Dependency Context
About
An MCP server that provides AI assistants with contextual access to your project's dependency documentation, enabling more accurate responses about libraries and frameworks used in your codebase.
Details
- Author
- Dsinghbailey
- GitHub stars
- 2
- Downloads
- 131
- Categories
- Other
Jump to
- Indexes dependency documentation from GitHub repositories
- Provides semantic search over indexed documentation
- Supports custom dependency configuration via dependency-context.json
- Falls back to package.json or requirements.txt automatically
- Includes both MCP tools and a standalone CLI mode
- Uses local vector embeddings for offline search
Setting up with Highlight
This MCP is not yet compatible with Highlight’s one-click setup. However, you can still use it with Highlight by following these steps:
- Download and install Highlight from highlightai.com/download
- Navigate to the plugins tab and select "Add Custom Plugin"
-
Configure the plugin with the settings below
Plugin Name
Dependency ContextCommand (node, npx, python, etc.)Please refer to the README for specific instructions on how to obtain API keys or other required environment variables.
- Enable "Start Automatically" if you want the plugin to start when Highlight launches
From the repository
Configure by creating a dependency-context.json file in your project root listing the dependencies to index. If none is present, it falls back to scanning package.json or requirements.txt. Add the MCP config to your editor (Cursor recommended), then prompt the AI to initialize dependency-context. A CLI mode is also available via npx or global install for direct documentation download.
Claude Desktop / Cursor
Paste into your MCP client config file to install this server.
{
"mcpServers": {
"dependency context": {
"dependency-context": {
"command": "npx",
"args": [
"-y",
"--package=dependency-context",
"dependency-context"
],
"env": {
"GITHUB_TOKEN": "YOUR_GITHUB_TOKEN_HERE",
"MODEL_NAME": "Xenova/all-MiniLM-L6-v2",
"DEBUG": "false",
"MIN_CHUNK_SIZE": "800",
"MAX_CHUNK_SIZE": "8000",
"CHUNKS_RETURNED": "5"
}
}
}
}
}McpServers
{
"dependency-context": {
"command": "npx",
"args": [
"-y",
"--package=dependency-context",
"dependency-context"
],
"env": {
"GITHUB_TOKEN": "YOUR_GITHUB_TOKEN_HERE",
"MODEL_NAME": "Xenova/all-MiniLM-L6-v2",
"DEBUG": "false",
"MIN_CHUNK_SIZE": "800",
"MAX_CHUNK_SIZE": "8000",
"CHUNKS_RETURNED": "5"
}
}
}Dependency Context
by @darianb
An MCP server and command line tool that provides AI assistants with contextual access to your project's dependency documentation, enabling more accurate responses about libraries and frameworks used in your codebase.
Configuration
The recommended way to specify which dependencies you want to index is by creating a custom dependency-context.json file in your project root. This allows you to:
- Only index the dependencies you're actively using and need help with
- Improve indexing speed by limiting the number of dependencies
- Focus the search results on the libraries that matter most
Create a dependency-context.json file in your project root with the following format:
{
"express": "^4.17.1",
"axios": "1.0.0"
}
If a dependency-context.json file is not present, Dependency Context falls back to scanning package.json or requirements.txt.
Quick Start
Option 1: Use MCP with AI Integration (Recommended for Cursor Users)
1. Add the MCP config to your editor (Cursor recommended):
{
"mcpServers": {
"dependency-context": {
"command": "npx",
"args": ["-y", "--package=dependency-context", "dependency-context"],
"env": {
"GITHUB_TOKEN": "YOUR_GITHUB_TOKEN_HERE", // Optional but recommended
"MODEL_NAME": "Xenova/all-MiniLM-L6-v2", // Optional, default shown
"DEBUG": "false", // Optional, default shown
"MIN_CHUNK_SIZE": "800", // Optional, default shown
"MAX_CHUNK_SIZE": "8000", // Optional, default shown
"CHUNKS_RETURNED": "5" // Optional, default shown
}
}
}
}
2. Enable the MCP in your editor
3. Prompt the AI to initialialize dependency-context. Make sure you are in "Agent" mode.
Can you initialize dependency-context?
4. Prompt the AI like you normally do. It will automatically pull in dependency-context when relevant
Option 2: Direct Download of Documentation (No Vector Database)
If you prefer to just download and browse the dependency documentation without vector search:
1. Install dependency-context:
npm install -g dependency-context
2. Download documentation for a project:
```bash
Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.
