Nia Context MCP

Supercharge Cursor, Cline, Windsurf, and other AI coding assistants with codebase understanding and documentation search

Nia MCP to 10x any coding agent (Cursor, Cline, Windsurf, etc.)

• Get your API key at app.trynia.ai - 25 requests are free!
• Works with Cursor, Claude Desktop, Continue.dev, VS Code Cline, and any MCP-compatible tool

✨ What You Can Do

Transform your coding experience by giving your AI assistant superpowers:

Index GitHub Repos

Store entire repositories for instant code context and understanding

Index Documentation

Search across any website or documentation with natural language

Web Search & Discovery

Find trending repos, libraries, and documentation with AI-powered search

Deep Research Agent

Perform comprehensive research and analysis on complex technical topics

Chat with Entire Codebases

Talk to ENTIRE documentation and repos, not just individual pages (10x’ing Cursor!)

Multi-Source Intelligence

Get context from code, documentation, and web sources in one query

🎯 Real-World Use Cases

Index Additional Repos + Company Docs

Ask Cursor: “Use nia mcp to search across our backend repo and API docs, then come back and implement this feature”

Discover & Research Technologies

“Find the best React state management libraries, research their pros and cons, then index the top 2 for deeper analysis”

Create Personal Knowledge Base

Index all the frameworks you use - React, Next.js, TypeScript docs - and query them instantly

Search Multiple Codebases

Compare implementations across different projects or find patterns across your entire organization

Code + Documentation Context

Get context from both source code AND documentation in one unified search

🎥 See It In Action

Documentation Tools

📦 Installation

1. Prerequisites

   • Python 3.8+
   • uvx (recommended) or pipx or pip
   • Your NIA API key from app.trynia.ai
2. Install the Package

   Option A: Using uvx (Recommended)

   # uvx will automatically handle the installation when you run it
   # No need to pre-install if you have uvx
   

3. Configure Your MCP Client

   Choose your coding assistant below for specific configuration steps

4. Replace Your API Key

   Get your API key from app.trynia.ai and replace YOUR_API_KEY in the configuration

5. Restart Your Client

   Restart your coding assistant to load the new MCP server

⚙️ Configuration for Your Coding Assistant

Cursor

1. Open MCP Configuration

   Open your Cursor MCP configuration file:

   • macOS: ~/.cursor/mcp.json
   • Windows: %APPDATA%\Cursor\mcp.json
   • Linux: ~/.config/cursor/mcp.json
2. Add NIA Configuration

   Option A: Using uvx

   {
     "mcpServers": {
       "nia": {
         "command": "uvx",
         "args": ["nia-mcp-server"],
         "env": {
           "NIA_API_KEY": "YOUR_API_KEY",
           "NIA_API_URL": "https://apigcp.trynia.ai/"
         }
       }
     }
   }
   

🛠️ Available Tools

Repository Management

• index_repository - Index GitHub repositories for intelligent search
• list_repositories - List all indexed repositories with status
• check_repository_status - Monitor indexing progress
• delete_repository - Remove indexed repositories
• rename_repository - Rename repositories for better organization
• search_codebase - Search code with natural language queries

Documentation Management

• index_documentation - Index websites and documentation
• list_documentation - List all indexed documentation sources
• check_documentation_status - Monitor documentation indexing
• delete_documentation - Remove indexed documentation
• rename_documentation - Rename documentation sources
• search_documentation - Search docs with natural language

Web Search & Research

• nia_web_search - AI-powered search for repos, docs, and content
• nia_deep_research_agent - Deep multi-step research and analysis

Project Initialization

• initialize_project - Set up NIA-enabled projects with IDE configs
• Support for Cursor, VS Code, Claude, Windsurf, Cline, and more
• Automatic creation of IDE-specific rules and configurations

📋 Complete Tool Reference

Repository Management Tools

index_repository

Index a GitHub repository for intelligent code search.

Parameters:

• repo_url (str): GitHub repository URL (e.g., https://github.com/owner/repo)
• branch (str, optional): Branch to index (defaults to main branch)

Usage: When indexing starts, use check_repository_status to monitor progress.

list_repositories

List all indexed repositories with their status.

Parameters: None

Returns: List of repositories with status, branch, and indexing information.

check_repository_status

Check the indexing status of a repository.

Parameters:

• repository (str): Repository in owner/repo format

Returns: Current status, progress, and indexing details.

delete_repository

Delete an indexed repository.

Parameters:

• repository (str): Repository in owner/repo format

Returns: Confirmation of deletion.

rename_repository

Rename an indexed repository for better organization.

Parameters:

• repository (str): Repository in owner/repo format
• new_name (str): New display name (1-100 characters)

Returns: Confirmation of rename operation.

search_codebase

Search indexed repositories using natural language.

Parameters:

• query (str): Natural language search query (use comprehensive questions for best results)
• repositories (List[str], optional): List of repositories to search (owner/repo format)
• include_sources (bool, default=True): Whether to include source code in results

Returns: Search results with relevant code snippets and explanations.

Documentation Management Tools

index_documentation

Index documentation or website for intelligent search.

Parameters:

• url (str): URL of the documentation site to index
• url_patterns (List[str], optional): URL patterns to include in crawling (e.g., [“/docs/”, “/guide/”])
• max_age (int, optional): Maximum age of cached content in seconds
• only_main_content (bool, default=True): Extract only main content (removes navigation, ads, etc.)

Usage: When indexing starts, use check_documentation_status to monitor progress.

list_documentation

List all indexed documentation sources.

Parameters: None

Returns: List of documentation sources with status and metadata.

check_documentation_status

Check the indexing status of a documentation source.

Parameters:

• source_id (str): Documentation source ID

Returns: Current status, progress, and indexing details.

delete_documentation

Delete an indexed documentation source.

Parameters:

• source_id (str): Documentation source ID to delete

Returns: Confirmation of deletion.

rename_documentation

Rename a documentation source for better organization.

Parameters:

• source_id (str): Documentation source ID
• new_name (str): New display name (1-100 characters)

Returns: Confirmation of rename operation.

search_documentation

Search indexed documentation using natural language.

Parameters:

• query (str): Natural language search query (use comprehensive questions for best results)
• sources (List[str], optional): List of documentation source IDs to search
• include_sources (bool, default=True): Whether to include source references in results

Returns: Search results with relevant documentation excerpts.

Web Search & Research Tools

nia_web_search

Search repositories, documentation, and other content using AI-powered search.

Use Cases:

• Finding specific repos/docs/content
• Looking for examples or implementations
• Searching for what’s available on a topic
• Simple, direct searches that need quick results
• Finding similar content to a known URL

Parameters:

• query (str): Natural language search query
• num_results (int, default=5): Number of results to return (max: 10)
• category (str, optional): Filter by category (“github”, “company”, “research paper”, “news”, “tweet”, “pdf”)
• days_back (int, optional): Only show results from the last N days
• find_similar_to (str, optional): URL to find similar content to

Returns: Search results with actionable next steps.

nia_deep_research_agent

Perform deep, multi-step research on a topic using advanced AI research capabilities.

Use Cases:

• Comparing multiple options (“compare X vs Y vs Z”)
• Analyzing pros and cons
• Questions with “best”, “top”, “which is better”
• Needing structured analysis or synthesis
• Complex questions requiring multiple sources
• Questions about trends, patterns, or developments
• Requests for comprehensive overviews

Parameters:

• query (str): Research question (use comprehensive questions for best results)
• output_format (str, optional): Structure hint (e.g., “comparison table”, “pros and cons list”)

Returns: Comprehensive research results with citations.

Tool Selection Guide

Use nia_web_search for:

• “Find RAG libraries” → Simple search
• “What’s trending in Rust?” → Quick discovery
• “Show me repos like LangChain” → Similarity search

Use nia_deep_research_agent for:

• “Compare RAG vs GraphRAG approaches” → Comparative analysis
• “What are the best vector databases for production?” → Evaluation needed
• “Analyze the pros and cons of different LLM frameworks” → Structured analysis

Project Initialization Tools

initialize_project

Initialize a NIA-enabled project with IDE-specific rules and configurations.

Parameters:

• project_root (str): Absolute path to the project root directory
• profiles (List[str], optional): List of IDE profiles to set up (default: [“cursor”])

Supported Profiles: cursor, vscode, claude, windsurf, cline, codex, zed, jetbrains, neovim, sublime

Returns: Status of initialization with created files and next steps.

Examples:

• Basic: initialize_project("/path/to/project")
• Multiple IDEs: initialize_project("/path/to/project", profiles=["cursor", "vscode"])
• Specific IDE: initialize_project("/path/to/project", profiles=["windsurf"])

🎯 Pro Tips for Maximum Productivity

Discovery → Research → Index Workflow

1. Use nia_web_search to find relevant libraries: “Find popular React state management libraries”
2. Use nia_deep_research_agent to analyze: “Compare Redux vs Zustand vs Jotai with pros and cons”
3. Index the top choices for deep code analysis: “Index zustand repository”

Multi-Source Queries

Combine searches across repositories and documentation:

“Search both the Next.js repo and docs for server-side rendering patterns, then implement a similar solution for our project”

Context-Aware Development

Use specific, detailed queries:

“Find error handling patterns in our backend repo, compare with best practices from indexed documentation, then refactor this function”

Knowledge Base Building

Index frequently used resources:

• Framework documentation (React, Vue, Angular)
• Company internal docs
• Best practice guides
• Reference implementations

Research-Driven Development

Use the deep research agent for:

• Technology selection decisions
• Architecture pattern analysis
• Performance optimization strategies
• Security best practices research

⚠️ Troubleshooting

"No API key provided"

Solution: Make sure you’ve replaced YOUR_API_KEY with your actual API key from app.trynia.ai

Double-check the environment variable is correctly set in your MCP configuration.

"Command not found: nia-mcp-server"

Solution: Try installing with pip first:

pip install nia-mcp-server

Or ensure uvx is properly installed and in your PATH.

"Invalid API key"

Solution:

• Double-check your API key at app.trynia.ai
• Ensure there are no extra spaces or characters
• Verify the API endpoint is set to https://apigcp.trynia.ai/

"Rate limit exceeded"

Solution: You’ve used your 25 free requests. Check your usage and upgrade your plan at app.trynia.ai

MCP Server Not Responding

Solution:

1. Check if the server is running properly
2. Verify your MCP client configuration
3. Restart your coding assistant
4. Check the console for error messages

🚀 What’s Next?

Build Your Knowledge Base

Start by indexing your most-used repositories and documentation sources

Experiment with Queries

Try different search approaches to find what works best for your workflow

Share with Your Team

Set up shared knowledge bases for your development team

Optimize Your Workflow

Integrate Nia MCP into your daily coding routine for maximum productivity

💬 Feedback & Support

We’d love your feedback! Help us improve Nia MCP and make coding agents even more powerful. Share your experience, suggestions, or even roast the product if you want - we really want to help developers code way, way faster!

Nia - 10x your coding agent

---

Need Help? Join our community or reach out through app.trynia.ai for support and updates.