Atlas / MCP servers / aadeshrao123 / Unreal

UnrealBLOCK

Name: Unreal
Author: aadeshrao123

mcp/aadeshrao123/unreal-4

MCP server by aadeshrao123 · ai agent · ai tools · blueprints · Audited 2026-10-08

AI bridge for Unreal Engine 5. Control the editor from Claude Code, Cursor, Windsurf, and any MCP client. Community edition of CodeFizz Editor Agent, which has 900+ commands across 28 categories on UE 5.6, 5.7 and 5.8.

Verdict

BLOCK

Grade

Trust score

69 /100

Exposed tools

200 135r · 119w · 32d

Transport

stdio

License

MPL-2.0

Stars

Overview

From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.

Control the Unreal Engine 5 editor from Claude Code, Cursor, Windsurf, or any MCP client. Build materials, wire Blueprint graphs, author Niagara systems, spawn actors, edit data tables and profile performance, all by asking for it. 288 commands across 15 categories, no leaving your terminal.

[](LICENSE) [](https://www.unrealengine.com/) [](https://modelcontextprotocol.io/) [](https://codefizz.dev/?utmsource=github&utmmedium=readme&utmcampaign=communitybadge)

### This is the community edition It is stable and feature complete for what it covers. It is not abandoned, and it is not in active development either: new subsystem work now goes into [CodeFizz Editor Agent](https://codefizz.dev/?utm_source=github&utm_medium=readme&utm_campaign=community_notice), the maintained commercial build. Everything here keeps working, issues still get read, and the MIT licence is permanent. If you want Control Rig, PCG, Sequencer, Mutable, Procedural Vegetation, audio and MetaSounds, Behavior Trees and EQS, or packaged binaries for UE 5.6 / 5.7 / 5.8, that lives in Pro.

Community edition vs Pro

Read from source at commit ff42563918ecOBSERVED · 2026-10-08

Connect

Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control.

claude-code (pypi)

claude mcp add unrealmcp -- None unrealmcp==1.3.1

Exposed tools (200)

135 read · 119 write · 32 destructive. Blast radius: 32 tools can delete or overwrite — an agent that can be talked into calling a tool can be talked into calling this one.

Tool	Risk	Description
`add_blueprint_node`	write	Add a node to a Blueprint graph.
`add_component_to_blueprint`	write	Add a component to a Blueprint
`add_data_table_row`	write	Add a new row, optionally setting initial field values.
`add_event_node`	write	Add an event node (ReceiveBeginPlay, ReceiveTick, ReceiveDestroyed, etc.).
`add_function_input`	write	Add an input parameter to a Blueprint function.
`add_function_output`	write	Add an output parameter to a Blueprint function.
`add_input_action_modifier`	write	Add a modifier to an Input Action.
`add_input_action_trigger`	write	Add a trigger to an Input Action.
`add_key_mapping`	write	Add a key-to-action mapping to an Input Mapping Context.
`add_mapping_modifier`	write	Add a modifier to a specific mapping in an Input Mapping Context.
`add_mapping_trigger`	write	Add a trigger to a specific mapping in an Input Mapping Context.
`add_mass_config_trait`	write	Add a SINGLE trait to a Mass Entity Config without replacing existing traits.
`add_material_comments`	write	Add comment boxes to a material graph for organization.
`add_material_expression`	write	Add a single expression node without clearing other nodes.
`add_material_function_input`	write	Add an input pin to a Material Function.
`add_material_function_output`	write	Add an output pin to a Material Function.
`add_niagara_component`	write	Add a NiagaraComponent to an existing actor in the level.
`add_niagara_custom_hlsl_input`	write	Add a typed input pin to a scratch pad module
`add_niagara_custom_hlsl_output`	write	Add a typed output pin to a scratch pad module
`add_niagara_emitter`	write	Add an emitter to a Niagara System.
`add_niagara_event_handler`	write	Add an event handler to an emitter that responds to events from a source emitter.
`add_niagara_graph_node`	write	Create a new node inside a Niagara scratch-pad / dynamic-input / module graph.
`add_niagara_map_get_pin`	write	Add a typed output pin to the first ParameterMapGet node in a scratch pad graph.
`add_niagara_map_set_pin`	write	Add a typed input pin to the first ParameterMapSet node in a scratch pad graph.
`add_niagara_module`	write	Add a module to an emitter
`add_niagara_node_pin`	write	Generic: add a typed dynamic pin to any node in a scratch pad graph.
`add_niagara_renderer`	write	Add a renderer to an emitter.
`add_niagara_script_parameter`	write	Add an input or output parameter to a Niagara script.
`add_niagara_simulation_stage`	write	Add a simulation stage to an emitter.
`add_niagara_user_parameter`	write	Add a user parameter to a Niagara System. Supports every type the editor
`add_statetree_binding`	write	Add a property binding connecting a source node
`add_statetree_condition`	write	Add a condition to a state
`add_statetree_evaluator`	write	Add a global evaluator to the StateTree.
`add_statetree_global_task`	write	Add a global task to the StateTree.
`add_statetree_parameter`	write	Add a tree-level parameter.
`add_statetree_state`	write	Add a state to the tree. Omit parent params for a root-level state.
`add_statetree_task`	write	Add a task to a state.
`add_statetree_transition`	write	Add a transition to a state.
`add_widget`	write	Add a new widget to a Widget Blueprint
`analyze_blueprint_graph`	read	Analyze a specific graph — nodes, connections, and execution flow.
`apply_and_save_niagara_scratch_pad`	write	Apply a scratch pad module
`apply_material_to_actor`	write	Apply a material to a level actor.
`apply_material_to_blueprint`	write	Apply a material to a Blueprint component.
`apply_niagara_scratch_pad`	write	Commit a scratch pad module
`build_material_function_graph`	read	Build the internal node graph of a Material Function in one atomic call.
`build_material_graph`	read	Build a complete material node graph in one atomic operation.
`cleanup_material_function`	destructive	Remove unconnected/orphaned nodes from a Material Function.
`cleanup_material_graph`	destructive	Delete orphaned and/or dead-end nodes from a material graph.
`clear_niagara_module_input`	destructive	Reset a module input to its default — equivalent to
`compile_blueprint`	read	Compile a Blueprint.
`compile_niagara_system`	read	Compile a Niagara System and report any errors.
`compile_statetree`	read	Compile a StateTree and return errors/warnings.
`connect_blueprint_nodes`	read	Connect two nodes in a Blueprint graph.
`connect_material_expressions`	read	Connect two expression nodes using their indices.
`connect_niagara_pins`	read	Wire one node
`control_niagara_effect`	read	Control a Niagara effect actor in the level.
`create_blueprint`	write	Create a new Blueprint asset from any C++ or Blueprint parent class.
`create_blueprint_function`	write	Create a new function in a Blueprint.
`create_blueprint_variable`	write	Create a variable in a Blueprint.
`create_data_asset`	write	Create a new data asset of any UDataAsset subclass.
`create_input_action`	write	Create a new UInputAction asset.
`create_input_mapping_context`	write	Create a new UInputMappingContext asset.
`create_material`	write	Create a new Material asset.
`create_material_function`	write	Create a new Material Function asset.
`create_material_instance`	write	Create a Material Instance from a parent material.
`create_niagara_module_asset`	write	Create a standalone Niagara script asset (module / dynamic input / function).
`create_niagara_scratch_pad_module`	write	Create a new scratch pad script on a Niagara System.
`create_niagara_system`	write	Create a new Niagara System asset.
`create_statetree`	write	Create a new StateTree asset with the specified schema.
`delete_actor`	destructive	Delete an actor from the level by name.
`delete_asset`	destructive	Delete an asset or directory from the Content Browser.
`delete_blueprint_function`	destructive	Delete a function from a Blueprint.
`delete_blueprint_node`	destructive	Delete a node from a Blueprint graph by GUID.
`delete_data_table_row`	destructive	Delete a row from a Data Table by name.
`delete_material_expression`	destructive	Delete a single expression node by index. Recompiles and saves automatically.
`delete_niagara_graph_node`	destructive	Delete a node from a Niagara graph (asset + edit-copy).
`delete_niagara_scratch_pad_module`	destructive	Delete a scratch pad module from a Niagara System.
`delete_niagara_system`	destructive	Delete a Niagara System asset.
`describe_niagara_type`	read	Return the full schema of any Niagara type — primitive, enum, struct, or data interface.
`disconnect_material_expression`	read	Disconnect a specific input pin on a material expression node.
`disconnect_niagara_pins`	read	Break all connections on a specific pin of a node in a scratch pad graph.
`duplicate_asset`	read	Duplicate an asset to a new location.
`duplicate_data_table_row`	read	Copy an existing row under a new name.
`duplicate_material_expression`	read	Duplicate a node (same type and properties, offset from original).
`duplicate_niagara_emitter`	read	Duplicate an emitter within a Niagara System.
`duplicate_niagara_scratch_pad_module`	read	Duplicate an existing scratch pad module on a system.
`duplicate_widget`	read	Deep-copy a widget including children. Auto-renames on conflicts.
`execute_python`	write	Execute arbitrary Python code inside the running Unreal Editor.
`find_actors`	read	Flexible actor search across the level — combine any number of filters.
`find_actors_by_name`	read	Find actors whose name matches the given pattern.
`find_assets`	read	Search the Asset Registry by class, path, and/or name pattern.
`find_niagara_scratch_pad_usage`	read	Reverse lookup: find where a scratch pad script is referenced in the stack.
`find_references`	read	Find assets that reference or are referenced by the given asset.
`get_actor_material_info`	read	Get material slot info for a level actor.
`get_actor_properties`	read	Get property values from a live actor instance placed in the world.
`get_actor_property_metadata`	read	Inspect type/clamp/enum metadata for properties on a placed actor.
`get_actors_in_level`	read	List all actors in the current level.
`get_asset_info`	read	Get asset metadata: class, package, and key properties.
`get_asset_properties`	read	Get all editable properties of an asset.
`get_available_material_pins`	read	Query all available material output pins for a specific material.
`get_available_materials`	read	List materials that can be applied to objects.
`get_blueprint_class_defaults`	read	Get all default property values from a Blueprint
`get_blueprint_function_details`	read	Get detailed info about Blueprint functions (empty name = all functions).
`get_blueprint_material_info`	read	Get material slot info for a Blueprint
`get_blueprint_variable_details`	read	Get detailed info about Blueprint variables (empty name = all variables).
`get_data_asset_properties`	read	Read ALL property values from a data asset (no visibility filter).
`get_data_table_row`	read	Get a single row from a Data Table by its row name.
`get_data_table_rows`	read	Get all rows and their field data from a Data Table asset.
`get_data_table_schema`	read	Get column names and types defined by the Data Table
`get_expression_type_info`	read	Look up pin names and editable properties for a node type WITHOUT creating one.
`get_input_action`	read	Read all properties, triggers, and modifiers of an Input Action.
`get_input_mapping_context`	read	Read all key mappings from an Input Mapping Context.
`get_mass_config_traits`	read	Read all traits from a Mass Entity Config asset with their properties expanded.
`get_material_errors`	read	Get shader compilation errors for a material.
`get_material_expression_info`	read	Get detailed info for a single node, including all available input/output pins.
`get_material_function_info`	read	Inspect a Material Function: inputs, outputs, internal nodes.
`get_material_graph_nodes`	read	Read expression nodes in a material graph.
`get_material_info`	read	Inspect a material
`get_material_instance_parameters`	read	Get all overridable parameters from a Material Instance (scalar, vector, texture, switch).
`get_material_property_connections`	read	Query which expression node feeds each material output slot.
`get_mcp_token_stats`	read	Get accumulated per-command token usage statistics.
`get_niagara_actors`	read	Get all Niagara actors in the current level.
`get_niagara_data_interface_schema`	read	Return the full editable-property schema of a UNiagaraDataInterface subclass.
`get_niagara_emitter_attributes`	read	Get attributes defined on an emitter at a given scope.
`get_niagara_emitters`	read	Get emitters in a Niagara System.
`get_niagara_event_handlers`	read	Get all event handlers configured on an emitter.
`get_niagara_graph_nodes`	read	Introspect every node inside a Niagara graph.
`get_niagara_module_input_binding`	read	Resolve the actual binding of each module input — Default/Local/Linked/Dynamic/Data.
`get_niagara_module_inputs`	read	Get all input parameters for a specific module.
`get_niagara_module_versions`	read	Check for outdated modules in an emitter and their available versions.
`get_niagara_modules`	read	Get modules in an emitter
`get_niagara_node_info`	read	Deep inspect a single Niagara node: full pin layout, connections, type-specific detail.
`get_niagara_node_type_info`	read	Get pin/property schema for a Niagara node type or script asset.
`get_niagara_particle_stats`	read	Get live particle counts and emitter execution state from running preview.
`get_niagara_playback_range`	read	Get the current timeline playback range and frame rate settings.
`get_niagara_rapid_iteration_parameters`	read	Get rapid iteration parameters (the actual configurable values on modules).
`get_niagara_renderer_info`	read	Get detailed information about a renderer on an emitter.
`get_niagara_renderer_properties`	read	Get editable properties of a renderer with current values and valid enum values.
`get_niagara_schema_actions`	read	Return the full graph context-menu actions for a scratch pad module.
`get_niagara_script_properties`	read	Read the details-panel properties of a Niagara script.
`get_niagara_system_errors`	read	Get compilation errors, warnings, and validation issues for a Niagara System.
`get_niagara_system_info`	read	Get detailed information about a Niagara System.
`get_niagara_user_parameters`	read	Get user-exposed parameters from a Niagara System asset or a level actor
`get_property_valid_types`	read	Return valid classes/structs/enum-values for a property dropdown.
`get_selected_actors`	read	Get all currently selected actors in the editor viewport.
`get_selected_assets`	read	Get currently selected assets in the Content Browser.
`get_slot_properties`	read	Read layout slot properties (type depends on parent: CanvasPanelSlot, HorizontalBoxSlot, etc.).
`get_statetree_bindings`	read	Get property bindings with full path details. Each binding shows source/target
`get_statetree_evaluators`	read	Get all global evaluators with their properties and instance data.
`get_statetree_full_info`	read	Get complete StateTree data in one call with verbosity control.
`get_statetree_global_tasks`	read	Get all global tasks with their properties.
`get_statetree_info`	read	Get compact overview of a StateTree: schema, state count, evaluator count, etc.
`get_statetree_node`	read	Get detailed info for a single node (task/evaluator/condition) by GUID.
`get_statetree_parameters`	read	Get tree-level parameters (name, type, value).
`get_statetree_state`	read	Get detailed info for a single state: properties, tasks, conditions, transitions.
`get_statetree_states`	read	Get state hierarchy as a JSON tree. Each state shows name, type, ID, task/transition counts.
`get_statetree_transition_targets`	read	Get all valid transition targets for a StateTree.
`get_widget_properties`	read	Read all property values from a widget (supports all FProperty types).
`get_widget_tree`	read	Read the complete widget hierarchy from a Widget Blueprint
`get_world_info`	read	Get current editor level info (world name, actor count, actor list).
`health_check`	read	Check if the UE5 editor bridge is running and responsive.
`import_asset`	write	Import an external file (texture, mesh, FBX, etc.) into the Content Browser.
`import_assets_batch`	write	Import multiple files into the Content Browser in one batch.
`layout_material_expressions`	read	Auto-layout all nodes using Unreal
`link_niagara_parameter`	read	Link a module input to a Niagara parameter (user, system, emitter, or particle scope).
`list_assets`	read	List assets in a Content Browser directory.
`list_data_asset_classes`	read	List all UDataAsset subclasses currently loaded in the editor.
`list_data_assets`	read	List data assets in a content path, optionally filtered by class.
`list_input_actions`	read	List all UInputAction assets, optionally filtered by name and path.
`list_input_keys`	read	List available FKey values (keyboard, mouse, gamepad keys).
`list_input_mapping_contexts`	read	List all UInputMappingContext assets, optionally filtered.
`list_material_expression_types`	read	List available material expression node types.
`list_modifier_types`	read	List all available UInputModifier subclasses with their editable properties.
`list_niagara_available_parameters`	write	List parameters that can be bound to a Map Get / Map Set pin.
`list_niagara_data_interface_functions`	read	Enumerate member functions on a Niagara data interface class.
`list_niagara_data_interfaces`	read	List available Niagara Data Interfaces (Grid2D, Grid3D, AudioSpectrum, etc.).
`list_niagara_emitter_templates`	read	List available emitter templates for use with add_niagara_emitter.
`list_niagara_input_source_menu`	read	Reproduce the stack-UI source dropdown for a specific input.
`list_niagara_modules`	read	List available Niagara module scripts for use with add_niagara_module.
`list_niagara_node_types`	read	Enumerate every Niagara node type that can be added to a graph.
`list_niagara_ops`	read	Enumerate all valid UNiagaraNodeOp operations from FNiagaraOpInfo registry.
`list_niagara_parameter_types`	read	Live query of FNiagaraTypeRegistry — every type the editor
`list_niagara_scratch_pad_modules`	read	List all scratch pad modules on a Niagara System.
`list_niagara_script_parameters`	read	List input + output parameters of a Niagara script.
`list_niagara_systems`	read	List Niagara System assets in the project.
`list_statetree_enum_values`	read	List valid enum values for StateTree properties.
`list_statetree_node_types`	read	List available task/evaluator/condition types for the StateTree
`list_trigger_types`	write	List all available UInputTrigger subclasses with their editable properties.
`list_widget_types`	read	List available UWidget subclasses. Use filter to narrow results.
`move_material_expression`	write	Move a material expression node to a new graph position.
`move_widget`	write	Move a widget to a different parent panel.
`open_asset`	read	Open an asset in the UE5 editor (blueprint, material, data table, widget, etc.).
`performance_analyze_insight`	read	Query profiling data from a loaded .utrace trace.
`performance_start_trace`	write	Start recording a performance trace from the running Unreal Editor.
`performance_stop_trace`	write	Stop the active trace recording and optionally load it for analysis.
`read_blueprint_content`	read	Read complete Blueprint content: event graph, functions, variables, components.
`recompile_material`	destructive	Force recompile and save a material. Reports success/failure.
`remove_input_action_modifier`	destructive	Remove a modifier from an Input Action by index.
`remove_input_action_trigger`	destructive	Remove a trigger from an Input Action by index.
`remove_key_mapping`	destructive	Remove a key mapping from an Input Mapping Context.

Trust audit

BLOCKgrade D · trust 69/100 Do not install this without reading the findings. The audit found something that could harm you or your machine.

Layer	What it checks	Result
L0	Provenance & inventory	PASS
L1	Static analysis of the code	FAIL
L2	Instruction surface (what it tells the agent)	WARN
L3	Class-specific surface	WARN
L4	Behavioural (sandbox)	SKIPPED

What the source does

Filesystem: declared (3 observation(s))
Network: none-observed
Shell: none-observed
Dependencies: pinned
Secrets in source: none-found

Findings (6)

HIGHFilesystem / path · fs.credential_store · CWE-22, CWE-59

install.sh:356

target="$HOME/.aws/amazonq/mcp.json"

Why it matters. touches a credential store

HIGHFilesystem / path · fs.credential_store · CWE-22, CWE-59

install.sh:357

mkdir -p "$HOME/.aws/amazonq"

Why it matters. touches a credential store

HIGHPrivilege escalation / persistence · priv.escalate · CWE-269, CWE-250

install.sh:20

echo " Linux: sudo apt install python3 python3-pip"

Why it matters. asks for elevated privileges

MEDIUMPrompt injection · scope.undeclared_system · CWE-94, CWE-1427

system use found in code, not declared in the description

Why it matters. the description does not admit a capability the code has

Fix. declare system use in the description, or remove it

MEDIUMFilesystem / path · mcp.destructive_tools · CWE-22, CWE-59

cleanup_material_function, cleanup_material_graph, clear_niagara_module_input, delete_actor, delete_asset, delete_blueprint_function, delete_blueprint_node, delete_data_table_row, delete_material_expr

Why it matters. 32 tool(s) can delete or overwrite

Fix. prefer a read-only mode or scoped tokens; the page states the blast radius

LOWCode injection · code.dynamic_import · CWE-78, CWE-94, CWE-95

.github/workflows/ci.yml:62

importlib.import_module(mod_name)

Gates applied: no_behavioural_pass.

Audited 2026-10-08 · audit v0.4.1 · source sha ff42563918ecfull audit observations/trust-audit/mcp-server/aadeshrao123__unreal-4.json · Report an issue / request a re-scan

Audit history

Every audit this server has had. A grade with a past is a grade somebody is still checking.

Date	Source	Verdict	Grade	Score	Change
2026-10-08	`ff42563918ec`	BLOCK	D	69	first audit

Questions

What is the Unreal MCP server?

What tools does Unreal expose?

200 in total: 135 read-only, 119 that write, and 32 that can delete or overwrite (cleanup_material_function, cleanup_material_graph, clear_niagara_module_input, delete_actor, delete_asset). Every one is listed on this page with its risk.

Is Unreal safe to connect to an agent?

No — not without reading the findings first. The audit graded it D (69/100) and found 3 critical or high issues in the source. Each one is listed on this page with the file and line it is on. Separately from the audit: 32 of its tools can destroy data, so scope the token you give it to what you actually need.

What credentials does Unreal need?

No credential environment variables were found in its source, so it appears to need none.

How does Unreal run?

It speaks stdio, so it runs as a local process your client starts. It is published on npm as unrealcli at 1.2.1.

How current is this page?

The grade is for one exact copy of the source (ff42563918ec), read on 2026-10-08. The repository is watched and re-audited when it changes.

Source	aadeshrao123/unreal-mcp
npm	`unrealcli@1.2.1`
PyPI	`unrealmcp`
Transports	stdio
License	MPL-2.0
Activity	pushed 63d ago