to v1.3.2 or later where the IAM policy boundary restrictions are enforced.
- 2
Audit IAM Roles
Review all af-e2e-* roles in sandbox accounts for unauthorized creation, modification, or unexpected trust relationships.
- 3
Tighten Permission Boundaries
Apply explicit IAM permission boundaries and SCPs to restrict role creation and privilege escalation paths regardless of application-level controls.
- 4
Rotate Credentials
Rotate credentials for the documented principal and any associated service accounts or API keys used by automation/agent pipelines.
- 5
Enable Monitoring
Implement CloudTrail/GuardDuty alerts for anomalous IAM CreateRole, AttachRolePolicy, and AssumeRole activity tied to sandbox environments.
CVE / Advisory IDs
Industries Most Exposed
Respond to this threat
Pro subscribers get a full AI-generated incident-response playbook for this threat — detection, containment, eradication, and recovery steps — plus an unlimited AI Threat Advisor for questions about your environment.