StateTree asset editing: read, modify states/tasks/conditions/transitions/bindings/evaluators/global tasks/colors/state parameters/root parameters, compile and validate. Actions: - read: Full dump of a StateTree asset: state hierarchy (with description, tag, customTickRate, color), tasks, conditi...
Risk signalsHigh parameter count (38 properties) · Admin/system-level operation
Part of the Ue server.
Free to start. No card required.
AI agents may call statetree to permanently remove or destroy resources in Ue. Without a policy, an autonomous agent could delete critical data in a loop with no way to undo the damage. PolicyLayer blocks destructive tools by default and requires explicit human approval before enabling them.
Without a policy, an AI agent could call statetree in a loop, permanently destroying resources in Ue. There is no undo for destructive operations. PolicyLayer blocks this tool by default and only allows it when a human explicitly approves the action.
Destructive tools permanently remove data. Block by default. Only enable with explicit approval workflows.
{
"version": "1",
"default": "deny",
"hide": [
"statetree"
]
} See the full Ue policy for all 22 tools.
These attack patterns abuse exactly the kind of access statetree gives an agent. Each links to the full case and the policy that stops it:
Other destructive tools across the catalogue. The same approach applies to each: deny by default, or require human approval.
StateTree asset editing: read, modify states/tasks/conditions/transitions/bindings/evaluators/global tasks/colors/state parameters/root parameters, compile and validate. Actions: - read: Full dump of a StateTree asset: state hierarchy (with description, tag, customTickRate, color), tasks, conditions, transitions, evaluators, global tasks, root params, bindings. Params: assetPath - list_states: List all states with IDs and paths. Params: assetPath - add_state: Add child state. Params: assetPath, stateId? (parent GUID, omit for root), name, stateType? (State|Group|LinkedAsset|Subtree), selectionBehavior?, insertIndex? - remove_state: Remove a state by ID. Params: assetPath, stateId - set_state_property: Set a property on a state. Params: assetPath, stateId, propertyName (name|type|selectionBehavior|bEnabled|weight|linkedAsset|description|tag|customTickRate|color), value. For tag: gameplay tag string or empty to clear. For customTickRate: number in seconds or empty to disable. For color: palette color display name, GUID, or empty to clear. - clear_state_nodes: Remove all tasks/conditions/transitions from a state. Params: assetPath, stateId - add_task: Add a task to a state. Params: assetPath, stateId, structType (C++ struct name e.g. FMyStateTreeTask or an engine-shipped task like FStateTreeRunParallelStateTreesTask), instanceProperties? - add_enter_condition: Add an enter condition to a state. Params: assetPath, stateId, structType (C++ struct name), instanceProperties?, operand? (And|Or) - remove_enter_condition: Remove an enter condition by index. Params: assetPath, stateId, conditionIndex - remove_task: Remove a task by index. Params: assetPath, stateId, taskIndex - set_task_instance_property: Set a property on a task's instance data. Params: assetPath, stateId, taskIndex, propertyName, value - set_task_property: Set a property on the task's node struct (FStateTreeTaskBase-level: bConsideredForCompletion, bTaskEnabled, bShouldStateChangeOnReselect). Distinct from set_task_instance_property which targets instance data. Params: assetPath, stateId, taskIndex, propertyName, value (string-encoded e.g. 'true'/'false') - add_transition: Add a transition to a state. Params: assetPath, stateId, trigger (OnStateCompleted|OnStateSucceeded|OnStateFailed|OnTick|OnEvent; combine with | e.g. OnStateSucceeded|OnStateFailed), transitionType (GotoState|NextState|Succeeded|Failed), eventTag?, targetStateId?, targetStatePath?, priority? (Low|Normal|Medium|High|Critical), delayDuration?, bDelayTransition? - add_transition_condition: Add a condition to an existing transition. Params: assetPath, stateId, transitionIndex, structType, instanceProperties?, operand? - remove_transition: Remove a transition by index. Params: assetPath, stateId, transitionIndex - add_binding: Add a property binding. Params: assetPath, sourceStructId, sourcePath, targetStructId, targetPath - remove_binding: Remove a property binding. Params: assetPath, targetStructId, targetPath - list_bindings: List all property bindings. Params: assetPath, structId? (filter) - add_evaluator: Add an evaluator to the StateTree (tree-level). Params: assetPath, structType (must derive from FStateTreeEvaluatorBase), instanceProperties? - remove_evaluator: Remove an evaluator by node ID. Params: assetPath, nodeId - set_evaluator_instance_property: Set a property on an evaluator's instance data. Params: assetPath, nodeId, propertyName, value - set_evaluator_property: Set a property on the evaluator's node struct (FStateTreeEvaluatorBase-level). Params: assetPath, nodeId, propertyName, value - add_global_task: Add a global task to the StateTree (tree-level). Params: assetPath, structType (must derive from FStateTreeTaskBase), instanceProperties? - remove_global_task: Remove a global task by node ID. Params: assetPath, nodeId - set_global_task_instance_property: Set a property on a global task's instance data. Params: assetPath, nodeId, propertyName, value - set_global_task_property: Set a property on a global task's node struct (FStateTreeTaskBase-level). Params: assetPath, nodeId, propertyName, value - list_colors: List all color palette entries for a StateTree. Params: assetPath. Returns: colors[] with id, displayName, color. - add_color: Add a new color to the StateTree palette. Params: assetPath, displayName, color? (FLinearColor string e.g. '(R=1.0,G=0.0,B=0.0,A=1.0)') - list_state_parameters: List parameters defined on a state. Params: assetPath, stateId. Returns: parameters[] with name, id, type, value; also bFixedLayout. - add_state_parameter: Add a parameter to a state's property bag. Rejects fixed-layout (linked) states. Params: assetPath, stateId, paramName, paramType (Bool|Int32|Int64|Float|Double|Name|String|Text) - remove_state_parameter: Remove a parameter from a state's property bag by name. Rejects fixed-layout (linked) states. Params: assetPath, stateId, paramName - set_state_parameter: Set the value of an existing state parameter. On fixed-layout states, also marks the parameter as overridden. Params: assetPath, stateId, paramName, value - set_root_parameters: Define root parameters (property bag). Params: assetPath, parameters[] ({name, type}) where type is float|int32|bool|string|name|double - compile: Compile a StateTree asset. Returns success, errors[], warnings[]. Params: assetPath - validate: Validate a StateTree asset without compiling. Params: assetPath. It is categorised as a Destructive tool in the Ue MCP Server, which means it can permanently delete or destroy data. Block by default and require explicit approval.
Register the Ue MCP server in PolicyLayer and add a rule for statetree: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Ue. Nothing to install.
statetree is a Destructive tool with critical risk. Critical-risk tools should be blocked by default and only enabled with explicit human approval.
Yes. Add a rate_limit block to the statetree rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for statetree. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
statetree is provided by the Ue MCP server (ue-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 22 Ue tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
4,600+ MCP servers and 31,000+ tools scanned and risk-classified.