tracked-search
A drop-in replacement for Claude's built-in web_search tool that automatically tracks usage and provides multiple search backends.
README
MCP Tracked Search
A drop-in replacement for Claude's built-in web_search tool that automatically tracks usage and provides multiple search backends.
Features
- Drop-in replacement: Uses the same
web_searchtool name and parameters - Dual search engines: Brave (primary) and DuckDuckGo (fallback/free)
- Automatic tracking: Counts daily and monthly usage for Brave API
- Smart fallback: Automatically switches to DuckDuckGo when Brave limits are reached
- Rate limiting: Prevents exceeding Brave API limits (166/day, 5000/month)
- Enhanced search: Optional
tracked_searchtool with API selection and more options - Usage stats: Check current usage with
search_usagetool
Setup
- Install dependencies:
npm install
- Build:
npm run build
- Set your Brave API key:
export BRAVE_API_KEY="your-brave-api-key"
- Add to Claude Desktop config (
~/Library/Application Support/Claude/claude_desktop_config.json):
{
"mcpServers": {
"tracked-search": {
"command": "node",
"args": ["/Users/bard/Code/mcp-tracked-search/build/index.js"],
"env": {
"BRAVE_API_KEY": "your-brave-api-key"
}
}
}
}
Usage
Once installed, the tools will be available in Claude:
web_search (drop-in replacement)
<web_search>
<query>latest AI news</query>
</web_search>
tracked_search (enhanced)
<tracked_search>
<query>latest AI news</query>
<api>brave</api> <!-- or "duckduckgo" -->
<count>20</count>
<freshness>24h</freshness>
</tracked_search>
search_usage (check stats)
<search_usage />
How it works
- Intercepts search requests before they go to the search API
- For Brave: Checks usage against daily (166) and monthly (5000) limits
- If Brave limit reached, automatically falls back to DuckDuckGo (free, no limits)
- Performs the search with the selected API
- Returns results in the same format as built-in web_search
- Stores Brave usage data in
~/.config/mcp-tracked-search/usage.json
API Selection Logic
web_searchtool: Always tries Brave first, falls back to DuckDuckGo if limits reachedtracked_searchtool: Can explicitly choose API with theapiparameter- DuckDuckGo: Free, no API key needed, no rate limits (but be respectful)
Benefits
- Never exceed Brave API limits accidentally
- Track actual usage vs estimated 20/day
- Optimize search usage based on real data
- Automatic fallback to free DuckDuckGo when limits reached
- Choose the best search engine for each query
- Single tool for multiple search backends
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.