sharplens-mcp
SharpLensMcp is an MCP server for Roslyn-powered semantic analysis, navigation, refactoring, and code generation for .NET/C# solutions.
Navigate code: symbol info, go-to-definition, references, callers, implementations, type hierarchies, overloads, and containing members.
Search semantically: glob symbol search, multi-filter queries by async/static/accessibility/attributes/parameters, type members, and attribute usages.
Analyze code: diagnostics, code fixes, data/control flow, complexity metrics, dead/unused code, change impact, type compatibility, file overviews, and method analysis.
Refactor safely: rename, change signatures, extract methods/interfaces/variables, organize usings, format documents, apply Roslyn code actions, and preview before applying.
Generate code: constructors, equality members, null checks, missing interface members, test stubs, and validate generated code before writing.
Audit architecture and quality: DI registrations, reflection usage, circular dependencies, NuGet dependencies, source generators, generated code, and project structure.
Manage workspace state: load solutions, check server health, sync documents after external edits, and inspect project/dependency graphs.
Enables comprehensive .NET/C# code analysis through Microsoft Roslyn, providing semantic understanding of C# codebases including navigation, refactoring, diagnostics, and code generation capabilities.
Provides tools to analyze NuGet package dependencies across .NET projects, listing packages per project and detecting dependency relationships within the solution.
SharpLensMcp
A Model Context Protocol (MCP) server providing 92 AI-optimized tools for .NET/C# semantic code analysis, navigation, refactoring, and code generation using Microsoft Roslyn.
Built for AI coding agents - provides compiler-accurate code understanding that AI cannot infer from reading source files alone.
Installation
Via NuGet (Recommended)
dotnet tool install -g SharpLensMcpThen run with:
sharplensVia npm
npx -y sharplens-mcpBuild from Source
dotnet build -c Release
dotnet publish -c Release -o ./publishRelated MCP server: LSP-MCP
Claude Code Setup
Install the tool (pick one):
dotnet tool install -g SharpLensMcp
# or
npx -y sharplens-mcpCreate
.mcp.jsonin your project root:
{
"mcpServers": {
"sharplens": {
"type": "stdio",
"command": "npx",
"args": ["-y", "sharplens-mcp"],
"env": {
"DOTNET_SOLUTION_PATH": "/path/to/your/Solution.sln (or .slnx)"
}
}
}
}Restart Claude Code to load the MCP server
Verify by asking Claude to run a health check on the Roslyn server
Why Use This with Claude Code?
Claude Code has native LSP support for basic navigation (go-to-definition, find references). SharpLensMcp adds deep semantic analysis:
Capability | Native LSP | SharpLensMcp |
Go to definition | ✅ | ✅ |
Find references | ✅ | ✅ |
Find async methods missing CancellationToken | ❌ | ✅ |
Impact analysis (what breaks?) | ❌ | ✅ |
Dead code detection | ❌ | ✅ |
Complexity metrics | ❌ | ✅ |
Safe refactoring with preview | ❌ | ✅ |
Batch operations | ❌ | ✅ |
Configuration
Environment Variable | Description | Default |
| Path to a | None (must call |
| Use absolute paths instead of relative |
|
| Logging verbosity: |
|
| Timeout for long-running operations |
|
| Maximum diagnostics to return |
|
| Cache the file-path to document lookup |
|
The pre-1.6.0 ROSLYN_* spellings of these variables are still read as a fallback; the SHARPLENS_* spelling wins when both are set. Boolean values are case-insensitive. A number that does not parse or is below its minimum falls back to the default, and one above its maximum is capped; each case logs a warning to stderr.
Multi-targeted projects
A project built for several target frameworks is one project to every tool: each result is reported once, with a targetFrameworks list where it exists only under some frameworks. projectName accepts the plain name (MyLib) for all frameworks, or a framework-qualified name (MyLib(net8.0)) for one.
If DOTNET_SOLUTION_PATH is not set, you must call the load_solution tool before using other tools.
Migrating from 1.5.x tool names
Tool names no longer carry the roslyn: prefix — the colon violates the MCP tool-name pattern (^[a-zA-Z0-9_-]{1,64}$), which some clients enforce. Every tool keeps its name minus the prefix:
1.5.x name | 1.6.0 name |
|
|
|
|
|
|
...same rule for all tools... | drop the |
tools/list publishes only the new names. Calls using the old prefixed names are still accepted as aliases for one release and will be removed in the following one.
AI Agent Configuration Tips
AI models may have trained bias toward using their native tools (Grep, Read, LSP) instead of MCP server tools, even when SharpLensMcp provides better capabilities.
To ensure optimal tool usage:
Claude Code: Add to your project's
CLAUDE.md:For C# code analysis, prefer SharpLensMcp tools over native tools: - Use `search_symbols` instead of Grep for finding symbols - Use `get_method_source` instead of Read for viewing methods - Use `find_references` for semantic (not text) referencesOther MCP clients: Configure tool priority in your agent's system prompt
The semantic analysis from Roslyn is more accurate than text-based search, especially for overloaded methods, partial classes, and inheritance hierarchies.
Agent Responsibility: Document Synchronization
Important: SharpLensMcp maintains an in-memory representation of your solution for fast queries. When files are modified externally (via Edit/Write tools), the agent is responsible for synchronizing changes.
When to call sync_documents:
Action | Call sync_documents? |
Used Edit tool to modify .cs files | ✅ Yes |
Used Write tool to create new .cs files | ✅ Yes |
Deleted .cs files | ✅ Yes |
Used SharpLensMcp refactoring tools (rename, extract, etc.) | ❌ No (auto-updated) |
Modified .csproj files | ❌ No (use |
Usage:
# After editing specific files
sync_documents(filePaths: ["src/MyClass.cs", "src/MyService.cs"])
# After bulk changes - sync all documents
sync_documents()A full sync also picks up new .cs files the project's compile items include. SharpLensMcp's apply tools write only the files an edit changes, keep each file's encoding and line endings, and refuse with STALE_DOCUMENT when a target file changed on disk since it was read — run sync_documents, then retry. They also refuse with UNSUPPORTED_ENCODING to write a file whose bytes its encoding cannot reproduce (a BOM-less file that is not valid UTF-8 on a system whose fallback code page is UTF-8, such as Linux); save it as UTF-8 first.
Why this design?
This mirrors how LSP (Language Server Protocol) works - the client (editor) notifies the server of changes. This approach:
Eliminates race conditions (agent controls timing)
Avoids file watcher complexity and platform quirks
Is faster than full solution reload
Gives agents explicit control over workspace state
If you don't sync: Queries may return stale data (old method signatures, missing new files, etc.)
Features
92 Semantic Analysis Tools - Navigation, refactoring, code generation, diagnostics, discovery, audit/quality
AI-Optimized Descriptions - Clear USAGE/OUTPUT/WORKFLOW patterns
Structured Responses - Consistent
success/error/dataformat withsuggestedNextToolsZero-Based Coordinates - Clear warnings to prevent off-by-one errors
Preview Mode - Safe refactoring with preview before apply
Batch Operations - Multiple lookups in one call to reduce context usage
Tool Categories
Navigation & Discovery (24 tools)
Tool | Description |
| Semantic info at position |
| Jump to symbol definition |
| All references; each classified read/write/invocation/cast/typeof/nameof/attribute; optional |
| Interface/abstract implementations |
| Impact analysis - who calls this? |
| Multi-hop callers/callees graph with depth bound + cycle detection |
| Reachability + the connecting call path(s) between two methods; follows dispatch, with barriers and a checkpoint |
| Inheritance chain |
| Glob pattern search ( |
| Multi-filter search (async, public, etc.) |
| All members by type name |
| Multiple types in one call |
| Detailed signature by name |
| Find all subclasses |
| Full inheritance chain |
| List attributes on a symbol |
| Enclosing symbol at position |
| All overloads of a method |
| Find types/members by attribute |
| Inspect NuGet/BCL/external assembly types — members + XML docs |
| Map a pasted stack trace to file/line/symbol, mangling undone |
| Extensions applying to a type — classic and C# 14 blocks |
| Full XML docs for a symbol with |
| Navigate to the base member / interface members a member implements |
Analysis (17 tools)
Tool | Description |
| Compiler errors/warnings + configured analyzer findings (StyleCop, Roslynator, NetAnalyzers); matches CI |
| Public-API breaking-change report vs a git ref |
| Which exceptions can escape a method, and where they're caught |
| Structural similarity search (token-shingle fingerprints) |
| Compute dead-code removals + newly unused usings (generation-only) |
| Unreachable basic blocks per method (real CFG, not heuristics) |
| Compute the usings that fix CS0246/CS0103 (generation-only) |
| Variable assignments and usage |
| Branching/reachability |
| What breaks if changed? |
| Can A assign to B? |
| What does this method call? |
| Dead code detection |
| Compile check without writing |
| Cyclomatic, nesting, LOC, cognitive |
| Project and namespace cycle detection |
| Unimplemented interface/abstract members |
Refactoring (16 tools)
Tool | Description |
| Safe rename across solution |
| Add/remove/reorder parameters |
| Extract with data flow analysis |
| Generate interface from class |
| From fields/properties |
| Compute the contents to move a type into its own file (generation-only) |
| Compute a partial-class split for selected members (generation-only) |
| Sort and remove unused |
| Batch organize multiple files |
| Batch format files in project |
| All Roslyn refactorings at position |
| Apply any refactoring by title |
| Generate interface stubs |
| Field to property |
| Inline temp variable |
| Extract expression to variable |
Code Generation (3 tools)
Tool | Description |
| Generate ArgumentNullException guards |
| Equals/GetHashCode/operators |
| Compilable test skeleton for a method (framework auto-detected) |
Compound Tools (7 tools)
Tool | Description |
| Full type info in one call |
| Signature + callers + outgoing calls + location |
| File summary with diagnostics |
| Source code by name |
| Multiple method sources in one call |
| How to create a type |
| Composite audit dashboard: diagnostics + unused + coupling + coverage per project |
Audit & Quality (10 tools)
Tool | Description |
| Detect over-coupled types via efferent + afferent coupling + member-count thresholds |
| Find public surface no test reaches — tests detected as xUnit, NUnit, MSTest, TUnit, Reqnroll and SpecFlow discover them, with setup, fixtures and step bindings |
| Which tests cover a symbol — the inverse of find_untested_code |
| Where a type is constructed ( |
| Where a type appears in is/as/pattern matches |
| Where an exception type is thrown (optionally derived) |
| Where an exception type is caught (optionally via a base clause) |
| async void / blocking-on-async / unforwarded CancellationToken |
| Enforce namespace/project dependency rules over the type graph |
| Naming audit honoring .editorconfig rules, with conventional defaults |
Discovery (3 tools)
Tool | Description |
| Scan DI service registrations |
| Detect reflection/dynamic usage |
| Surface [InterceptsLocation] call rerouting, generated code included |
Infrastructure (12 tools)
Tool | Description |
| Server status + load fidelity ( |
| PackageReferences/ProjectReferences the compiler never needs |
| Compute the fix for every instance of a diagnostic id (generation-only) |
| Load .sln/.slnx/.slnf; reports partial loads ( |
| Sync file changes into loaded solution |
| Solution structure |
| Project dependencies |
| Available fixes for a diagnostic |
| Apply a specific code fix |
| NuGet package listing per project |
| List active source generators |
| View generated source code |
Other MCP Clients
For MCP clients other than Claude Code, add to your configuration:
{
"mcpServers": {
"sharplens": {
"command": "sharplens",
"args": [],
"env": {
"DOTNET_SOLUTION_PATH": "/path/to/your/Solution.sln (or .slnx)"
}
}
}
}Usage
Load a solution: Call
load_solutionwith path to.slnor.slnxfile (or setDOTNET_SOLUTION_PATH)Analyze code: Use any of the 92 tools for navigation, analysis, refactoring, audit
Refactor safely: Preview changes before applying with
preview: true
Architecture
MCP Client (AI Agent)
| stdin/stdout (JSON-RPC 2.0)
v
SharpLensMcp
- Protocol handling
- 92 AI-optimized tools
|
v
Microsoft.CodeAnalysis (Roslyn)
- MSBuildWorkspace
- SemanticModel
- SymbolFinderRequirements
.NET 8.0 SDK or later — works with .NET 8, 9, 10, and future versions. Analyzes any .NET 8+ project/solution.
MCP-compatible AI agent
FAQ
Why does the tool target net8.0 — can it analyze my .NET 9 / .NET 10 project?
Yes. net8.0 is the tool's own runtime floor — the Roslyn 5.x packages it builds on require it — not a ceiling on what it can analyze. RollForward lets the installed tool run on newer .NET runtimes, and MSBuildWorkspace loads each project's real target framework from its csproj, so one install analyzes solutions targeting .NET 8, 9, 10, and beyond.
Development
Adding New Tools
Add the method to the matching
src/RoslynService.*.cspartial (Navigation, Analysis, Refactoring, CallAnalysis, …) and return through the shared response envelope:
public async Task<object> YourToolAsync(string param1, int? param2 = null,
CancellationToken cancellationToken = default)
{
EnsureSolutionLoaded();
// Your logic...
return CreateSuccessResponse(
data: new { /* results */ },
suggestedNextTools: new[] { "next_tool_hint" }
);
}Register one
ToolDefinitioninsrc/ToolRegistry.cs— its name, description, input schema, theReadOnlyflag (mutating tools passReadOnly: falseand receive adestructiveHintannotation), and a handler that binds arguments throughJsonRpcParametersand calls your method. The registry drives bothtools/listand dispatch; there is no separate switch to edit. Two tests keep it honest:ToolsListGoldenTestslocks the published schema byte-for-byte (re-capture the golden when a schema change is intentional), andToolSchemaParityTestsasserts every parameter the handler reads is declared in the schema.Build and publish:
dotnet build -c Release
dotnet publish -c Release -o ./publishAdd both test levels — a unit test of the
RoslynServicemethod against a deterministic fixture, AND a wire test through the MCP dispatcher (intests/SharpLensMcp.Tests/Mcp/) with exact value locks plus an error path. This is non-negotiable; see Testing.
Testing
Every test must satisfy the Testing Charter (C1–C9) in tests/SharpLensMcp.Tests/TESTING.md — the standing contract. The headline rules:
Lock exact values (C1): assert a concrete name / count / substring / error code /
(line, column)— neverNotBeNull/> 0/ a type-only check as the sole assertion.Both levels per tool (C4): a unit test against a
Fixtures/*.csfixture and a dispatcher (wire) test that unwrapscontent[0].text, plus an error path.Right casing (C3): in-process Newtonsoft yields PascalCase
error.Code/meta.TotalCount; the MCP wire yields camelCase. Read the casing your test's path actually produces.Deterministic (C7): the suite is serialized via
xunit.runner.json; fixture mutators always restore; the timeout test uses a forced-cancellation seam, not a timing race.Out-of-process spine (C6):
StdioIntegrationTestsvalue-pins one tool per category over the real binary and runs thetools/listgolden over the stdio pipe.Pre-commit gate (C9): build-clean + green is necessary but not sufficient — re-read each changed test and confirm it fails on a wrong answer.
Run the suite:
dotnet test -c ReleaseKey Files
File | Purpose |
| Tool implementations split by concern across ~30 partials (Navigation, Analysis, Refactoring, Inspection, Validation, TypeDiscovery, Discovery, ExternalApi, Quality, Metrics, CodeActions, CodeGeneration, Compound, CallAnalysis, ExceptionFlow, StackTrace, ApiSurface, SimilarCode, …) — each file's name predicts its contents |
| MCP protocol mechanics: JSON-RPC parse loop, |
| The tool surface: one |
| Typed JSON-RPC argument accessors and the |
| Typed records used by the audit composite, constructor generator, and signature-change parser (one type per file) |
License
MIT - See LICENSE for details.
Available Tools
62 toolsroslyn:add_null_checksA
Add ArgumentNullException.ThrowIfNull guard clauses for nullable parameters.
USAGE: Position cursor on a method with reference type parameters. OUTPUT: Generated guard clauses inserted at method start. IMPORTANT: Uses ZERO-BASED coordinates (editor line - 1).
Inputs: line (required), column (required), filePath (required), preview.
roslyn:analyze_change_impactA
Analyze what would break if you change a symbol. Identifies breaking changes before you make them.
USAGE: analyze_change_impact(filePath, line, column, changeType="rename|changeType|addParameter|removeParameter") OUTPUT: List of impacted locations, whether change is safe, and specific issues at each location. IMPORTANT: Uses ZERO-BASED coordinates (editor line - 1).
Inputs: line (required), column (required), filePath (required), changeType (required), +1 more.
roslyn:analyze_control_flowA
Analyze branching and reachability in a code region.
Returns: entryPoints, exitPoints, returnStatements, endPointIsReachable.
USAGE: analyze_control_flow("path/to/file.cs", startLine=10, endLine=25)
Inputs: endLine (required), filePath (required), startLine (required).
roslyn:analyze_data_flowB
Analyze variable assignments and usage in a code region.
Returns: variablesDeclared, alwaysAssigned, dataFlowsIn/Out, readInside/Outside, writtenInside/Outside, captured.
USAGE: analyze_data_flow("path/to/file.cs", startLine=10, endLine=25)
Inputs: endLine (required), filePath (required), startLine (required).
roslyn:analyze_methodA
Get comprehensive method analysis in ONE CALL: signature + callers + outgoing calls + location.
USAGE: analyze_method("MyService", "ProcessData") or analyze_method("MyClass", "Calculate", includeCallers=true, includeOutgoingCalls=true)
Inputs: typeName (required), methodName (required), maxCallers, includeCallers, +2 more.
roslyn:apply_code_action_by_titleA
Apply a code action by its title. Supports exact and partial matching.
USAGE: apply_code_action_by_title(filePath, line, column, title) OUTPUT: Changed files with preview or applied changes WORKFLOW: (1) Call get_code_actions_at_position first, (2) Apply with preview=true, (3) Apply with preview=false IMPORTANT: Uses ZERO-BASED coordinates (editor line - 1).
Inputs: line (required), title (required), column (required), filePath (required), +3 more.
roslyn:apply_code_fixA
Apply automated code fix for a diagnostic. WORKFLOW: (1) Call with no fixIndex to list available fixes, (2) Call with fixIndex and preview=true to preview changes, (3) Call with preview=false to apply. IMPORTANT: Uses ZERO-BASED coordinates.
Inputs: line (required), column (required), filePath (required), diagnosticId (required), +2 more.
roslyn:change_signatureA
Change a method signature and preview impact on all call sites.
ACTIONS: add (new param), remove, rename, reorder parameters. WORKFLOW: (1) Call with preview=true (default) to see affected call sites, (2) Review changes, (3) Call with preview=false to apply. OUTPUT: oldSignature, newSignature, list of call sites needing updates. IMPORTANT: Uses ZERO-BASED coordinates (editor line - 1).
Inputs: line (required), column (required), changes (required), filePath (required), +1 more.
roslyn:check_type_compatibilityA
Check if one type can be assigned to another. Use before generating assignments or casts.
USAGE: check_type_compatibility(sourceType="MyDerivedClass", targetType="MyBaseClass") OUTPUT: compatible (bool), requiresCast (bool), conversionKind, and explanation of why/why not.
Inputs: sourceType (required), targetType (required).
roslyn:dependency_graphA
Visualize project dependencies as a graph. Shows which projects reference which, detects circular dependencies.
OUTPUT: format="json" returns structured data with nodes/edges. format="mermaid" returns diagram syntax. USE CASE: Understand solution architecture, find circular dependencies, plan refactoring.
Inputs: format.
roslyn:encapsulate_fieldA
Convert a field to a property with getter/setter.
USAGE: Position cursor on a field declaration. OUTPUT: Generated property wrapping the field. IMPORTANT: Uses ZERO-BASED coordinates (editor line - 1).
Inputs: line (required), column (required), filePath (required), preview.
roslyn:extract_interfaceA
Generate an interface from a class or struct. Extracts all public instance members (methods, properties, events).
USAGE: Position on class declaration, provide interfaceName="IMyService". OUTPUT: Generated interface code ready to insert. Useful for dependency injection and testability. IMPORTANT: Uses ZERO-BASED coordinates (editor line - 1).
Inputs: line (required), column (required), filePath (required), interfaceName (required), +1 more.
roslyn:extract_methodA
Extract selected statements into a new method. Uses data flow analysis to determine parameters and return type.
USAGE: Specify startLine/endLine range containing complete statements inside a method. OUTPUT: extractedCode (the new method), replacementCode (the call to insert), detected parameters and return type. WORKFLOW: (1) Preview with preview=true, (2) Apply with preview=false. IMPORTANT: Uses ZERO-BASED coordinates (editor line - 1).
Inputs: endLine (required), filePath (required), startLine (required), methodName (required), +2 more.
roslyn:extract_variableA
Extract an expression to a local variable.
USAGE: Position cursor on or select an expression. OUTPUT: Expression extracted to a new local variable. IMPORTANT: Uses ZERO-BASED coordinates (editor line - 1).
Inputs: line (required), column (required), filePath (required), endLine, +2 more.
roslyn:find_attribute_usagesA
Find all types and members decorated with a specific attribute.
USAGE: find_attribute_usages(attributeName: "Authorize") USAGE: find_attribute_usages(attributeName: "HttpGet", projectName: "MyApi")
OUTPUT: List of symbols with the attribute, their kind, arguments, and source location. Use for: finding all API endpoints, authorization points, serialization config, test fixtures.
Inputs: attributeName (required), maxResults, projectName.
roslyn:find_callersA
Find all methods/properties that call or reference a specific symbol (inverse of find_references). Essential for impact analysis: 'If I change this method, what code will be affected?' IMPORTANT: Uses ZERO-BASED coordinates.
Inputs: line (required), column (required), filePath (required), maxResults.
roslyn:find_circular_dependenciesA
Detect cycles in project or namespace dependency graphs.
USAGE: find_circular_dependencies() — project-level cycles USAGE: find_circular_dependencies(level: "namespace") — namespace-level cycles
OUTPUT: Dependency graph with any detected cycles listed. Use for: architecture analysis, identifying tightly coupled components.
Inputs: level.
roslyn:find_implementationsA
Find all implementations of an interface or abstract class. Returns implementing types with their locations. IMPORTANT: Uses ZERO-BASED coordinates (editor line - 1).
Inputs: line (required), column (required), filePath (required), maxResults.
roslyn:find_referencesA
Find all references to a symbol across the entire solution. Returns file paths, line numbers, and code context for each reference. IMPORTANT: Uses ZERO-BASED coordinates (editor line 10 = pass line 9).
Inputs: line (required), column (required), filePath (required), maxResults.
roslyn:find_reflection_usageA
Detect dynamic/reflection-based type and method usage that is invisible to static reference searches.
USAGE: find_reflection_usage() USAGE: find_reflection_usage(projectName: "MyApp", maxResults: 50)
OUTPUT: List of reflection API calls with the API used, context, and location. Use for: finding hidden dependencies before refactoring, security audits, understanding dynamic behavior.
Inputs: maxResults, projectName.
roslyn:find_unused_codeA
Find unused types, methods, properties, and fields in a project or entire solution. Returns symbols with zero references (excluding their declaration).
USAGE: find_unused_code() for entire solution, or find_unused_code(projectName="MyProject") for specific project. OUTPUT: List of unused symbols with location, kind, and accessibility. Default limit: 50 results.
Inputs: maxResults, projectName, includePrivate, includeInternal, +1 more.
roslyn:format_document_batchB
Format multiple documents in a project using Roslyn's NormalizeWhitespace. Ensures consistent indentation, spacing, and line breaks. PREVIEW mode by default - set preview=false to apply changes.
Inputs: preview, projectName, includeTests.
roslyn:generate_constructorA
Generate a constructor from fields and/or properties of a type.
USAGE: Position on class/struct declaration. Use includeProperties=true for auto-properties. OUTPUT: constructorCode string ready to paste, parameter list, and field assignments. IMPORTANT: Uses ZERO-BASED coordinates (editor line - 1).
Inputs: line (required), column (required), filePath (required), includeProperties, +1 more.
roslyn:generate_equality_membersA
Generate Equals, GetHashCode, and == / != operators for a type.
USAGE: Position cursor on a class or struct declaration. OUTPUT: Generated equality members comparing all instance fields and properties. IMPORTANT: Uses ZERO-BASED coordinates (editor line - 1).
Inputs: line (required), column (required), filePath (required), preview, +1 more.
roslyn:get_attributesB
Find all symbols with specific attributes.
USAGE:
Find obsolete: get_attributes("Obsolete")
Find serializable: get_attributes("Serializable")
Scope to project: get_attributes("Obsolete", scope="project:MyProject")
Scope to file: get_attributes("Obsolete", scope="file:MyClass.cs")
Inputs: attributeName (required), scope, maxResults.
roslyn:get_base_typesA
Get full inheritance chain BY NAME.
USAGE: get_base_types("MyService") returns: MyService → BaseService → ... → Object
Inputs: typeName (required).
roslyn:get_code_actions_at_positionA
Get ALL available code actions (fixes + refactorings) at a position. This is the master tool that exposes 100+ Roslyn refactorings.
USAGE: get_code_actions_at_position(filePath, line, column) or with selection: add endLine, endColumn OUTPUT: List of actions with title, kind (fix/refactoring), equivalenceKey WORKFLOW: (1) Call this to see available actions, (2) Use apply_code_action_by_title to apply one IMPORTANT: Uses ZERO-BASED coordinates (editor line - 1).
Inputs: line (required), column (required), filePath (required), endLine, +3 more.
roslyn:get_code_fixesA
Get available code fixes for a specific diagnostic. Returns list of fix titles and descriptions. WORKFLOW: (1) get_diagnostics to find issues, (2) get_code_fixes to see options, (3) apply_code_fix to fix.
Inputs: line (required), column (required), filePath (required), diagnosticId (required).
roslyn:get_complexity_metricsA
Get complexity metrics for a method or entire file.
METRICS: cyclomatic (decision points), nesting (max depth), loc (lines), parameters (count), cognitive (Sonar-style) USAGE: get_complexity_metrics(filePath) for file, or add line/column for specific method OUTPUT: Per-method breakdown with all requested metrics IMPORTANT: Uses ZERO-BASED coordinates (editor line - 1).
Inputs: filePath (required), line, column, metrics.
roslyn:get_containing_memberA
Get information about the containing method/property/class at a position. Returns the enclosing symbol's name, kind, and signature. Useful for understanding context. IMPORTANT: Uses ZERO-BASED coordinates (editor line - 1).
Inputs: line (required), column (required), filePath (required).
roslyn:get_derived_typesA
Find all types inheriting from a base type BY NAME.
USAGE:
Find all subclasses: get_derived_types("BaseService")
Direct children only: get_derived_types("BaseClass", includeTransitive=false)
Inputs: baseTypeName (required), maxResults, includeTransitive.
roslyn:get_diagnosticsA
Get compiler errors, warnings, and info messages for a file or entire project. Returns: list of diagnostics with id, message, severity, and location. Use before committing to catch issues.
Inputs: filePath, severity, projectPath, includeHidden.
roslyn:get_di_registrationsA
Scan for dependency injection service registrations (AddScoped, AddTransient, AddSingleton, etc.).
USAGE: get_di_registrations() USAGE: get_di_registrations(projectName: "MyApi")
OUTPUT: List of DI registrations with lifetime, service type, implementation type, and location. Use for: understanding service wiring, finding missing registrations, auditing lifetimes.
Inputs: projectName.
roslyn:get_file_overviewA
Get comprehensive file overview in ONE CALL: diagnostics summary + type declarations + namespace + line count.
USAGE: get_file_overview("path/to/MyClass.cs")
Inputs: filePath (required).
roslyn:get_generated_codeA
View the source code produced by a source generator.
USAGE: get_generated_code(projectName: "MyApp", generatedFileName: "MyType.g.cs")
OUTPUT: Full source code of the generated file. Use get_source_generators first to discover available generated files.
Inputs: projectName (required), generatedFileName (required).
roslyn:get_instantiation_optionsA
Get all ways to create an instance of a type: constructors, factory methods, and builder patterns.
USAGE: get_instantiation_options(typeName="HttpClient") OUTPUT: List of constructors with signatures, static factory methods, and hints (e.g., "implements IDisposable").
Inputs: typeName (required).
roslyn:get_method_overloadsA
Get all overloads of a method. Returns list of signatures with parameter details. Use when you need to choose between overloads. IMPORTANT: Uses ZERO-BASED coordinates (editor line - 1).
Inputs: line (required), column (required), filePath (required).
roslyn:get_method_signatureB
Get detailed method signature BY NAME including parameters, return type, nullability, and modifiers.
USAGE: get_method_signature("MyClass", "ProcessData") or with overload selection: get_method_signature("MyClass", "ProcessData", overloadIndex=1)
Inputs: typeName (required), methodName (required), overloadIndex.
roslyn:get_method_sourceA
Get the actual source code of a method by type and method name. Eliminates need for file Read.
USAGE: get_method_source(typeName="MyService", methodName="ProcessData") OUTPUT: Full method source including signature, body, location (file + line numbers), and line count.
Inputs: typeName (required), methodName (required), overloadIndex.
roslyn:get_method_source_batchA
Get source code for multiple methods in a single call (batch optimization).
USAGE: get_method_source_batch(methods: [{typeName: 'ServiceA', methodName: 'Process'}, {typeName: 'ServiceB', methodName: 'Handle'}]) OUTPUT: Results array with source for each method, plus errors array for any that failed. BENEFIT: One call instead of multiple - reduces round trips when tracing code flows.
Inputs: methods (required), maxMethods.
roslyn:get_missing_membersA
Get all interface and abstract members that must be implemented for a type.
USAGE: Position on a class that implements interfaces or extends abstract classes. OUTPUT: List of missing members with exact signatures ready to copy. Use before implementing interfaces. IMPORTANT: Uses ZERO-BASED coordinates (editor line - 1).
Inputs: line (required), column (required), filePath (required).
roslyn:get_nuget_dependenciesA
List NuGet package references per project with versions.
USAGE: get_nuget_dependencies() USAGE: get_nuget_dependencies(projectName: "MyApp")
OUTPUT: List of projects with their NuGet packages, versions, and asset settings. Use for: dependency audits, version checks, understanding external dependencies.
Inputs: projectName.
roslyn:get_outgoing_callsA
Get all methods and properties that a method calls. Helps understand method dependencies and behavior. Returns list of called symbols with locations. IMPORTANT: Uses ZERO-BASED coordinates (editor line - 1).
Inputs: line (required), column (required), filePath (required), maxDepth.
roslyn:get_project_structureA
Get solution/project structure. IMPORTANT: For large solutions (100+ projects), use summaryOnly=true or projectNamePattern to avoid token limit errors. Maximum output is limited to 25,000 tokens.
Inputs: maxProjects, summaryOnly, includeDocuments, includeReferences, +1 more.
roslyn:get_source_generatorsA
List active source generators and their generated output per project.
USAGE: get_source_generators() USAGE: get_source_generators(projectName: "MyApp")
OUTPUT: List of generators with their assembly info and generated files. Use for: understanding generated code, debugging generator issues.
Inputs: projectName.
roslyn:get_symbol_infoA
Get detailed semantic information about a symbol at a specific position. IMPORTANT: Uses ZERO-BASED coordinates. If your editor shows 'Line 14, Column 5', pass line=13, column=4. Returns symbol kind, type, namespace, documentation, and location.
Inputs: line (required), column (required), filePath (required).
roslyn:get_type_hierarchyA
Get the inheritance hierarchy (base types and derived types) for a type. Returns baseTypes chain and derivedTypes list. IMPORTANT: Uses ZERO-BASED coordinates (editor line - 1).
Inputs: line (required), column (required), filePath (required), maxDerivedTypes.
roslyn:get_type_membersA
Get all members (methods, properties, fields, events) of a type BY NAME.
USAGE PATTERNS:
Basic: get_type_members("MyClass") - list all members
With inheritance: get_type_members("MyService", includeInherited=true)
Filter by kind: get_type_members("MyClass", memberKind="Method")
Verbosity control: verbosity="summary" (names only), "compact" (default, + signatures), "full" (+ docs, attrs)
WORKS WITH: Fully-qualified ("MyNamespace.MyClass"), simple ("MyClass"), or partial names.
Inputs: typeName (required), verbosity, maxResults, memberKind, +1 more.
roslyn:get_type_members_batchA
Get members for multiple types in a single call (batch optimization).
USAGE: get_type_members_batch(typeNames: ['ServiceA', 'ServiceB', 'ControllerC']) OUTPUT: Results for each type with members, or error if type not found BENEFIT: One call instead of multiple - reduces context usage for AI agents
Inputs: typeNames (required), verbosity, memberKind, includeInherited, +1 more.
roslyn:get_type_overviewA
Get comprehensive type overview in ONE CALL: type info + base types (first 3) + member counts.
USAGE: get_type_overview("MyService") - returns everything you need to understand a type quickly.
Inputs: typeName (required).
roslyn:go_to_definitionA
Fast navigation to symbol definition. Returns the definition location without finding all references. IMPORTANT: Uses ZERO-BASED coordinates (editor line 10 = pass line 9).
Inputs: line (required), column (required), filePath (required).
roslyn:health_checkA
Check the health and status of the Roslyn MCP server and workspace. Returns: server status, solution loaded state, project count, and memory usage. Call this first to verify the server is ready.
Inputs: no named parameters.
roslyn:implement_missing_membersA
Generate stub implementations for interface/abstract members.
USAGE: Position cursor on class declaration that implements interface or extends abstract class. OUTPUT: Generated stub code for all missing members. IMPORTANT: Uses ZERO-BASED coordinates (editor line - 1).
Inputs: line (required), column (required), filePath (required), preview.
roslyn:inline_variableA
Inline a variable, replacing all usages with its value.
USAGE: Position cursor on a variable declaration or usage. OUTPUT: Variable removed and all usages replaced with the expression. IMPORTANT: Uses ZERO-BASED coordinates (editor line - 1).
Inputs: line (required), column (required), filePath (required), preview.
roslyn:load_solutionA
Load a .NET solution for analysis. MUST be called before using any other analysis tools. Returns: projectCount, documentCount, and load time. Use health_check to verify current state.
Inputs: solutionPath (required).
roslyn:organize_usingsA
Sort and remove unused using directives in a file. Returns the modified file content. Automatically removes unused usings and sorts alphabetically.
Inputs: filePath (required).
roslyn:organize_usings_batchA
Organize using directives for multiple files in a project. Supports file pattern filtering (e.g., '.cs', 'Services/.cs'). PREVIEW mode by default - set preview=false to apply changes.
Inputs: preview, filePattern, projectName.
roslyn:rename_symbolA
Safely rename a symbol (type, method, property, etc.) across the entire solution. Uses Roslyn's semantic analysis to ensure all references are updated. SUPPORTS PREVIEW MODE - always preview first! IMPORTANT: Uses ZERO-BASED coordinates. Default shows first 20 files with summary verbosity.
Inputs: line (required), column (required), newName (required), filePath (required), +3 more.
roslyn:search_symbolsA
Search for types, methods, properties, etc. by name across the solution. Supports glob patterns (e.g., 'Handler' finds classes ending with 'Handler', 'Get' finds symbols starting with 'Get'). Use ? for single character wildcard. PAGINATION: Returns totalCount and hasMore. Use offset to paginate through results.
Inputs: query (required), kind, offset, maxResults, +1 more.
roslyn:semantic_queryA
Advanced semantic code query with multiple filters. Find symbols based on their semantic properties.
EXAMPLES:
Async methods without CancellationToken: isAsync=true, parameterExcludes=["CancellationToken"]
Public static methods: accessibility="Public", isStatic=true
Classes with [Obsolete]: kinds=["Class"], attributes=["ObsoleteAttribute"]
FILTERS: All specified filters are combined with AND logic. Omit a filter to skip it. Returns symbol details with locations.
Inputs: type, kinds, isAsync, isStatic, +7 more.
roslyn:sync_documentsA
Synchronize document changes from disk into the loaded solution. Call this after using Edit/Write tools to ensure Roslyn has fresh content.
USAGE:
sync_documents(filePaths: ["src/Foo.cs", "src/Bar.cs"]) - sync specific files
sync_documents() - sync ALL documents (refresh entire solution)
WHEN TO CALL:
After using Edit tool to modify .cs files
After using Write tool to create new .cs files
After deleting .cs files
NOT needed after using SharpLensMcp refactoring tools (they auto-update)
HANDLES: Modified files (updates content), new files (adds to solution), deleted files (removes from solution). Much faster than load_solution - only updates documents, doesn't re-parse projects.
Inputs: filePaths.
roslyn:validate_codeA
Check if code would compile without writing to disk. Use to validate generated code before applying.
USAGE: validate_code(code="public void Foo() {}", contextFilePath="path/to/file.cs") to check with existing usings. OUTPUT: compiles (bool), errors list with line numbers. Essential before inserting AI-generated code.
Inputs: code (required), standalone, contextFilePath.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 62 tools
Most tools have distinct purposes, but some pairs like find_references and find_callers could be confused without reading descriptions. Overall, the descriptions help differentiate.
Tools use a consistent snake_case verb_noun pattern, with minor exceptions like 'go_to_definition' instead of 'get_definition'. The pattern is mostly uniform.
With 62 tools, the surface is overly large. Many specialized and batch tools could be consolidated. This exceeds the typical well-scoped range of 3-15 tools.
The tool set covers a broad range of Roslyn features including analysis, refactoring, navigation, and diagnostics. Minor gaps exist (e.g., no individual add-using tool), but overall coverage is strong.
Maintenance
Related MCP Connectors
A comprehensive Model Context Protocol (MCP) server that enables AI assistants to interact with yo…
The Cortex MCP server provides read-only access to real-time engineering context from the Cortex developer portal, allowing AI coding assistants to answer natural language questions about your organization's catalog (microservices, libraries, domains, teams, infrastructure), scorecards (engineering standards and best practices), initiatives (goals and deadlines), and Engineering Intelligence metrics. It includes tools for querying documentation, tracking personal entities, and accessing AI-assisted insights across the entire Cortex ecosystem.
Nifty's MCP server — exposes tasks, projects, messages, and files as tools for AI agents.
MCP server for your apps' tools and custom tools, plus hosted AI agents and approval-gated workflows
Related MCP Servers
- AlicenseAqualityAmaintenanceRoslyn-based MCP server providing semantic code intelligence for .NET codebases — type hierarchies, call sites, DI registrations, and reflection usage for Claude Code6775MIT
- FlicenseAqualityDmaintenanceMinimal MCP server bridging a Roslyn C# language server to AI agents, exposing tools for diagnostics, call hierarchy, and type hierarchy.1-
- AlicenseNot gradedqualityBmaintenanceA semantic map of your .NET solution for AI coding agents. Analyzes a solution with Roslyn into a queryable code graph exposing 11 read-only tools (find_symbol, impact_analysis, find_implementations, etc.) over MCP. 100% local, no telemetry, MIT licensed.25MIT
- AlicenseNot gradedqualityBmaintenanceA local .NET MCP server for coding agents, providing persistent semantic indexing, typed-graph DI wiring resolution, reduced context, and compiler-backed verification for .NET projects.37Apache 2.0