Contextplus

v1.0.0Coding Agentsstable

Semantic Intelligence for Large-Scale Engineering. Context+ is an MCP server designed for developers who demand 99% accuracy. By combining RAG, Tree-sitter AST, Spectral Clustering, and Obsidian-style linking, Context+ turns a massive codebase into a

mcp-server
Share:
1,902
Stars
0
Downloads
0
Weekly
0/5

What is Contextplus?

Contextplus is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to semantic intelligence for large-scale engineering. context+ is an mcp server designed for developers who demand 99% accuracy. by combining rag, tree-sitter ast, spectral clustering, and obsidian-style...

Semantic Intelligence for Large-Scale Engineering. Context+ is an MCP server designed for developers who demand 99% accuracy. By combining RAG, Tree-sitter AST, Spectral Clustering, and Obsidian-style linking, Context+ turns a massive codebase into a

This server falls under the Coding Agents category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • Semantic Intelligence for Large-Scale Engineering. Context+

Use Cases

Analyze massive codebases with 99% accuracy using advanced AST and clustering.
Build semantic intelligence graphs for complex engineering projects.
forloopcodes

Maintainer

LicenseMIT
Languagetypescript
Versionv1.0.0
UpdatedMay 21, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

Manual Installation

npx contextplus

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use Contextplus

Context+ is a semantic intelligence MCP server for large-scale engineering codebases that combines RAG (Retrieval-Augmented Generation), Tree-sitter AST parsing, spectral clustering, and Obsidian-style linking to give AI assistants deep, accurate understanding of code structure. It exposes tools for semantic code search, blast radius analysis (what changes when you edit a function), static analysis, and a persistent memory graph with decay scoring — all without requiring cloud services. Developers working on large repositories use it to navigate unfamiliar code, plan refactors, and maintain context across long-running agent sessions.

Prerequisites

  • Node.js 18+ and npm (or Bun) installed
  • An MCP-compatible client such as Claude Desktop or Claude Code
  • Ollama running locally with a model like nomic-embed-text (for local embeddings), OR an OpenAI-compatible API key for cloud embeddings
  • A codebase directory to analyze — Context+ supports 43 programming languages via Tree-sitter
  • Sufficient disk space for the local vector index (scales with codebase size)
1

Run Context+ directly with npx

Context+ can be started without a global install using npx. This automatically downloads and runs the latest version.

npx -y contextplus
2

Generate IDE configuration

Use the init command to automatically write the correct MCP config for your editor. This produces the correct JSON for Claude Desktop or Claude Code.

npx -y contextplus init claude
3

Configure the embedding provider

Set environment variables to choose your embedding backend. For local embeddings use Ollama; for cloud embeddings use an OpenAI-compatible provider.

# For Ollama (local, no API key needed):
export CONTEXTPLUS_EMBED_PROVIDER=ollama
export OLLAMA_EMBED_MODEL=nomic-embed-text

# For OpenAI-compatible:
export CONTEXTPLUS_EMBED_PROVIDER=openai
export CONTEXTPLUS_OPENAI_API_KEY=sk-...
export CONTEXTPLUS_OPENAI_BASE_URL=https://api.openai.com/v1
export CONTEXTPLUS_OPENAI_EMBED_MODEL=text-embedding-3-small
4

Point Context+ at your codebase

Set the working directory in the MCP config args to the root of the codebase you want to analyze. Context+ will index the repository on first run.

5

Add Context+ to your MCP client configuration

Copy the generated configuration into your claude_desktop_config.json or Claude Code settings, including the embedding environment variables.

Contextplus Examples

Client configuration

Add this block to claude_desktop_config.json. Set CONTEXTPLUS_EMBED_PROVIDER and the appropriate model variables for your chosen embedding backend.

{
  "mcpServers": {
    "contextplus": {
      "command": "npx",
      "args": ["-y", "contextplus"],
      "env": {
        "CONTEXTPLUS_EMBED_PROVIDER": "ollama",
        "OLLAMA_EMBED_MODEL": "nomic-embed-text"
      }
    }
  }
}

Prompts to try

Once Context+ is running and has indexed your codebase, use these prompts to navigate and understand it.

- "Search for all functions that handle user authentication in this codebase"
- "What is the blast radius if I change the signature of the processPayment function?"
- "Give me a skeleton of the UserService class"
- "Find all identifiers related to database connection pooling"
- "Run static analysis on the src/api directory and list potential issues"
- "Create a restore point before I start this refactor"

Troubleshooting Contextplus

Semantic search returns no results after indexing

Verify the embedding provider is configured correctly. For Ollama, run 'ollama list' to confirm the embed model is pulled (e.g., 'ollama pull nomic-embed-text'). For OpenAI, check that CONTEXTPLUS_OPENAI_API_KEY is set and valid.

Context+ is slow to start on a large codebase

The first run builds the AST index and vector embeddings for every file, which takes time proportional to codebase size. Subsequent starts are faster as the index is cached. Exclude large generated directories (node_modules, .git, dist) if possible.

npx contextplus fails with a network error

Check your internet connection and npm registry access. Try 'npm cache clean --force' then retry. Alternatively, install globally with 'npm install -g contextplus' and run 'contextplus' directly.

Frequently Asked Questions about Contextplus

What is Contextplus?

Contextplus is a Model Context Protocol (MCP) server that semantic intelligence for large-scale engineering. context+ is an mcp server designed for developers who demand 99% accuracy. by combining rag, tree-sitter ast, spectral clustering, and obsidian-style linking, context+ turns a massive codebase into a It connects AI assistants to external tools and data sources through a standardized interface.

How do I install Contextplus?

Follow the installation instructions on the Contextplus GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.

Which AI clients work with Contextplus?

Contextplus works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is Contextplus free to use?

Yes, Contextplus is open source and available under the MIT license. You can use it freely in both personal and commercial projects.

Browse More Coding Agents MCP Servers

Explore all coding agents servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Quick Config Preview

{ "mcpServers": { "contextplus": { "command": "npx", "args": ["-y", "contextplus"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use Contextplus?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides