Context+

v0.2.3Coding Agentsstable

Context+ is a semantic intelligence server that transforms codebases into searchable, hierarchical feature graphs using RAG, Tree-sitter AST, and spectral clustering. It provides tools for deep code discovery, blast radius analysis, and memory graph

mcp-serverclaude-codecli-toolmcpmcp-tools
Share:
1,902
Stars
0
Downloads
0
Weekly
0/5

What is Context+?

Context+ is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to context+ is a semantic intelligence server that transforms codebases into searchable, hierarchical feature graphs using rag, tree-sitter ast, and spectral clustering. it provides tools for deep code d...

Context+ is a semantic intelligence server that transforms codebases into searchable, hierarchical feature graphs using RAG, Tree-sitter AST, and spectral clustering. It provides tools for deep code discovery, blast radius analysis, and memory graph

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

Features

  • Context+ is a semantic intelligence server that transforms c

Use Cases

Transform codebases into searchable, hierarchical feature graphs using RAG and AST parsing.
Perform deep code discovery and blast radius analysis.
Build memory graphs for enhanced code understanding.
ForLoopCodes

Maintainer

LicenseMIT License
Languagetypescript
Versionv0.2.3
UpdatedMay 21, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

NPM

npx -y @neuledge/context

Manual Installation

npx -y @neuledge/context

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 Context+

Context+ (@neuledge/context) is a semantic intelligence MCP server that transforms large codebases into searchable, hierarchical feature graphs by combining RAG retrieval, Tree-sitter AST parsing across 43 languages, and spectral clustering. It exposes tools for deep code discovery, blast radius analysis, static analysis, and a persistent knowledge graph with memory nodes and typed relation edges — giving AI coding assistants persistent, accurate context beyond a single conversation window. It is designed for developers and AI agents working on large or unfamiliar repositories where exhaustive file-by-file reading would exceed context limits.

Prerequisites

  • Node.js 18+ and npm installed
  • An MCP-compatible client such as Claude Desktop or Claude Code
  • Ollama running locally with the nomic-embed-text model (for local embeddings), OR an OpenAI-compatible API key
  • A codebase directory accessible on the local filesystem
  • No cloud account required for local embedding mode
1

Run with npx (no install required)

The @neuledge/context package can be run directly via npx without a global install. This is the recommended way to get started.

npx -y @neuledge/context
2

Generate the MCP configuration for Claude

Use the built-in init command to automatically produce the correct configuration snippet for Claude Desktop or Claude Code.

npx -y @neuledge/context init claude
3

Set embedding provider environment variables

Choose your embedding backend. Ollama runs locally with no API costs; the OpenAI provider uses the API but may give higher quality embeddings.

# Local (Ollama):
export CONTEXTPLUS_EMBED_PROVIDER=ollama
export OLLAMA_EMBED_MODEL=nomic-embed-text

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

Add to Claude Desktop configuration

Paste the generated configuration into claude_desktop_config.json, ensuring the environment variables for your chosen embedding backend are included in the env block.

5

Restart your MCP client

Restart Claude Desktop or Claude Code so it picks up the new server. On first activation, Context+ will index the target codebase.

Context+ Examples

Client configuration

Add this to claude_desktop_config.json. Adjust environment variables to match your embedding provider choice.

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

Prompts to try

After the server has indexed your codebase, use these prompts to explore and understand it.

- "Show me the context tree for the top-level features of this repository"
- "Semantically search for code related to rate limiting"
- "What is the blast radius of changing the database schema for the users table?"
- "Add a memory node noting that the auth module uses JWT with RS256 signing"
- "Retrieve all memory nodes related to API security across this project"
- "List available restore points before I start refactoring"

Troubleshooting Context+

Tools return empty results or indexing seems to have not run

On first use, Context+ must index the codebase which can take several minutes for large repos. Check terminal output for indexing progress. If embeddings are failing silently, verify the CONTEXTPLUS_EMBED_PROVIDER variable is set and the provider is reachable.

Ollama embedding fails with connection refused

Ensure Ollama is running ('ollama serve') and the model is downloaded ('ollama pull nomic-embed-text'). Ollama defaults to http://127.0.0.1:11434 — check with 'curl http://127.0.0.1:11434/api/tags'.

npx -y @neuledge/context fails to resolve the package

Check npm registry access: 'npm view @neuledge/context'. If the package is not found, try installing globally: 'npm install -g @neuledge/context' and run 'context' directly. Verify Node.js 18+ with 'node --version'.

Frequently Asked Questions about Context+

What is Context+?

Context+ is a Model Context Protocol (MCP) server that context+ is a semantic intelligence server that transforms codebases into searchable, hierarchical feature graphs using rag, tree-sitter ast, and spectral clustering. it provides tools for deep code discovery, blast radius analysis, and memory graph It connects AI assistants to external tools and data sources through a standardized interface.

How do I install Context+?

Install via npm with the command: npx -y @neuledge/context. Then add the server configuration to your AI client's JSON config file (e.g., claude_desktop_config.json or .cursor/mcp.json).

Which AI clients work with Context+?

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

Is Context+ free to use?

Yes, Context+ is open source and available under the MIT License 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": { "context": { "command": "npx", "args": ["-y", "@neuledge/context"] } } }

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

Read the full setup guide →

Ready to use Context+?

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