Official Perplexity AI search MCP server
## Perplexity MCP Server: AI-Powered Search Engine The **Perplexity MCP Server** integrates the conversational AI search engine directly into Google Antigravity, enabling AI assistants to perform research queries with synthesized, sourced answers. This integration brings intelligent information retrieval to your development workflow. ### Why Perplexity MCP? - **Synthesized Answers**: Get comprehensive answers synthesized from multiple sources - **Source Citations**: All responses include cited sources for verification - **Real-Time Information**: Access current information with fresh web searches - **Follow-Up Queries**: Maintain context for iterative research conversations - **Pro Search**: Deep research mode for comprehensive topic exploration ### Key Features #### 1. Research Queries ```python from anthropic import Anthropic client = Anthropic() response = client.messages.create( model="claude-sonnet-4-20250514", max_tokens=1024, messages=[{ "role": "user", "content": "Research the current state of WebAssembly adoption in production environments" }], tools=[{ "name": "perplexity_search", "description": "AI-powered search" }] ) ``` #### 2. Deep Research ```python # Comprehensive research response = client.messages.create( model="claude-sonnet-4-20250514", max_tokens=1024, messages=[{ "role": "user", "content": "Conduct deep research on GraphQL vs REST API design patterns with current best practices" }], tools=[{"name": "perplexity_pro", "description": "Deep research mode"}] ) ``` #### 3. Follow-Up Questions ```python # Continue research thread response = client.messages.create( model="claude-sonnet-4-20250514", max_tokens=1024, messages=[{ "role": "user", "content": "Follow up on that research: what are the performance implications?" }], tools=[{"name": "perplexity_followup", "description": "Follow-up queries"}] ) ``` #### 4. Focused Search ```python # Domain-specific search response = client.messages.create( model="claude-sonnet-4-20250514", max_tokens=1024, messages=[{ "role": "user", "content": "Search academic papers about transformer architecture optimizations" }], tools=[{"name": "perplexity_academic", "description": "Academic search"}] ) ``` ### Configuration ```json { "mcpServers": { "perplexity": { "command": "npx", "args": ["-y", "@anthropic/mcp-perplexity"], "env": { "PERPLEXITY_API_KEY": "your-perplexity-api-key" } } } } ``` ### Use Cases **Technical Research**: Get synthesized answers to complex technical questions with sources. **Competitive Analysis**: Research competitors, technologies, and market trends comprehensively. **Documentation**: Gather information for technical writing with verified sources. **Learning**: Explore new topics with comprehensive, well-sourced explanations. The Perplexity MCP Server brings AI-powered research capabilities directly into your development workflow, enabling intelligent information synthesis.
{
"mcpServers": {
"perplexity": {
"mcpServers": {
"perplexity": {
"env": {
"PERPLEXITY_API_KEY": "YOUR_PERPLEXITY_API_KEY"
},
"args": [
"-y",
"perplexity-mcp"
],
"command": "npx"
}
}
}
}
}