{"description":"Trending threats, MITRE ATT\u0026CK coverage, and detection metadata. Fed continuously.","favicon":"https://feed.craftedsignal.io/favicon-32x32.png","feed_url":"https://feed.craftedsignal.io/tags/trust-policy/feed.json","home_page_url":"https://feed.craftedsignal.io/","icon":"https://feed.craftedsignal.io/apple-touch-icon.png","items":[{"_cs_actors":[],"_cs_cpes":[],"_cs_cves":[],"_cs_exploited":false,"_cs_has_poc":false,"_cs_poc_references":[],"_cs_products":["@hulumi/policies (\u003c 1.3.2)","GitHub Actions"],"_cs_severities":["critical"],"_cs_tags":["github","aws","iam","oidc","trust-policy"],"_cs_type":"advisory","_cs_vendors":["GitHub","AWS"],"content_html":"\u003cp\u003e@hulumi/policies versions before 1.3.2 contain a critical vulnerability related to AWS IAM trust policies. The vulnerability arises because the software's trust policy inspector only checks for exact AWS IAM \u003ccode\u003eStringLike\u003c/code\u003e and \u003ccode\u003eStringEquals\u003c/code\u003e condition operator keys. Attackers can bypass the intended security checks by using set-qualified operators such as \u003ccode\u003eForAnyValue:StringLike\u003c/code\u003e. These set-qualified operators can obscure wildcard GitHub Actions OIDC \u003ccode\u003esub\u003c/code\u003e conditions, which are supposed to be validated by the mandatory guardrail. This bypass can potentially lead to unauthorized access to AWS resources. The vulnerability is patched in version 1.3.2, which introduces proper evaluation of set-qualified string operators and rejects unsafe GitHub OIDC \u003ccode\u003esub\u003c/code\u003e conditions.\u003c/p\u003e\n\u003ch2 id=\"attack-chain\"\u003eAttack Chain\u003c/h2\u003e\n\u003col\u003e\n\u003cli\u003eAn attacker gains control of a GitHub Actions workflow.\u003c/li\u003e\n\u003cli\u003eThe attacker modifies the workflow to request an OIDC token from the GitHub Actions environment.\u003c/li\u003e\n\u003cli\u003eThe attacker crafts a malicious AWS IAM role trust policy using a set-qualified condition operator (e.g., \u003ccode\u003eForAnyValue:StringLike\u003c/code\u003e) to bypass the guardrail implemented by @hulumi/policies. This crafted policy obscures the intended \u003ccode\u003eStringLike\u003c/code\u003e or \u003ccode\u003eStringEquals\u003c/code\u003e checks on the \u003ccode\u003esub\u003c/code\u003e claim of the OIDC token.\u003c/li\u003e\n\u003cli\u003eThe attacker's workflow uses the AWS CLI or SDK with the crafted trust policy to assume the IAM role.\u003c/li\u003e\n\u003cli\u003eDue to the bypassed validation, the attacker's request succeeds, and the workflow obtains temporary AWS credentials associated with the IAM role.\u003c/li\u003e\n\u003cli\u003eThe attacker leverages the acquired AWS credentials to perform unauthorized actions within the AWS environment.\u003c/li\u003e\n\u003cli\u003eThe attacker may exfiltrate sensitive data, modify infrastructure configurations, or deploy malicious resources, depending on the permissions granted to the assumed IAM role.\u003c/li\u003e\n\u003c/ol\u003e\n\u003ch2 id=\"impact\"\u003eImpact\u003c/h2\u003e\n\u003cp\u003eSuccessful exploitation of this vulnerability allows attackers to bypass the intended OIDC-based access controls for AWS resources. This can lead to unauthorized access to sensitive data, infrastructure modifications, or deployment of malicious resources within the AWS environment. The severity is critical because it directly undermines the trust relationship between GitHub Actions and AWS, potentially affecting any organization using @hulumi/policies for managing these integrations.\u003c/p\u003e\n\u003ch2 id=\"recommendation\"\u003eRecommendation\u003c/h2\u003e\n\u003cul\u003e\n\u003cli\u003eUpgrade \u003ccode\u003e@hulumi/policies\u003c/code\u003e to version 1.3.2 or later to incorporate the patch that properly evaluates set-qualified string operators.\u003c/li\u003e\n\u003cli\u003eManually review existing AWS IAM role trust policies that rely on @hulumi/policies to identify and remediate any instances where set-qualified operators are used in a way that could bypass the intended validation of GitHub OIDC \u003ccode\u003esub\u003c/code\u003e conditions.\u003c/li\u003e\n\u003cli\u003eImplement monitoring and alerting on AWS IAM role assumption events to detect any suspicious activity that may indicate exploitation attempts, using the principle of least privilege to limit the impact of compromised roles.\u003c/li\u003e\n\u003cli\u003eConsider using tools and services that provide continuous monitoring of IAM configurations and compliance with security best practices.\u003c/li\u003e\n\u003c/ul\u003e\n","date_modified":"2026-05-21T20:48:15Z","date_published":"2026-05-21T20:48:15Z","id":"https://feed.craftedsignal.io/briefs/2026-05-21-hulumi-policies-bypass/","summary":"@hulumi/policies versions before 1.3.2 are vulnerable to a critical trust policy bypass where set-qualified operators such as ForAnyValue:StringLike could hide wildcard GitHub Actions OIDC sub conditions from the mandatory guardrail, potentially allowing unauthorized access to AWS resources.","title":"@hulumi/policies GitHub OIDC Trust Policy Bypass via AWS Set-Qualified Condition Operators","url":"https://feed.craftedsignal.io/briefs/2026-05-21-hulumi-policies-bypass/"}],"language":"en","title":"CraftedSignal Threat Feed - Trust-Policy","version":"https://jsonfeed.org/version/1.1"}