Splunk SOAR versions below 8.8.0 contain a vulnerability where an operator of one configured external identity provider can authenticate as an existing user associated with a different configured identity provider. This occurs because Splunk SOAR authenticates accounts by username without verifying the identity provider ownership, allowing potential privilege escalation if the targeted user has the Administrator role. The vulnerability requires multiple configured identity providers with distinct trust domains and overlapping usernames.
Splunk App for Stream versions below 8.1.7 contain an XML External Entity (XXE) vulnerability in the vocabulary upload REST API endpoint accessible on the management port 8089. This flaw allows an unauthenticated attacker who can reach the API to upload a crafted XML vocabulary file that can read local files accessible to the Splunk process or send outbound requests. The vulnerability arises because the XML parser does not enforce an application-level policy to block external entity resolution.
In Splunk SOAR versions below 8.8.0, an authenticated user with edit permission on a workbook can configure a service-level agreement (SLA) only at the workbook phase or task level. This causes the system to immediately flag the task as overdue and not display the configured deadline in the work queue. The flaw arises because the absence of a phase-level or task-level SLA is treated as immediately due, potentially obscuring real deadline breaches.
In Splunk App for Stream versions below 8.1.7, an unauthenticated user can cause the application to reset its in-memory indexer cache. This occurs because the Splunk Web handler processes cache-reset requests without requiring user authentication. The vulnerability has a medium severity rating with a CVSS score of 5.3.
Splunk App for Stream versions below 8.1.7 contain a vulnerability where a user with read access to splunkd_access.log or the _internal Splunk index can recover the Stream forwarder authentication key. This key is exposed because it is sent in an internal request URL that is logged in splunkd_access.log. An attacker with this key could impersonate a trusted Stream forwarder.
Splunk SOAR versions below 8.8.0 contain a vulnerability where a user with permission to author and run playbooks can add messages to the log of a different playbook run, potentially causing that run to fail. This includes runs outside the labels the user is authorized to access. The issue arises because the product does not verify that playbook API requests use the trusted runtime context for the current playbook run.
Splunk AI Assistant versions below 2.0.0 contain an authorization bypass vulnerability that allows users without 'admin' or 'power' roles to access limited information about indexes and data sources beyond their authorized scope. This occurs because the assistant does not fully enforce authorization controls on all retrieved information.
Splunk SOAR versions below 8.8.0 contain a vulnerability where an authenticated user with permission to edit and run playbooks can view child playbook status and loop-state metadata for playbook runs outside their authorized scope. This occurs because the playbook status API does not properly bind the requested run to the current execution context before returning metadata.
In Splunk SOAR versions below 8.8.0, a vulnerability exists where a user with permissions to edit and run a playbook on a permitted label can cause the system to return custom-function status and callback state from a different playbook run. This occurs due to insufficient binding between the custom-function-status request and the initiating playbook run.
Splunk SOAR versions below 8.8.0 contain a vulnerability where an authenticated user with permissions to edit and run playbooks on a container label can associate a custom-function callback with another playbook run. This flaw allows the callback to modify the state of a container beyond the user's authorized scope due to improper binding of playbook API requests to the trusted playbook run and container context.
Splunk SOAR versions below 8.8.0 contain a vulnerability where an authenticated user with permission to edit and run playbooks can access or modify persisted playbook run data beyond their authorized scope. This includes sensitive data such as playbook inputs, outputs, secrets, and intermediate results. The flaw arises because the playbook automation APIs do not enforce data operation restrictions tied to the active playbook run. This can lead to unauthorized data disclosure and corruption of playbook run results.
Splunk SOAR versions below 8.8.0 contain a vulnerability where a user with access to the configured load balancer account can replace a load balancer helper. This helper is later executed by Splunk SOAR with root privileges during cluster lifecycle operations. Because Splunk SOAR does not verify the integrity or ownership of this helper before execution, an attacker could achieve arbitrary code execution with root privileges if a cluster lifecycle operation is performed by an administrator.
Splunk SOAR versions below 8.8.0 contain a vulnerability where an authenticated user with permission to create and run playbooks can start child playbooks on containers outside their authorized label or repository scope. This occurs because the identity used to authorize the child playbook run is not bound to the parent playbook execution context, potentially exposing container data and affecting automation beyond the user's scope.
Splunk SOAR versions below 8.8.0 contain a vulnerability where users with roles granting Playbooks view, edit, execute, and delete permissions can cause cross-repository playbook references to execute attacker-controlled code. This occurs because the system allows binding of existing cross-repository playbook references to a different source control repository if a deleted repository name is reused.
Splunk SOAR versions below 8.8.0 contain a vulnerability where an authenticated user with permission to author and run playbooks can access action data beyond their authorized scope and modify the status and message of other playbook runs. This occurs due to insufficient authorization checks in the playbook API.
In Splunk SOAR versions below 8.8.0, users with roles that allow editing and running playbooks could exploit a vulnerability to execute non-read-only actions from a Safe Mode playbook, causing unauthorized state changes. This occurs because the system does not verify that the playbook run context in an action request matches the server-provided context.
Splunk SOAR versions prior to 8.8.0 contain a vulnerability where an authenticated user with permissions to edit and run playbooks can disrupt automation batches for other events. This occurs because the playbook API requests are not properly bound to the server-provided execution context, allowing interference with concurrent automation processes.
Splunk SOAR versions below 8.8.0 contain a vulnerability where a user with permission to edit and run playbooks for a specific container label can manipulate playbook automation for containers they are not authorized to view. This occurs because the Playbook Automation API does not verify that a supplied playbook run identifier is bound to the correct execution context. Exploitation requires an existing running playbook on a container label outside the user's permitted view.
In Splunk SOAR versions below 8.8.0, an authenticated user with the system_settings:view permission but without assets:view permission could access sensitive asset and application inventory data, tenant identifiers, connector run messages, and stored configuration snapshots via the ingestion status REST API. This occurs because the API enforces system_settings:view permission but does not enforce assets:view permission or tenant restrictions when returning asset-derived ingestion records.
In Splunk SOAR versions below 8.8.0, an authenticated user with permissions to view and edit containers can import a container archive that causes artifacts to appear linked to cases or notes incorrectly. This occurs because the import process accepts artifact workflow state from the archive rather than deriving it from actual case associations and notes. This discrepancy can lead REST API clients and filters to use inaccurate artifact states.