Deepcontext

v1.0.0Developer Toolsstable

<tool> <purpose>Prepares a codebase for intelligent search by creating a searchable index</purpose> <when_to_use> <scenario>Call this first before searching any new codebase</scenario> <s

aiai-agentsclaudeclaude-codecode
Share:
277
Stars
0
Downloads
0
Weekly
0/5

What is Deepcontext?

Deepcontext is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to <tool> <purpose>prepares a codebase for intelligent search by creating a searchable index</purpose> <when_to_use> <scenario>call this first before searching any new codebase</scenario> <s

<tool> <purpose>Prepares a codebase for intelligent search by creating a searchable index</purpose> <when_to_use> <scenario>Call this first before searching any new codebase</scenario> <s

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

Features

  • index_codebase
  • search_codebase
  • get_indexing_status
  • clear_index

Use Cases

Create searchable codebase indexes
Enable intelligent code search
Prepare codebases for AI analysis
LicenseApache-2.0
Languagetypescript
Versionv1.0.0
UpdatedMay 21, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

NPM

npx -y @wildcard-ai/deepcontext

Manual Installation

npx -y @wildcard-ai/deepcontext

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 Deepcontext

DeepContext is an MCP server that creates semantic search indexes of codebases, enabling AI agents to find relevant code by meaning rather than by exact text matching. It uses Tree-sitter to parse TypeScript, Python, and JavaScript files, combines vector similarity search (via Turbopuffer) with BM25 full-text search, and applies Jina reranking to surface the most relevant results. Developers add it to Claude Code or Codex CLI so the agent can intelligently navigate large unfamiliar codebases — locating the right functions, modules, and patterns — without exhausting the context window by loading entire files.

Prerequisites

  • Node.js 18+ (for npx) or a local clone of the repository for self-hosted mode
  • A Wildcard API key from https://wild-card.ai/deepcontext (for cloud mode)
  • Self-hosted only: a Turbopuffer API key and a Jina AI API key
  • Claude Code, Codex CLI, or another MCP-compatible client
1

Get your Wildcard API key

Visit https://wild-card.ai/deepcontext and sign up to obtain a WILDCARD_API_KEY. This key is required for the cloud-hosted indexing and search service.

2

Add DeepContext to Claude Code

Run the Claude Code mcp add command to register the server. This stores the configuration in your project's .mcp.json file.

claude mcp add deepcontext \
  -e WILDCARD_API_KEY=your-wildcard-api-key \
  -- npx @wildcard-ai/deepcontext@latest
3

Add to Claude Desktop configuration

Alternatively, add DeepContext to your Claude Desktop config file for use across all conversations.

{
  "mcpServers": {
    "deepcontext": {
      "command": "npx",
      "args": ["-y", "@wildcard-ai/deepcontext@latest"],
      "env": {
        "WILDCARD_API_KEY": "your-wildcard-api-key"
      }
    }
  }
}
4

Index your codebase

Once the server is connected, ask Claude to index the current codebase. DeepContext will scan and index all TypeScript, Python, and JavaScript files while excluding test files, generated code, and build artifacts.

5

Monitor indexing progress

Use the get_indexing_status tool to check how many files have been indexed. Large codebases may take a few minutes.

6

Search by meaning

Once indexing completes, ask questions about the codebase in natural language. DeepContext's hybrid search finds code semantically matching your query.

Deepcontext Examples

Client configuration

Claude Desktop configuration for the DeepContext MCP server in cloud mode.

{
  "mcpServers": {
    "deepcontext": {
      "command": "npx",
      "args": ["-y", "@wildcard-ai/deepcontext@latest"],
      "env": {
        "WILDCARD_API_KEY": "your-wildcard-api-key"
      }
    }
  }
}

Prompts to try

Example prompts for using DeepContext once your codebase is indexed.

- "Index this codebase"
- "What is the indexing status?"
- "Find all functions that handle user authentication"
- "Show me the code responsible for database connection pooling"
- "Where is error handling implemented for API responses?"
- "Find any code related to rate limiting or throttling"
- "Clear the index and re-index from scratch"

Troubleshooting Deepcontext

WILDCARD_API_KEY is invalid or authentication fails

Double-check the key copied from https://wild-card.ai/deepcontext. Ensure there are no extra spaces in the environment variable value. Generate a new key from the dashboard if needed.

Indexing stalls or get_indexing_status shows 0 files after a long wait

Confirm the server has read access to the codebase directory. If the project is very large, DeepContext automatically excludes test and build directories — check that source files are not all inside an excluded path pattern.

Self-hosted mode fails with missing API keys

Self-hosted deployment requires both TURBOPUFFER_API_KEY (for vector storage) and JINA_API_KEY (for embeddings and reranking), in addition to the local server path. Set all three environment variables before starting the server.

Frequently Asked Questions about Deepcontext

What is Deepcontext?

Deepcontext is a Model Context Protocol (MCP) server that <tool> <purpose>prepares a codebase for intelligent search by creating a searchable index</purpose> <when_to_use> <scenario>call this first before searching any new codebase</scenario> <s It connects AI assistants to external tools and data sources through a standardized interface.

How do I install Deepcontext?

Install via npm with the command: npx -y @wildcard-ai/deepcontext. 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 Deepcontext?

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

Is Deepcontext free to use?

Yes, Deepcontext is open source and available under the Apache-2.0 license. You can use it freely in both personal and commercial projects.

Browse More Developer Tools MCP Servers

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

Quick Config Preview

{ "mcpServers": { "deepcontext": { "command": "npx", "args": ["-y", "@wildcard-ai/deepcontext"] } } }

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

Read the full setup guide →

Ready to use Deepcontext?

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