Earch MCP
Enables web search and local business data retrieval through the Brave Search API. Provides tools for web search, local points of interest, and rich search results with customizable parameters.
README
Earch MCP
The Universal MCP Server exposes tools for your workflows and is designed for prompt-first usage in MCP-compatible clients.
Installation
Prerequisites
- Node.js 18+
- Set
EARCH_MCP_API_KEYin your environment (orBRAVE_API_KEY)
Get an API key
- If your tools require an external API, obtain a key from the provider’s docs/console.
- Otherwise, you can skip this step.
Build locally
cd /path/to/earch-mcp
npm i
npm run build
Setup: Claude Code (CLI)
Use this one-liner (replace with your real values):
claude mcp add Earch MCP -s user -e EARCH_MCP_API_KEY="sk-your-real-key" -- npx earch-mcp
To remove:
claude mcp remove Earch MCP
Setup: Cursor
Create .cursor/mcp.json in your client (do not commit it here):
{
"mcpServers": {
"earch-mcp": {
"command": "npx",
"args": ["earch-mcp"],
"env": { "EARCH_MCP_API_KEY": "sk-your-real-key" },
"autoStart": true
}
}
}
Other Clients and Agents
<details> <summary>VS Code</summary>
Install via URI or CLI:
code --add-mcp '{"name":"earch-mcp","command":"npx","args":["earch-mcp"],"env":{"EARCH_MCP_API_KEY":"sk-your-real-key"}}'
</details>
<details> <summary>Claude Desktop</summary>
Follow the MCP install guide and reuse the standard config above.
</details>
<details> <summary>LM Studio</summary>
- Command: npx
- Args: ["earch-mcp"]
- Env: EARCH_MCP_API_KEY=sk-your-real-key
</details>
<details> <summary>Goose</summary>
- Type: STDIO
- Command: npx
- Args: earch-mcp
- Enabled: true
</details>
<details> <summary>opencode</summary>
Example ~/.config/opencode/opencode.json:
{
"$schema": "https://opencode.ai/config.json",
"mcp": {
"earch-mcp": {
"type": "local",
"command": ["npx", "earch-mcp"],
"enabled": true
}
}
}
</details>
<details> <summary>Qodo Gen</summary>
Add a new MCP and paste the standard JSON config.
</details>
<details> <summary>Windsurf</summary>
See docs and reuse the standard config above.
</details>
Setup: Codex (TOML)
Example (Serena):
[mcp_servers.serena]
command = "uvx"
args = ["--from", "git+https://github.com/oraios/serena", "serena", "start-mcp-server", "--context", "codex"]
This server (minimal):
[mcp_servers.earch-mcp]
command = "npx"
args = ["earch-mcp"]
# Optional environment variables:
# EARCH_MCP_API_KEY = "sk-your-real-key"
# MCP_NAME = "earch-mcp"
Configuration (Env)
- EARCH_MCP_API_KEY: Your API key (or BRAVE_API_KEY)
- MCP_NAME: Server name override (default: earch-mcp)
Available Tools
- web.search
- inputs: { q: string, count?: number (<=50), country?: string, safesearch?: "off"|"moderate"|"strict", freshness?: "pd"|"pw"|"pm"|"py" }
- outputs: JSON from Brave Web Search API
- local.pois
- inputs: { ids: string[1..20] }
- outputs: JSON from Brave Local POIs API
- local.descriptions
- inputs: { ids: string[1..20] }
- outputs: JSON from Brave Local Descriptions API
- web.rich
- inputs: { q: string }
- outputs: JSON containing hint and rich callback result
Example invocation (MCP tool call)
{
"tool": "web.search",
"arguments": { "q": "brave search" }
}
Troubleshooting
- 401 auth errors: check EARCH_MCP_API_KEY or BRAVE_API_KEY
- Ensure Node 18+
- Local runs: npx earch-mcp after npm run build
- Inspect publish artifacts: npm pack --dry-run
References
- MCP SDK: https://modelcontextprotocol.io/docs/sdks
- Architecture: https://modelcontextprotocol.io/docs/learn/architecture
- Server Concepts: https://modelcontextprotocol.io/docs/learn/server-concepts
- Server Spec: https://modelcontextprotocol.io/specification/2025-06-18/server/index
- Brave Search API: https://api.search.brave.com/res/v1/web/search
- Brave Local API: https://api.search.brave.com/res/v1/local/pois
- Brave Rich API: https://api.search.brave.com/res/v1/web/rich
Name Consistency & Troubleshooting
- Always use CANONICAL_ID (earch-mcp) for identifiers and keys.
- Use CANONICAL_DISPLAY (Earch MCP) only for UI labels.
- Do not mix legacy keys after registration.
Consistency Matrix:
- npm package name → earch-mcp
- Binary name → earch-mcp
- MCP server name (SDK metadata) → earch-mcp
- Env default MCP_NAME → earch-mcp
- Client registry key → earch-mcp
- UI label → Earch MCP
Conflict Cleanup:
- Remove any stale keys (e.g., old display names) and re-add with earch-mcp only.
- Cursor: configure in the UI; this project intentionally omits .cursor/mcp.json.
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.