MCP Servers · isaacphi

mcp-language-server

mcp-language-server bridges Language Server Protocol (LSP) tools with MCP-enabled AI clients, allowing LLMs to access semantic code analysis features like definition lookup, references, diagnostics, and rename operations. Written in Go, it works with language servers for Go, Rust, Python, TypeScript, C/C++, and others via stdio communication.

Talk to DEV.co Discuss Implementation

Source: GitHub — github.com/isaacphi/mcp-language-server

1.6k

GitHub stars

137

Forks

Primary language

BSD-3-Clause

License (OSI-approved)

Key facts

Objective fields from the source. Values we can't verify are shown as “Unknown” rather than guessed.

Field	Value
Repository	isaacphi/mcp-language-server
Owner	isaacphi
Primary language	Go
License	BSD-3-Clause — OSI-approved
Stars	1.6k
Forks	137
Open issues	61
Latest release	v0.1.1 (2025-05-16)
Last updated	2026-03-01
Source	https://github.com/isaacphi/mcp-language-server

What mcp-language-server is

A Go-based MCP server that wraps LSP implementations, exposing tools for semantic code navigation (definition, references, hover, diagnostics, rename, edit_file). Communicates with underlying language servers over stdio and exposes results to MCP clients. Uses mcp-go for MCP protocol and incorporates gopls code for LSP handling.

Quickstart

Get the mcp-language-server source

Clone the repository and explore it locally.

terminalbash

1git clone https://github.com/isaacphi/mcp-language-server.git2cd mcp-language-server3# follow the project's README for install & configuration

Need it deployed, integrated, or customized instead? DEV.co ships production installs.

Best use cases

AI-assisted code navigation and refactoring

LLMs need reliable symbol lookup, cross-reference finding, and rename operations. This server provides those capabilities at scale across large codebases without duplicating LSP logic.

Multi-language codebase analysis in Claude or compatible AI clients

Projects using Go, Rust, Python, TypeScript, or C/C++ can offload semantic queries to their respective language servers, giving AI clients precise code context.

Automated refactoring and code generation workflows

The edit_file tool enables programmatic multi-line edits; combined with rename and definition tools, this supports large-scale automated code transformations.

Implementation considerations

Requires manual installation of Go binary and at least one language server (gopls, rust-analyzer, pyright, etc.). PATH and environment variable configuration is language-specific.
Each language server must be independently installed and configured; the adapter does not bundle them. Workspace path must be specified at startup.
61 open issues suggest active development; changes to MCP or LSP specs may require updates. Test coverage appears present (Go Tests badge visible).
LSP communication is wrapped; performance depends on the underlying language server's responsiveness and resource usage. Large workspaces may cause latency.
Configuration is JSON-based (claude_desktop_config.json or equivalent). No built-in UI or configuration wizard; manual text editing required.

When to avoid it — and what to weigh

You need a production-hardened, long-term supported tool — Project is marked 'beta software' and was created Dec 2024. Only one release (v0.1.1, May 2025). Stability and backwards compatibility guarantees are unknown.
Your language server does not communicate over stdio — The adapter explicitly requires stdio-based LSP communication. Language servers using TCP, websockets, or other transports are not supported.
You need off-the-shelf support for obscure or emerging languages — Compatibility depends on a working LSP implementation for your language. Only documented examples cover Go, Rust, Python, TypeScript, and C/C++.
You cannot isolate or sandbox the language server process — The server spawns and runs language server binaries (gopls, rust-analyzer, etc.) in the same environment. Security posture depends on language server vetting and system isolation.

License & commercial use

BSD-3-Clause ('New' or 'Revised'). A permissive OSI license allowing commercial use, modification, and distribution with attribution and no warranty. Attribution file present for gopls-derived code.

BSD-3-Clause is a standard permissive license compatible with commercial software. No explicit restrictions on commercial use detected. However, verify that the underlying language servers (gopls, rust-analyzer, etc.) also permit your intended use case, as they have separate licenses.

DEV.co evaluation signals

Editorial assessment — not user reviews. Directional, with an explicit confidence level.

Signal	Assessment
Maintenance	Active
Documentation	Adequate
License clarity	Clear
Deployment complexity	Moderate
DEV.co fit	Good
Assessment confidence	Medium

Security considerations

The server spawns and communicates with language server binaries (gopls, rust-analyzer, etc.). Security posture depends entirely on those tools. No input validation, sandboxing, or access control described. Workspace path is user-specified; no restriction on what can be indexed. Language servers have full filesystem access and may execute arbitrary code (e.g., build scripts). Requires trust in all dependency binaries. No security audit or disclosure policy documented.

Alternatives to consider

Direct LSP integration in AI clients

Some MCP clients may natively support LSP or implement custom language-specific integrations without an adapter layer. Reduces one dependency.

Language-specific code analysis libraries (e.g., tree-sitter, custom AST parsers)

If you control the AI client, building language-specific analysis directly avoids spawning external processes and reduces setup complexity.

Anthropic's Files API or similar document-aware models

For static code analysis and context, large context-window models with semantic indexing may reduce the need for real-time LSP queries.

Software development agency

Build on mcp-language-server with DEV.co software developers

Evaluate mcp-language-server if you're building Claude integrations or MCP clients that require reliable symbol lookup, cross-references, and refactoring. Confirm language server availability and test in a staging environment first.

Talk to DEV.co

Related open-source tools

Surfaced by semantic similarity across the DEV.co open-source index.

mcptools github-mcp-server apify-mcp-server

Related on DEV.co

Explore the category and the services that help you build with it.

MCP Servers AI development Custom software and code tools Talk to DEV.co

mcp-language-server FAQ

Do I need to install the language server separately?

Yes. mcp-language-server is an adapter; it requires you to install (gopls, rust-analyzer, pyright, etc.) beforehand and ensure they are in PATH or specified by full path in config.

What MCP clients are supported?

Documentation explicitly shows Claude Desktop. Compatibility with other MCP clients (e.g., Cursor, other Anthropic integrations) is not documented. Likely works with any MCP-compliant client, but test required.

Can I use this in a CI/CD pipeline or server environment?

Unknown. No documented deployment examples for headless servers, containers, or CI systems. Only documented use is with Claude Desktop (macOS path examples). Requires evaluation for your environment.

What happens if the language server crashes?

Not documented. No resilience, restart, or fallback strategy described. Likely the MCP server or client will encounter an error; monitoring and recovery are unspecified.

Software developers & web developers for hire

DEV.co helps companies turn open-source tools like mcp-language-server into production software. Our software development services cover the full lifecycle — architecture, web development, integration, and maintenance — delivered by software developers and web developers who ship. Engage our software development agency to implement or customize it for your mcp servers stack.

Need LSP-backed semantic code tools for AI workflows?

Talk to DEV.co Schedule a Consultation