RefactorMCP
Automated refactoring tools for C# code transformation using Roslyn.
RefactorMCP
RefactorMCP is a Model Context Protocol server that exposes Roslyn-based refactoring tools for C#.
Usage
Run the console application directly or host it as an MCP server:
dotnet run --project RefactorMCP.ConsoleApp
For usage examples see EXAMPLES.md.
Available Refactorings
- Extract Method – create a new method from selected code and replace the original with a call (expression-bodied methods are not supported).
- Introduce Field/Parameter/Variable – turn expressions into new members; fails if a field already exists.
- Convert to Static – make instance methods static using parameters or an instance argument.
- Move Static Method – relocate a static method and keep a wrapper in the original class.
- Move Instance Method – move one or more instance methods to another class and delegate from the source. If a moved method no longer accesses instance members, it is made static automatically. Provide a
methodNameslist along with optionalconstructor-injectionsandparameter-injectionsto control dependencies. - Move Multiple Methods (instance) – move several methods and keep them as instance members of the target class. The source instance is injected via the constructor when required.
- Move Multiple Methods (static) – move multiple methods and convert them to static, adding a
thisparameter. - Make Static Then Move – convert an instance method to static and relocate it to another class in one step.
- Move Type to Separate File – move a top-level type into its own file named after the type.
- Make Field Readonly – move initialization into constructors and mark the field readonly.
- Transform Setter to Init – convert property setters to init-only and initialize in constructors.
- Constructor Injection – convert method parameters to constructor-injected fields or properties.
- Safe Delete – remove fields or variables only after dependency checks.
- Extract Class – create a new class from selected members and compose it with the original.
- Inline Method – replace calls with the method body and delete the original.
- Extract Decorator – create a decorator class that delegates to an existing method.
- Create Adapter – generate an adapter class wrapping an existing method.
- Add Observer – introduce an event and raise it from a method.
- Use Interface – change a method parameter type to one of its implemented interfaces.
- List Tools – display all available refactoring tools as kebab-case names.
Metrics and summaries are also available via the metrics:// and summary:// resource schemes.
Contributing
- Run
dotnet testto ensure all tests pass. - Format the code with
dotnet formatbefore opening a pull request.
License
Licensed under the Mozilla Public License 2.0.
Related Servers
Scout Monitoring MCP
sponsorPut performance and error data directly in the hands of your AI assistant.
Alpha Vantage MCP Server
sponsorAccess financial market data: realtime & historical stock, ETF, options, forex, crypto, commodities, fundamentals, technical indicators, & more
JavaScript Sandbox
Provides a secure JavaScript execution environment for running code snippets.
Refine Prompt
Refines and structures prompts for large language models using the Anthropic API.
Gateway MCP Server
A gateway server that intelligently routes MCP requests to multiple backend servers.
Flux Schnell MCP Server
Generate images using the Flux Schnell model via the Replicate API.
Cashfree MCP Server
Integrate AI tools and agents with Cashfree's Payment Gateway, Payouts, and SecureID APIs.
Mesh Agent MCP Server
Connects to Heurist Mesh APIs to provide access to various blockchain and web3 tools.
Qase MCP Server
An MCP server for interacting with the Qase test management platform.
MetaMCP
A proxy server that combines multiple MCP servers into a single endpoint, routing requests to the appropriate underlying server.
Remote MCP Server (Authless)
An example of a remote MCP server deployable on Cloudflare Workers, without authentication.
SeedDream 3.0
Generate images using Bytedance's SeedDream 3.0 model via the FAL AI platform.