Featured

Deploy OpenClaw in 60 seconds — 20% off logoDeploy OpenClaw in 60 seconds — 20% off

Launch OpenClaw on Hostinger in about 60 seconds and keep your agent live 24/7. Our referral link gives you 20% off, no coupon code needed.

Launch on Hostinger
Run your Hermes agent on Hostinger, fully managed logoRun your Hermes agent on Hostinger, fully managed

Launch Hermes on Hostinger in one click, fully managed, no VPS knowledge needed. Use code ZACAARON10 for 10% off.

Launch on Hostinger
Crawl and scrape any site into clean data, 10% off logoCrawl and scrape any site into clean data, 10% off

Firecrawl crawls and scrapes any site into clean markdown for your agent. Get 1,000 free credits, and new users get 10% off their first purchase.

Try Firecrawl free
6,000+ web scrapers for your AI agent, start free logo6,000+ web scrapers for your AI agent, start free

Apify gives your agent live web data: 6,000+ prebuilt scrapers and actors, MCP-ready. Sign up free with $5 in usage credits.

Try Apify free
One API to scrape, enrich, and extract the internet. logoOne API to scrape, enrich, and extract the internet.

Context.dev gives your agents a single API to scrape, enrich, and extract live web data — no proxies, no parsers, no maintenance.

Start building free
SetupClaw: done-for-you OpenClaw for founders & exec teams logoSetupClaw: done-for-you OpenClaw for founders & exec teams

White-glove OpenClaw for founders and exec teams (4–50+ employees): we install, harden, integrate your tools, and maintain it — secured from day one.

Get it set up for you
SEO data APIs for your agent, $1 free credit logoSEO data APIs for your agent, $1 free credit

DataForSEO gives your agent live access to SERP results, keyword data, backlinks, and on-page SEO data through one API. New accounts get a $1 credit, good for up to 20,000 keyword or backlink lookups.

Try DataForSEO free
Reach 48,000+ AI builders

A flat monthly placement in front of developers actively installing AI tools. No lock-in, cancel anytime.

Advertise here

Works with

Claude CodeClaude DesktopCursorVS CodeClineCodex CLIOpenClaw+ any MCP client

Install to Claude Code

This server doesn't publish a one-line install command. Follow the setup in the source repository.

Summary

MCP server providing flight search, web search, URL fetching with JS rendering, and documentation search tools.

README.md

Bhinneka MCP Toolkit

A FastMCP server that provides Google Flights utilities, SearXNG web search, safe URL fetching with optional JS rendering, and Context7 documentation search — all via the Model Context Protocol (MCP).

Features

  • Flight Search: Find flights between airports with flexible search criteria
  • Airport Lookup: Search airports by code, name, or city with comprehensive database
  • Price Optimization: Get cheapest flight options sorted by price
  • Best Flights: Access Google's recommended flight selections
  • SearXNG Search: Web/news/image search via a JSON SearXNG backend
  • URL Fetch: Safe static fetch (HTML-to-text) + optional Playwright JS rendering for SPAs
  • Context7 Docs: Search libraries and retrieve documentation text
  • Multiple Transports: Support for stdio (development) and HTTP (production) modes
  • CLI Interface: Built-in testing and management commands

Installation

Using uvx (Recommended)

# Install and run directly with uvx
uvx bhinneka --help

Using pip

pip install bhinneka

Using Docker (For Remote Deployment)

# Build and run with Docker
docker build -t bhinneka .
docker run -p 8000:8000 bhinneka

# Or use docker-compose for easier management
docker-compose up

Quick Start with uvx

MCP Server Modes

# Default mode (stdio transport for Claude Desktop and MCP clients)
uvx bhinneka serve

# HTTP mode (for remote access)
uvx bhinneka serve --transport http --host 0.0.0.0 --port 8000

# Check server status and capabilities
uvx bhinneka status

Testing Commands

# Test flight search
uvx bhinneka test-search LAX JFK 2025-12-25

# Test airport search
uvx bhinneka test-airports "Los Angeles"

# Show version information
uvx bhinneka version

Short Command Alias

The package also provides short aliases: bn (new) and legacy gf.

# All commands work with the shorter alias
uvx bhinneka serve
# is equivalent to
uvx --from bhinneka bn serve

MCP Integration

With MCP Inspector

For development and testing with MCP Inspector:

uvx bhinneka serve

Then connect your MCP client to the stdio transport.

HTTP Server Mode

For production use or remote MCP clients:

uvx bhinneka serve --transport http --host 0.0.0.0 --port 8000

MCP clients can connect to: http://your-server:8000/mcp

Authentication for HTTP deployments: see AUTH.md for Google OAuth setup and email/domain allowlisting.

With Claude Desktop

To use this MCP server with Claude Desktop, add the following configuration to your Claude Desktop settings:

{
  "mcpServers": {
    "bhinneka": {
      "command": "uvx",
      "args": [
        "bhinneka",
        "serve"
      ]
    }
  }
}

This configuration will:

  • Automatically install the latest version using uvx
  • Run the server with stdio transport (default for Claude Desktop)
  • Make flight search tools available in Claude Desktop

After adding this configuration, restart Claude Desktop and you'll have access to all the flight search tools directly in your conversations.

Alternative Configuration with HTTP Transport

For advanced users who prefer HTTP transport for remote access:

{
  "mcpServers": {
    "bhinneka": {
      "command": "uvx",
      "args": [
        "bhinneka",
        "serve",
        "--transport",
        "http",
        "--host",
        "127.0.0.1",
        "--port",
        "8000"
      ]
    }
  }
}

Available MCP Tools

flights_search

Search for flights between two airports.

Parameters:

  • origin (string): Origin airport code (e.g., "LAX")
  • destination (string): Destination airport code (e.g., "JFK")
  • departure_date (string): Departure date in YYYY-MM-DD format
  • return_date (string, optional): Return date for round-trip flights
  • seat_class (string): Seat class - "economy", "premium_economy", "business", "first"
  • adults (integer): Number of adult passengers (default: 1)
  • children (integer, optional): Number of child passengers
  • infants (integer, optional): Number of infant passengers
  • max_results (integer): Maximum results to return (default: 6)

flights_find_airports

Search for airports by code, name, or city.

Parameters:

  • query (string): Search query (airport code, name, or city)
  • limit (integer): Maximum results to return (default: 10)

flights_get_cheapest

Get flights sorted by price (lowest first).

Parameters: Same as search_flights

flights_get_best

Get Google's recommended "best" flights balancing price, duration, and convenience.

Parameters: Same as search_flights

get_server_status

Get server status and capabilities information.

Parameters: None

SearXNG Web Search Tools

These tools require SEARXNG_BASE_URL to be set (e.g., https://your-searxng.example).

  • searx_web_search(query, engines?, language?, time_range?, safesearch?, max_results?)

General web search. Returns a concise, readable list.

  • searx_images_search(query, engines?, language?, safesearch?, max_results?)

Image-focused results; includes image URLs where available.

  • searx_news_search(query, engines?, language?, time_range?, max_results?)

News category search with optional time range.

  • searx_search_json(query, category?, engines?, language?, time_range?, safesearch?, max_results?)

Returns compact JSON string for programmatic use.

URL Fetch Tools

Fetch web pages safely with optional JS rendering.

  • fetch_url(url, text_only=true, render_js=false, timeout=120, max_bytes=2000000, follow_redirects=true, extract_links=false, return_json=false)

Static fetch by default. Blocks localhost/private networks. Returns readable text with title/description/lang. Add extract_links=true to include absolute anchors.

  • fetch_url_rendered(url, text_only=true, timeout=120, max_bytes=2000000, follow_redirects=true, extract_links=false, return_json=false)

Same as fetch_url but with render_js=true for SPA pages (uses Playwright/Chromium). Heavier and slower; use only when needed.

Notes:

  • With text_only=true (default), HTML/CSS/JS are stripped.
  • SPAs often require render_js=true to capture dynamic content.
  • Non-HTML types (JSON, text/plain) are returned as pretty text.

Context7 Documentation Tools

Search libraries and retrieve documentation text from Context7.

  • context7_search(query, client_ip?, api_key?, return_json=false)

Searches Context7 libraries by query. Returns a readable list or JSON when return_json=true.

  • context7_fetch(library_id, tokens?, topic?, type_hint?, client_ip?, api_key?)

Fetches documentation for a specific library ID. Defaults to type=txt. Returns the documentation text or an error message if unavailable.

Context7 Documentation Tools

Search libraries and retrieve documentation text from Context7.

  • context7_search(query, client_ip?, api_key?, return_json=false)

Searches Context7 libraries by query. Returns a readable list or JSON when return_json=true.

  • context7_fetch(library_id, tokens?, topic?, type_hint?, client_ip?, api_key?)

Fetches documentation for a specific library ID. Defaults to type=txt. Returns the documentation text or an error message if unavailable.

Environment variables:

  • CONTEXT7_BASE_URL (default: https://context7.com/api)
  • CONTEXT7_API_KEY (optional): Bearer token for private access
  • CONTEXT7_TIMEOUT (seconds, default: 15)
  • CLIENT_IP_ENCRYPTION_KEY (optional, hex-64). If present, client IP may be included (unencrypted fallback if encryption not available).

Notes:

  • You can pass client_ip and api_key per-call; env vars are used by default.
  • The tools handle rate limits (429), unauthorized (401), and not found (404) with clear messages.

Development Setup

Requirements

  • Python 3.8+
  • rye (recommended) or pip

Setup with rye

# Clone the repository
git clone https://github.com/tistaharahap/google-flights-mcp
cd google-flights-mcp

# Install dependencies
rye sync

# Run the MCP server
rye run python -m bhinneka.cli serve

# Run tests
rye run python -m bhinneka.cli test-search LAX JFK 2025-12-25

# Code formatting and linting
rye run ruff check
rye run ruff format

Configuration

Transport Options

  • stdio: For MCP Inspector and direct client integration
  • streamable-http: For remote access and production deployments

Environment Variables by Tool

General

  • PLAYWRIGHT_BROWSERS_PATH (Docker sets /ms-playwright)

Flights (flights_*)

  • No keys required. Uses public data via fast-flights.

SearXNG (searx_*)

  • SEARXNG_BASE_URL (required), e.g., https://your-searxng.example
  • SEARXNG_TIMEOUT (seconds, default: 8)
  • SEARXNG_MAX_RESULTS (default: 10)
  • SEARXNG_LANGUAGE (default: en)

URL Fetch (fetch_url, fetch_url_rendered)

  • No environment variables required. JS rendering requires Playwright + Chromium installed (Dockerfile handles this).

Context7 (context7_*)

  • CONTEXT7_BASE_URL (default: https://context7.com/api)
  • CONTEXT7_API_KEY (optional): Bearer token for private access
  • CONTEXT7_TIMEOUT (seconds, default: 15)
  • CLIENT_IP_ENCRYPTION_KEY (optional, hex-64). If present, client IP may be included (encryption currently not applied; plaintext header is sent when provided).

Docker Deployment

Building and Running with Docker

# Build the Docker image
docker build -t bhinneka .

# Run the container
docker run -d \
  --name bhinneka \
  -p 8000:8000 \
  --restart unless-stopped \
  bhinneka

# Check container logs
docker logs bhinneka

# Stop the container
docker stop bhinneka

Using Docker Compose (Recommended)

The repository includes a docker-compose.yml file for easier deployment:

# Start the service
docker-compose up -d

# View logs
docker-compose logs -f

# Stop the service
docker-compose down

# Rebuild and restart
docker-compose up --build -d

Production Deployment with Nginx

For production deployments, use the nginx profile for reverse proxy and SSL termination:

# Start with nginx reverse proxy
docker-compose --profile production up -d

# This will start both the MCP server and nginx
# Configure SSL certificates in ./ssl/ directory
# Customize nginx.conf for your domain and SSL setup

Docker Configuration

The Docker setup includes:

  • Multi-stage build for optimized image size
  • Non-root user for enhanced security
  • Health checks for container monitoring
  • HTTP transport configured for remote access
  • Port 8000 exposed for MCP client connections

Environment Variables

You can customize the deployment using environment variables:

# Custom host and port
docker run -p 9000:9000 \
  -e GOOGLE_FLIGHTS_HOST=0.0.0.0 \
  -e GOOGLE_FLIGHTS_PORT=9000 \
  bhinneka

# Or in docker-compose.yml:
environment:
  - GOOGLE_FLIGHTS_HOST=0.0.0.0
  - GOOGLE_FLIGHTS_PORT=8000

Connecting to Dockerized MCP Server

When running in Docker, the server is accessible via HTTP transport:

  • Local access: http://localhost:8000/mcp
  • Remote access: http://your-server-ip:8000/mcp

Configure your MCP client to connect to the HTTP endpoint instead of stdio transport.

Data Source

This server uses the fast-flights library to access Google Flights data. The library provides real-time flight information without requiring API keys.

License

[Add your license information here]

See related servers & alternatives →

Related MCP servers

Browse all →

Related guides

Hand-picked reading to help you choose and use Search servers.