What can you do with this server?

The unreal-api-mcp server gives AI agents accurate, version-specific Unreal Engine C++ API documentation to prevent hallucinated signatures, wrong #include paths, and deprecated API usage. * Search the Unreal Engine API: Find classes, structs, enums, functions, properties, or delegates by keyword (e.g., "character movement", "spawn actor"), with optional filtering by member type. * Get exact function signatures: Retrieve the precise C++ signature of any UE function by fully-qualified name (e.g., AActor::GetActorLocation), including parameters and return types. * Resolve #include paths: Look up the correct #include directive for any class, struct, or type (e.g., ACharacter → #include "GameFramework/Character.h"). * Get full class references: Retrieve a complete reference card for a class listing all its public members (e.g., UCharacterMovementComponent, APlayerController). * Check deprecation warnings: Verify whether an API is deprecated and get replacement hints (e.g., K2_AttachRootComponentTo → use AttachToComponent()). Coverage includes all Engine Runtime, Editor, and Developer modules; built-in plugins (Enhanced Input, Gameplay Abilities/GAS, Niagara, Common UI, Chaos, and hundreds more); and Blueprint graph internals (158 UK2Node subclasses, UEdGraphSchema_K2, KismetCompiler, etc.). It supports multiple UE versions (e.g., 5.5, 5.6, 5.7) with automatic version-specific database downloads, and works with Claude Code, Cursor, Windsurf, and any MCP-compatible tool. Third-party and marketplace plugins are not covered.

Which integrations are available for this server?

Provides accurate Unreal Engine C++ API documentation, including function signatures, include paths, class references, deprecation warnings, and keyword search, enabling AI agents to write correct UE C++ code without hallucinating APIs.

How do I use unreal-api-mcp?

1. Click on "Install Server". 2. Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state. 3. In the chat, type @ followed by the MCP server name and your instructions, e.g., "@unreal-api-mcp get include path for ACharacter" That's it! The server will respond to your query, and you can continue using it as needed. Here is a step-by-step guide with screenshots.

unreal-api-mcp

by Codeturion

Overview Schema Related Servers Score Discussions

Python

Local

unreal-api-mcp

PyPI Version PyPI Downloads MCP Registry GitHub Stars GitHub Last Commit Detect New UE Release License: MIT Python 3.10+

MCP server that gives AI agents accurate Unreal Engine C++ API documentation. Saves tokens, context, and time. Prevents hallucinated signatures, wrong #include paths, and deprecated API usage.

Works with Claude Code, Cursor, Windsurf, or any MCP-compatible AI tool. No Unreal Engine installation required. Check the supported versions. New versions are detected and built automatically every week.

Quick Start

Add to your MCP config (.mcp.json, mcp.json, or your tool's MCP settings), setting UNREAL_VERSION to match your project:

{
  "mcpServers": {
    "unreal-api": {
      "command": "uvx",
      "args": ["unreal-api-mcp"],
      "env": {
        "UNREAL_VERSION": "5.5"
      }
    }
  }
}

Set this to match your project's UE version. See supported versions for all available databases.

On first run the server downloads the correct database to ~/.unreal-api-mcp/. Patch versions (e.g. "5.7.3") fall back to the major.minor database (e.g. "5.7") if a patch-specific one isn't available.

Related MCP server: unreal-project-mcp

How It Works

Version detection. The server figures out which UE version to serve:

Priority	Source	Example
1	`UNREAL_VERSION` env var	`"5.5"`, `"5.7"`, `"5.7.3"`
2	`UNREAL_PROJECT_PATH`	Reads `.uproject` `EngineAssociation` field (e.g. `5.5.1` or `5.7`)

Set one of these to match your project. Without either, the server defaults to UE 5.7.

Database download. If the database for that version isn't cached locally, it downloads from GitHub (one time). For patch versions, falls back to the major.minor database if needed. Also checks for updates on startup.
Serve. All tool calls query the version-specific SQLite database. Exact lookups return in <1ms, searches in <5ms.

Each version has its own database with the correct signatures, deprecation warnings, and member lists for that release.

Tools

Tool	Purpose	Example
`search_unreal_api`	Find APIs by keyword	"character movement", "spawn actor", "K2Node"
`get_function_signature`	Exact signature with parameters and return type	`AActor::GetActorLocation`
`get_include_path`	Resolve `#include` for a type	"ACharacter" -> `#include "GameFramework/Character.h"`
`get_class_reference`	Full class reference card	"APlayerController", "UK2Node_SpawnActorFromClass", "UEdGraphSchema_K2"
`get_deprecation_warnings`	Check if an API is obsolete	"K2_AttachRootComponentTo" -> Use AttachToComponent() instead

Coverage

All Engine Runtime, Editor, Developer modules, plus built-in plugins (Enhanced Input, Gameplay Abilities, Common UI, Niagara, Chaos, and hundreds more).

Includes Blueprint graph internals: 158 UK2Node subclasses, UEdGraphSchema_K2, BlueprintGraph, KismetCompiler, and GraphEditor modules (1,120+ entries). If you're writing custom K2 nodes or editor tooling, it's indexed.

See the full list of supported versions and databases on the db-v1 release page. New versions are detected and built automatically every Monday via CI.

Record breakdown (UE 5.7):

Type	Count	Source
Classes (UCLASS)	10,075	`AActor`, `ACharacter`, `UGameplayStatics`, ...
Structs (USTRUCT)	9,014	`FHitResult`, `FVector`, `FTransform`, ...
Enums (UENUM)	3,475	`EMovementMode`, `ECollisionChannel`, ...
Functions (UFUNCTION)	23,414	Signatures with params, return types, specifiers
Properties (UPROPERTY)	66,340	Types, specifiers, doc comments
Delegates	2,406	Dynamic multicast, delegate declarations

Does not cover third-party plugins or marketplace assets. For those, rely on project source.

Benchmarks

Measured, not promised: 14 research questions across 2 testbeds, answered by 3 agent configs, every answer judged against ground truth verified beforehand (API facts against the docs database, source facts against the UE 5.8 engine tree). The full harness lives in docs/benchmark/ and re-runs with one command.

Config	Correct	Partial	Wrong
MCP + targeted Read	14/14	0	0
Skilled (Grep+Read)	11/14	2	1
Naive (full Reads)	12/14	1	1

The gap is not where you would expect. On well-documented includes and signatures, and on grep-able implementation facts inside a 14,000 line engine file, all three configs tie. Modern models know the documented UE surface and search local source competently. The gap opens on recent deprecations. Asked whether UKismetSystemLibrary::IsSplitScreen is still valid in 5.8, both non-MCP agents got it wrong: one declined, one asserted it was fine. It is deprecated in favor of HasMultipleLocalPlayers. Both also mischaracterized the 5.5-era change that made direct AActor::NetUpdateFrequency access deprecated. The MCP agent answered all three deprecation questions correctly, because the deprecation flag and replacement hint are in the database.

Why correctness rather than raw output? Agentic tools search code well now. Claude Code has shipped a Grep tool from the start, so an agent can find anything that is actually in your files. The problem is the parts that are not in your files: exact signatures, #include paths, and especially fresh deprecations. Those are not reliably in a model's memory either, and getting one wrong costs you a broken build or a silently deprecated call.

2 testbeds: 8 pure Unreal API lookups (exact signatures, #include paths, deprecations) and 6 questions answered inside CharacterMovementComponent.cpp (about 14,000 lines) from the UE 5.8 engine source
3 configs, same model (Sonnet) and turn limit: MCP tools + Grep/Read, Grep/Read only, Read only
The API sweep runs with no engine source present, the realistic case when writing UE C++ in your own project, so non-MCP configs answer from model memory. The source sweep gives every config the same files to grep
Ground truth verified before any runs; answers judged by a separate model session against that ground truth
Run it yourself: python docs/benchmark/run.py --cwd <dir> --questions <file> (results from July 2026; agent behavior moves, so re-run before quoting). Point it at your own UE project to reproduce the project-research scenario

Measured on UE 5.7 database (114,724 records), 50 iterations per query:

Query	Median	p95
Exact FQN lookup (`get_function_signature`)	<1ms	<1ms
FTS search: specific function name	<1ms	<1ms
FTS search: keyword ("spawn actor")	1ms	1ms
Include path resolution	2ms	2ms
Class reference (full member list)	22ms	23ms
Deprecation check	24ms	25ms

CLAUDE.md Snippet

Add this to your project's CLAUDE.md (or equivalent instructions file). This step is important. Without it, the AI has the tools but won't know when to reach for them.

## Unreal Engine API Lookup (unreal-api MCP)

Use the `unreal-api` MCP tools to verify UE C++ API usage instead of guessing. **Do not hallucinate signatures or #include paths.**

| When | Tool | Example |
|------|------|---------|
| Unsure about a function's parameters or return type | `get_function_signature` | `get_function_signature("AActor::GetActorLocation")` |
| Need the `#include` for a type | `get_include_path` | `get_include_path("ACharacter")` |
| Want to see all members on a class | `get_class_reference` | `get_class_reference("UCharacterMovementComponent")` |
| Searching for an API by keyword | `search_unreal_api` | `search_unreal_api("spawn actor")` |
| Checking if an API is deprecated | `get_deprecation_warnings` | `get_deprecation_warnings("K2_AttachRootComponentTo")` |
| Writing custom K2 nodes or editor tools | `get_class_reference` | `get_class_reference("UK2Node_SpawnActorFromClass")`, `get_class_reference("UEdGraphSchema_K2")` |

**Rules:**
- Before writing a UE API call you haven't used in this conversation, verify the signature with `get_function_signature`
- Before adding a `#include`, verify with `get_include_path` if unsure
- Covers: all Engine Runtime/Editor modules, built-in plugins (Enhanced Input, GAS, CommonUI, Niagara, etc.), Blueprint graph internals (UK2Node subclasses, EdGraphSchema, BlueprintGraph, KismetCompiler)
- Does NOT cover: third-party plugins or marketplace assets

Setup Details

Instead of setting UNREAL_VERSION, you can point to your Unreal project. The server reads the EngineAssociation field from your .uproject file:

{
  "mcpServers": {
    "unreal-api": {
      "command": "uvx",
      "args": ["unreal-api-mcp"],
      "env": {
        "UNREAL_PROJECT_PATH": "F:/Unreal Projects/MyProject"
      }
    }
  }
}

Using pip install:

pip install unreal-api-mcp

{
  "mcpServers": {
    "unreal-api": {
      "command": "unreal-api-mcp",
      "args": [],
      "env": {
        "UNREAL_VERSION": "5.5"
      }
    }
  }
}

Variable	Purpose	Example
`UNREAL_VERSION`	UE version to serve	`5.5`, `5.7`, `5.7.3`
`UNREAL_PROJECT_PATH`	Auto-detect version from .uproject	`F:/Unreal Projects/MyProject`
`UNREAL_INSTALL_PATH`	Override UE install path (for `ingest` only)	`H:/UE_5.6`

If you want to build a database from your own Unreal Engine installation instead of downloading:

# Build for a specific version
python -m unreal_api_mcp.ingest --unreal-version 5.6 --unreal-install "H:/UE_5.6"
python -m unreal_api_mcp.ingest --unreal-version 5.5 --unreal-install "H:/UE_5.5"

Databases are written to ~/.unreal-api-mcp/unreal_docs_{version}.db by default.

unreal-api-mcp/
├── src/unreal_api_mcp/
│   ├── server.py          # MCP server (5 tools)
│   ├── db.py              # SQLite + FTS5 database layer
│   ├── version.py         # Version detection + DB download
│   ├── header_parser.py   # Parse Unreal C++ headers (UCLASS, UFUNCTION, etc.)
│   ├── unreal_paths.py    # Locate UE installs + discover modules
│   └── ingest.py          # CLI ingestion pipeline
└── pyproject.toml

Databases are stored in ~/.unreal-api-mcp/ (downloaded on first run).

Troubleshooting

Problem	Fix
"Could not download UE X database"	Check internet connection. Or build locally: `python -m unreal_api_mcp.ingest --unreal-version 5.6 --unreal-install H:/UE_5.6`
Wrong API version being served	Set `UNREAL_VERSION` explicitly. Check stderr: `unreal-api-mcp: serving UE <version>`
Server won't start	Check `python --version` (needs 3.10+). Check path: `which unreal-api-mcp` or `where unreal-api-mcp`
Third-party plugins return no results	Marketplace/third-party plugins are not indexed. Only built-in Engine and Plugin APIs are covered.

Contact

Need a custom MCP server for your engine or framework? I build MCP tools that cut token waste and prevent hallucinations for AI-assisted game development. If you want something similar for your team's stack, reach out.

fuatcankoseoglu@gmail.com

License

MIT

Free to use, fork, modify, and share for any personal or non-commercial purpose. Commercial use requires permission.

Available Tools

3 tools

get_class_referenceB

Get all public members of an Unreal Engine class.

Args: class_name: The class name (e.g. "AActor", "ACharacter", "UGameplayStatics").

ParametersJSON Schema

Name	Required	Description	Default
`class_name`	Yes

Output Schema

ParametersJSON Schema

Name	Required	Description
`result`	Yes

TDQS

B3.4/5.0

Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations provided, the description carries full responsibility for behavioral disclosure. It only states the operation (get public members) but does not reveal whether the tool is read-only, if it requires specific permissions, how it handles errors (e.g., invalid class name), or any rate limits or side effects.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is extremely concise at two lines, front-loading the core purpose. The separate 'Args' block is clear and minimal. Every sentence adds value, with no redundancy.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the tool's simplicity (single parameter, output schema present), the description covers the essential information. It does not, however, mention error handling or assumptions about class existence. This is a minor gap but overall sufficient for an experienced developer or agent.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 0% for the single parameter class_name, so the description must compensate. It provides a brief description and examples (e.g., AActor, ACharacter), which add value beyond the raw schema. However, it does not explain naming conventions, case sensitivity, or the effect of invalid inputs. This is adequate but not comprehensive.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the verb 'Get' and the resource 'all public members of an Unreal Engine class,' making the tool's purpose immediately obvious. The provided examples (e.g., AActor) and the distinction from sibling tools (deprecation, function signature, etc.) further reinforce uniqueness.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines2/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description offers no guidance on when to use this tool versus alternatives like search_unreal_api or get_function_signature. It does not provide context, exclusions, or prerequisites. The agent is left to infer usage from the tool name alone.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

get_deprecation_warningsA

Check if an Unreal Engine API is deprecated.

Args: name: API name to check (function, class, property, etc.).

ParametersJSON Schema

Name	Required	Description	Default
`name`	Yes

Output Schema

ParametersJSON Schema

Name	Required	Description
`result`	Yes

TDQS

A3.8/5.0

Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations are provided, so the description must cover behavioral aspects. It only states it checks deprecation, with no mention of side effects, rate limits, or that it is a read-only operation. The output schema exists but the description adds no behavioral context beyond the basic function.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is concise and front-loaded, containing one sentence for the purpose and a brief parameter doc. No unnecessary words, making it efficient for an agent to parse.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the tool has a single parameter and an output schema, the description is adequate for a simple check. However, it lacks an example or note about return format, though the output schema compensates. It fits well with sibling tools.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

The schema description coverage is 0%, but the description provides clear semantics for the single parameter 'name', explaining it can be a function, class, property, etc. This adds meaningful guidance beyond the raw schema definition.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the tool checks deprecation status of an Unreal Engine API, using the verb 'check' and specifying the resource. It distinguishes from sibling tools like 'get_class_reference' and 'get_function_signature' which serve different purposes.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description does not provide explicit when-to-use or when-not-to-use guidance. While the intent is clear from the name and context, there is no mention of alternatives or prerequisites, relying on the implied context of sibling tools.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

get_include_pathA

Get the #include path for an Unreal Engine class, struct, or type.

Args: name: Class or type name (e.g. "AActor", "FHitResult", "ECollisionChannel").

ParametersJSON Schema

Name	Required	Description	Default
`name`	Yes

Output Schema

ParametersJSON Schema

Name	Required	Description
`result`	Yes

TDQS

A3.6/5.0

Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations are provided, so the description carries full burden. It states the function but does not disclose edge cases (e.g., what happens if the name is invalid, case sensitivity, or whether multiple results exist). Minimal behavioral context beyond the basic operation.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is short and to the point, with the main action front-loaded. It could be slightly more structured by integrating the parameter explanation into the first sentence, but overall it is efficient and avoids unnecessary words.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness3/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

The description is adequate for a simple tool with one parameter and an output schema (which covers return format). However, it lacks information about error handling or validation, leaving some gaps about tool behavior in edge cases.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Despite 0% schema description coverage, the description provides examples (e.g., 'AActor', 'FHitResult') and clarifies the parameter expects a class/struct/type name, adding significant meaning beyond the schema's bare type definition.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

Clearly states the tool retrieves the #include path for Unreal Engine classes/structs/types. The verb 'Get' and resource 'include path' are specific, and it is distinct from siblings like 'get_class_reference' or 'search_unreal_api'.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Implied usage from sibling list but no explicit guidance on when to use this over alternatives. Lacks any 'when not to use' or prerequisite information, though the purpose is straightforward.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Install Server

RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

TDQS

A3.9/5.0

Disambiguation5/5

Each tool has a clearly distinct purpose: retrieving class members, checking deprecation status, and getting include paths. No overlap or ambiguity.

Naming Consistency5/5

All tool names follow a consistent 'get_<action>' pattern with clear noun suffixes ('reference', 'warnings', 'path'), making them predictable and easy to understand.

Tool Count5/5

Three tools is an appropriate size for a focused API query server. Each tool addresses a common developer need without adding unnecessary complexity.

Completeness4/5

The tools cover core use cases for Unreal Engine API exploration, but could benefit from additional functionality like searching for APIs or listing available classes. Still, the surface is functional and satisfies the stated purpose.

Maintenance

ActivitySlowing

ResponsivenessSyncing

Resources

GitHub Repository

Need Help?

Related Servers

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Related MCP Connectors

ref-tools-ref-tools-mcp
ai.smithery
Provide your AI coding tools with token-efficient access to up-to-date technical documentation for…
Documentation Access Developer Tools Code Analysis
21,164
llmref A
org.llmref
Versioned documentation registry and semantic search for AI tools and coding assistants.
Documentation Access Search Developer Tools
42
Tseha AFeatured
io.tseha
Serves your design system and coding standards to coding agents, so they stop guessing.
Developer Tools Documentation Access
93
TeamShared B
io.github.xhad
Shared memory for coding agents. Stop re-explaining your codebase every session.
Knowledge & Memory Developer Tools
1041

Related MCP Servers

Unreal Engine Code Analyzer
Code Analysis Developer Tools
ayeletstudioindia
A
license
Not graded
quality
D
maintenance
Provides deep source code analysis for Unreal Engine codebases, allowing AI assistants to understand C++ class structures, search code, and analyze subsystems.
Updated a year ago (2025-08-06 17:32 UTC)
158
MIT
unreal-project-mcp
Code Analysis Developer Tools
tumourlove
A
license
A
quality
F
maintenance
Indexes Unreal Engine project C++ source, config files, dependencies, gameplay tags, replication, asset references, and log categories into a SQLite database and exposes tools for AI assistants to query structural and config info.
Updated 6 months ago (2026-03-06 09:36 UTC)
17
MIT
unreal-source-mcp
Code Analysis Developer Tools
tumourlove
A
license
Not graded
quality
F
maintenance
Indexes Unreal Engine source code into a local database, providing AI coding assistants with deep structural queries like class hierarchies, call graphs, and full-text search across all engine source files.
Updated 6 months ago (2026-03-06 09:36 UTC)
MIT
unreal-python-mcp
Code Execution Documentation Access
self-taught-code-tokushima
F
license
A
quality
D
maintenance
Enables AI coding assistants to browse Unreal Python API documentation, search classes and functions, and execute Python code in Unreal Editor.
Updated 7 months ago (2026-02-09 01:12 UTC)
8

Latest Blog Posts

Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)
By Om-Shree-0709 on July 25, 2026.
mcp
Agent Identity
OAuth 2.1
Your AI Chatbot Just Exposed Your CEO's Salary to an Intern
By Om-Shree-0709 on July 2, 2026.
Agent Identity
MCP Security
OAuth Delegation
Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)
By Om-Shree-0709 on June 30, 2026.
Agentic Ai
Prompt Injection
WebAssembly

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/Codeturion/unreal-api-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server

unreal-api-mcp

Quick Start

How It Works

Tools

Coverage

Benchmarks

CLAUDE.md Snippet

Setup Details

Troubleshooting

See Also

Contact

License

Available Tools

Output Schema

Output Schema

Output Schema

Maintenance

Resources

Looking for Admin?

Related MCP Connectors

Related MCP Servers

Latest Blog Posts

MCP directory API