Trawlia MCP server
{
"mcpServers": {
"trawlia": {
"command": "npx",
"args": [
"-y",
"trawlia-mcp@latest"
],
"env": {
"TRAWLIA_API_KEY": "<trawlia_api_key>",
"TRAWLIA_BASE_URL": "https://..."
}
}
}
}{
"mcpServers": {
"trawlia": {
"command": "npx",
"args": [
"-y",
"trawlia-mcp@latest"
],
"env": {
"TRAWLIA_API_KEY": "<trawlia_api_key>",
"TRAWLIA_BASE_URL": "https://..."
}
}
}
}{
"mcpServers": {
"trawlia": {
"command": "npx",
"args": [
"-y",
"trawlia-mcp@latest"
],
"env": {
"TRAWLIA_API_KEY": "<trawlia_api_key>",
"TRAWLIA_BASE_URL": "https://..."
}
}
}
}{
"servers": {
"trawlia": {
"command": "npx",
"args": [
"-y",
"trawlia-mcp@latest"
],
"env": {
"TRAWLIA_API_KEY": "${input:trawlia_api_key}",
"TRAWLIA_BASE_URL": "${input:trawlia_base_url}"
}
}
},
"inputs": [
{
"id": "trawlia_api_key",
"type": "promptString",
"description": "Trawlia api key",
"password": true
},
{
"id": "trawlia_base_url",
"type": "promptString",
"description": "Trawlia base url",
"password": true
}
]
}{
"mcpServers": {
"trawlia": {
"command": "npx",
"args": [
"-y",
"trawlia-mcp@latest"
],
"env": {
"TRAWLIA_API_KEY": "<trawlia_api_key>",
"TRAWLIA_BASE_URL": "https://..."
}
}
}
}file .mcp.json (in your project root)file claude_desktop_config.jsonfile ~/.cursor/mcp.jsonfile .vscode/mcp.jsonfile ~/.codeium/windsurf/mcp_config.jsoninstall npx -y trawlia-mcpversion 0.1.0
What it does
Trawlia is an MCP server that provides web search and page extraction tools, returning results as readable text rather than raw JSON. It is for agents and AI applications that need to search the live web or extract content from specific URLs.
- Search the live web with natural language queries and get ranked, deduplicated results with snippets or full page text.
- Extract readable content from a list of URLs with ads, banners, and modals removed.
- Perform news-focused searches to find recent articles with publication dates.
Tools this server exposes
| Tool | Does |
|---|---|
| trawlia_search | Search the live web with configurable depth (basic or advanced), topic (general or news), result count, domain filters, and time range. |
| trawlia_extract | Fetch and extract readable content from multiple URLs at once, with formatting elements removed. |
Before you enable it
An MCP server runs with whatever access you give it. Start with a development account rather than production, and check which tools the server exposes before you let an agent call them unattended.