Derivation rules
2,433 declarative match → where → emit rules
(1,787 derived, 646 explicit).
Filter by cloud, emitted edge, node type, or text.
Signing with a key that anchors an auth system forges the trusted identity.
Admin of a SIGN_VERIFY key can self-grant kms:Sign and forge signatures.
Swap an existing function's execution role via UpdateFunctionConfiguration.
Create a new Lambda bound to any role passable to lambda, then invoke it.
Cross-account principal that can run code in a function enters this account.
Executing as a more-privileged Lambda role is privilege escalation.
Replace function code, then run it, to execute as the function's role.
Inject code via env/layer/handler config change, then run, as the role.
A function leaks its execution-role credentials to any code it runs.
Code running as the function can fetch the execution-role token.
Function URL with AuthType NONE + public resource policy is internet-invocable.
An LF admin (or GRANT OPTION holder) granting Lake Formation SELECT / INSERT / DATA_LOCATION_ACCESS to a principal ARN in a foreign AWS account creates cross- account data-lake trust: the external principal can query this account's governed data via their own analytics services.
A principal who can read LF-governed table data (via Athena/Glue/EMR/Redshift Spectrum) and can write query output to an external S3 destination (UNLOAD / Glue job target / EMR write) can exfiltrate the governed data out of the account's LF-controlled boundary.
An effective Lake Formation SELECT grant on a governed table, combined with IAM permission to call an analytics service (Athena, Glue, EMR, Redshift Spectrum, SageMaker), enables reading the underlying S3 data via LF-vended credentials (lakeformation:GetDataAccess called internally by the service).
lakeformation:GrantPermissions lets a principal grant LF data-lake permissions (SELECT / INSERT / DATA_LOCATION_ACCESS / etc.) to ANY principal, including themselves. This sub-rule captures the self-grant case (principal -> principal), the privilege escalation primitive in Lake Formation.
lakeformation:PutDataLakeSettings lets a principal add themselves to the Lake Formation DataLakeAdmins list, after which ALL LF permission checks are bypassed for that principal - they can grant any data permission to any principal, read any governed table, and modify any registered location. Emits CanGrantPermission (nature:both) to feed into can-control linchpin for escalation roll-up.
lakeformation:UpdateResource + iam:PassRole to lakeformation.amazonaws.com lets a principal swap the IAM role Lake Formation uses to vend temporary S3 credentials for a registered location, redirecting all subsequent data access through a role of the attacker's choosing.
A Lightsail bucket access key authenticates to the bucket's data plane.
Standing up an instance from someone else's snapshot exposes the snapshot's on-disk contents within the account.
Arbitrary container deployment runs as the Lightsail-managed service execution context.
OS access to a Lightsail instance -> read IMDS -> act as the associated role.
A Lightsail instance with an associated IAM role exposes that role's session via IMDS.
Pushing an attacker image into the Lightsail private registry supplies the code a deployment will run.
Granting resource access to a bucket makes its objects readable by an instance/account principal.
Disabling Macie (macie2:DisableMacie) permanently disables the service, deletes all configurations and findings, and removes sensitive-data classification from the account - a weaken-defenses / cover-tracks / destructive primitive.
Reading Macie findings (macie2:GetFindings) reveals the precise S3 objects and locations where sensitive data - credentials, API keys, PII - was detected. This is a recon / targeting primitive.