DeepContext

Provides symbol-aware semantic search for large codebases via MCP with index, search, status, and clear operations.

typescript

244

GitHub Stars

typescript

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

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

DeepContext MCP adds symbol-aware semantic search to coding agents, enabling precise context for large codebases by indexing code and returning highly relevant chunks for fast, accurate completions.

How to use

You connect your coding agent to the DeepContext MCP server to enable semantic search over your codebase. Once connected, you can index a codebase and then query using natural language or keywords to find relevant code chunks. The agent will present precise results that respect code structure and imports, improving accuracy and reducing token usage.

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

# Add to ~/.codex/config.toml
[mcp_servers.deepcontext]
command = "npx"
args = ["-y", "@wildcard-ai/deepcontext@latest"]
env = { "WILDCARD_API_KEY" = "your-wildcard-api-key" }

After you configure the MCP client, index your current directory to build a searchable semantic model of your codebase. Then you can perform queries that leverage semantic relationships within your project.

index this codebase

How to install

Prerequisites are minimal for using the DeepContext MCP with supported clients. You will need a Wildcard API key and access to a compatible MCP client (Claude Code or Codex). Follow these steps to get started.

Generate an API key on your Wildcard DeepContext page. 2) Copy the API key. 3) Choose your MCP client and run the appropriate installation command shown below. 4) Index your codebase using the indexing command.

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

Self hosting and notes

If you want to host DeepContext locally and connect it directly to a vector store and embedding provider, you can follow the self-hosting flow. This requires Node.js for the MCP integration and access to vector storage and embedding services.

git clone https://github.com/Wildcard-Official/deepcontext-mcp.git
cd deepcontext
npm install
npm run build

Integration with Claude Code or Codex for a local setup looks like this. You run the integration command and provide required API keys.

claude mcp add deepcontext-local \
  -e TURBOPUFFER_API_KEY=your-turbopuffer-key \
  -e JINA_API_KEY=your-jina-key \
  -- node /path/to/deepcontext/dist/standalone-mcp-integration.js

Security and considerations

Manage your API keys securely. Do not hard‑code keys into code or scripts. Use environment variable management practices and rotate keys as needed. When indexing large codebases, monitor resource usage and adjust indexing settings to balance performance and costs.

Available tools

index_codebase

Creates a searchable index of your codebase for semantic search.

search_codebase

Finds relevant code using natural language or keyword queries.

get_indexing_status

Shows indexing status and file counts for your codebases.

clear_index

Removes all indexed data for a codebase.