Lets your AI coding assistant jump to definitions, find every usage of a symbol, and rename across a codebase, using the language servers you already run. It works even when the assistant gets line and column numbers wrong.
Claude Code
$ npx -y skills add ktnyt/cclsp --agent claude-code
Codex
$ npx skills add ktnyt/cclsp -a codex
Cursor
$ npx skills add ktnyt/cclsp -a cursor
Gemini CLI
$ npx skills add ktnyt/cclsp -a gemini-cli
GitHub Copilot
$ npx skills add ktnyt/cclsp -a github-copilot
OpenCode
$ npx skills add ktnyt/cclsp -a opencode
Windsurf
$ npx skills add ktnyt/cclsp -a windsurf
cclsp is a Model Context Protocol (MCP) server that seamlessly integrates LLM-based coding agents with Language Server Protocol (LSP) servers. LLM-based coding agents often struggle with providing accurate line/column numbers, which makes naive attempts to integrate with LSP servers fragile and frustrating. cclsp solves this by intelligently trying multiple position combinations and providing robust symbol resolution that just works, no matter how your AI assistant counts lines.
find_definitionfind_referencesrename_symbolrename_symbol_strictget_diagnosticsrestart_serverWhen using AI-powered coding assistants like Claude, you often need to navigate codebases to understand symbol relationships. cclsp bridges the gap between Language Server Protocol capabilities and Model Context Protocol, enabling:
cclsp provides an interactive setup wizard that automates the entire configuration process. Choose your preferred method:
Run the interactive setup wizard:
# One-time setup (no installation required)
npx cclsp@latest setup
# For user-wide configuration
npx cclsp@latest setup --user
The setup wizard will:
.claude/cclsp.json in current directory--user): Creates global config in ~/.config/claude/cclsp.jsonIf you prefer manual configuration:
Install cclsp:
npm install -g cclsp
Install language servers (see Language Server Installation)
Create configuration file:
# Use the interactive generator
cclsp setup
# Or create manually (see Configuration section)
Add to Claude MCP:
claude mcp add cclsp npx cclsp@latest --env CCLSP_CONFIG_PATH=/path/to/cclsp.json
The setup wizard shows installation commands for each LSP, but you can also install them manually:
📦 Common Language Servers
npm install -g typescript-language-server typescript
pip install "python-lsp-server[all]"
# Or basic installation: pip install python-lsp-server
go install golang.org/x/tools/gopls@latest
rustup component add rust-analyzer
rustup component add rust-src
# Ubuntu/Debian
sudo apt install clangd
# macOS
brew install llvm
# Windows: Download from LLVM releases
gem install solargraph
npm install -g intelephense
For more languages and detailed instructions, run npx cclsp@latest setup and select "Show detailed installation guides".
Configure in your MCP client (e.g., Claude Code):
{
"mcpServers": {
"cclsp": {
"command": "cclsp",
"env": {
"CCLSP_CONFIG_PATH": "/path/to/your/cclsp.json"
}
}
}
}
{
"mcpServers": {
"cclsp": {
"command": "node",
"args": ["/path/to/cclsp/dist/index.js"],
"env": {
"CCLSP_CONFIG_PATH": "/path/to/your/cclsp.json"
}
}
}
}
For easy setup, use the interactive configuration generator:
# Using npx (recommended for one-time setup)
npx cclsp@latest setup
# If installed globally
cclsp setup
# Or run directly with the development version
bun run setup
The interactive tool will:
Alternatively, create an cclsp.json configuration file manually:
{
"servers": [
{
"extensions": ["py", "pyi"],
"command": ["uvx", "--from", "python-lsp-server", "pylsp"],
"rootDir": ".",
"initializationOptions": {
"settings": {
"pylsp": {
"plugins": {
"jedi_completion": { "enabled": true },
"jedi_definition": { "enabled": true },
"jedi_hover": { "enabled": true },
"jedi_references": { "enabled": true },
"jedi_signature_help": { "enabled": true },
"jedi_symbols": { "enabled": true },
"pylint": { "enabled": false },
"pycodestyle": { "enabled": false },
"pyflakes": { "enabled": false }
}
}
}
}
},
{
"extensions": ["js", "ts", "jsx", "tsx"],
"command": ["npx", "--", "typescript-language-server", "--stdio"],
"rootDir": "."
}
]
}
Configuration Options:
extensions: Array of file extensions this server handlescommand: Command array to spawn the LSP serverrootDir: Working directory for the LSP server (optional, defaults to ".")restartInterval: Auto-restart interval in minutes (optional)initializationOptions: LSP server initialization options (optional)The initializationOptions field allows you to customize how each LSP server initializes. This is particularly useful for servers like pylsp (Python) that have extensive plugin configurations, or servers like devsense-php-ls that require specific settings.
📋 More Language Server Examples
{
"servers": [
{
"extensions": ["go"],
"command": ["gopls"],
"rootDir": "."
},
{
"extensions": ["rs"],
"command": ["rust-analyzer"],
"rootDir": "."
},
{
"extensions": ["c", "cpp", "cc", "h", "hpp"],
"command": ["clangd"],
"rootDir": "."
},
{
"extensions": ["java"],
"command": ["jdtls"],
"rootDir": "."
},
{
"extensions": ["rb"],
"command": ["solargraph", "stdio"],
"rootDir": "."
},
{
"extensions": ["php"],
"command": ["intelephense", "--stdio"],
"rootDir": "."
},
{
"extensions": ["cs"],
"command": ["omnisharp", "-lsp"],
"rootDir": "."
},
{
"extensions": ["swift"],
"command": ["sourcekit-lsp"],
"rootDir": "."
}
]
}
# Run in development mode
bun run dev
# Run tests
bun test
# Run manual integration test
bun run test:manual
# Lint code
bun run lint
# Format code
bun run format
# Type check
bun run typecheck
The server exposes these MCP tools:
find_definitionFind the definition of a symbol by name and kind in a file. Returns definitions for all matching symbols.
Parameters:
file_path: The path to the filesymbol_name: The name of the symbolsymbol_kind: The kind of symbol (function, class, variable, method, etc.) (optional)find_referencesFind all references to a symbol across the entire workspace. Returns references for all matching symbols.
Parameters:
file_path: The path to the file where the symbol is definedsymbol_name: The name of the symbolsymbol_kind: The kind of symbol (function, class, variable, method, etc.) (optional)include_declaration: Whether to include the declaration (optional, default: true)rename_symbolRename a symbol by name and kind in a file. This tool now applies the rename to all affected files by default. If multiple symbols match, returns candidate positions and suggests using rename_symbol_strict.
Parameters:
file_path: The path to the filesymbol_name: The name of the symbolsymbol_kind: The kind of symbol (function, class, variable, method, etc.) (optional)new_name: The new name for the symboldry_run: If true, only preview the changes without applying them (optional, default: false)Note: When dry_run is false (default), the tool will:
.bak extensionrename_symbol_strictRename a symbol at a specific position in a file. Use this when rename_symbol returns multiple candidates. This tool now applies the rename to all affected files by default.
Parameters:
file_path: The path to the fileline: The line number (1-indexed)Stars on GitHub
A development methodology for coding agents that makes the agent brainstorm before building, follow plans, run parallel work and finish branches in a disciplined way. You get fewer skipped steps and more consistent results from your coding assistant.
@obraHelps you take a change from a rough idea or bug report through to a tested, reviewed result, with each step handled by a focused skill. Its skills cover pinning down domain terms, designing module interfaces, diagnosing hard bugs, and reviewing changes against your repo's coding standards.
@mattpocockGives your coding agent a planning, testing and review routine to follow on every task. It plans before building, checks changes with tests, has a fresh reviewer look for regressions, and saves repeated wins as reusable skills.
@affaan-mDeveloper tools for working on the React codebase itself. Provides skills for extracting error messages, managing and debugging feature flags across release channels, running Flow type checks, and fixing lint and formatting issues before committing.
@facebookLets your AI coding assistant jump to definitions, find every usage of a symbol, and rename across a codebase, using the language servers you already run. It works even when the assistant gets line and column numbers wrong. Install it with the command on its page. It includes architecture, hands-on-test, playwright-cli. Its skills do not fire on their own yet. Free and open source.
Is this plugin yours?
Claim it with GitHubSubmit a pluginPromote it