### Configure Perplexity MCP Server

Source: https://context7.com/dainfernalcoder/perplexity-mcp/llms.txt

Configure the Perplexity MCP server in your MCP client settings. This example shows basic setup with an environment variable for the API key.

```json
{
  "mcpServers": {
    "perplexity": {
      "command": "npx",
      "args": ["-y", "perplexity-mcp"],
      "env": {
        "PERPLEXITY_API_KEY": "your_api_key"
      }
    }
  }
}
```

--------------------------------

### Use Perplexity Search Tool

Source: https://context7.com/dainfernalcoder/perplexity-mcp/llms.txt

Examples of using the 'search' tool with the Perplexity MCP server. Includes basic queries, recommended technical details, and forcing the search model.

```javascript
// Basic search query
const result = await use_mcp_tool({
  server_name: "perplexity",
  tool_name: "search",
  arguments: {
    query: "What is the capital of France?"
  }
});
```

```javascript
// Search with specific technical details (recommended)
const technicalResult = await use_mcp_tool({
  server_name: "perplexity",
  tool_name: "search",
  arguments: {
    query: "What is the default port for PostgreSQL 15.4 on Ubuntu 22.04?"
  }
});
```

```javascript
// Force using search model even if query seems complex
const forcedResult = await use_mcp_tool({
  server_name: "perplexity",
  tool_name: "search",
  arguments: {
    query: "How does React's virtual DOM work?",
    force_model: true  // Prevents automatic routing to reason tool
  }
});
```

```javascript
// Response includes content and sources
// {
//   content: [{
//     type: "text",
//     text: "The capital of France is Paris...\n\n## Sources\n1: https://..."
//   }]
// }
```

--------------------------------

### Configure Perplexity MCP Server (Node.js)

Source: https://github.com/dainfernalcoder/perplexity-mcp/blob/main/README.md

Configure the MCP server to use the Perplexity server by specifying the command, arguments, and environment variables, including the Perplexity API key. This example uses Node.js.

```json
{
  "mcpServers": {
    "perplexity": {
      "command": "node",
      "args": ["/path/to/perplexity-server/build/index.js"],
      "env": {
        "PERPLEXITY_API_KEY": "YOUR_API_KEY_HERE"
      },
      "disabled": false,
      "autoApprove": []
    }
  }
}
```

--------------------------------

### Configure Perplexity MCP Server (NPX)

Source: https://github.com/dainfernalcoder/perplexity-mcp/blob/main/README.md

Configure the MCP server to use the Perplexity server via NPX, which avoids local installation. Specify the command, arguments, and environment variables, including the Perplexity API key.

```json
{
  "mcpServers": {
    "perplexity": {
      "command": "npx",
      "args": [
        "-y",
        "perplexity-mcp"
      ],
      "env": {
        "PERPLEXITY_API_KEY": "your_api_key"
      }
    }
  }
}
```

--------------------------------

### Conduct Deep Research with Focus Areas

Source: https://context7.com/dainfernalcoder/perplexity-mcp/llms.txt

Use the 'deep_research' tool for comprehensive research. Specify the main query and optional `focus_areas` to guide the research towards specific subtopics.

```javascript
// Comprehensive research with focus areas
const researchResult = await use_mcp_tool({
  server_name: "perplexity",
  tool_name: "deep_research",
  arguments: {
    query: "The impact of quantum computing on cryptography",
    focus_areas: [
      "Post-quantum cryptographic algorithms",
      "Timeline for quantum threats to current encryption",
      "Practical mitigation strategies for organizations"
    ]
  }
});
```

--------------------------------

### Compare Technologies with Reason Tool

Source: https://context7.com/dainfernalcoder/perplexity-mcp/llms.txt

Use the 'reason' tool to compare and contrast technologies. Provide a detailed query outlining the comparison points and desired output format.

```javascript
// Compare technologies with reasoning
const comparisonResult = await use_mcp_tool({
  server_name: "perplexity",
  tool_name: "reason",
  arguments: {
    query: "Compare and contrast REST and GraphQL APIs, explaining their pros and cons for a mobile application with limited bandwidth"
  }
});
```

--------------------------------

### Configure Perplexity MCP Server (Command-Line API Key)

Source: https://github.com/dainfernalcoder/perplexity-mcp/blob/main/README.md

Configure the MCP server to use the Perplexity server, passing the API key directly as a command-line argument. This is a fallback if the API key cannot be read from the environment.

```json
{
  "mcpServers": {
    "perplexity": {
      "command": "node",
      "args": [
        "/path/to/perplexity-server/build/index.js",
        "--api-key",
        "your_api_key_here"
      ],
      "disabled": false,
      "autoApprove": []
    }
  }
}
```

--------------------------------

### Use Perplexity Search Tool

Source: https://github.com/dainfernalcoder/perplexity-mcp/blob/main/README.md

Use this tool for quick searches and basic information lookup. It's best for straightforward questions requiring direct answers. The `force_model` option is optional.

```javascript
const result = await use_mcp_tool({
  server_name: "perplexity",
  tool_name: "search",
  arguments: {
    query: "What is the capital of France?",
    force_model: false // Optional: force using this model even if query seems complex
  }
});
```

--------------------------------

### Perplexity API Key Configuration Options

Source: https://context7.com/dainfernalcoder/perplexity-mcp/llms.txt

Provides three methods for configuring the Perplexity API key: environment variable (preferred), command-line argument, or .env file with --cwd flag. Ensure the API key format is correct.

```json
// Option 1: Environment variable (preferred)
{
  "mcpServers": {
    "perplexity": {
      "command": "npx",
      "args": ["-y", "perplexity-mcp"],
      "env": {
        "PERPLEXITY_API_KEY": "pplx-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
      }
    }
  }
}
```

```json
// Option 2: Command-line argument
{
  "mcpServers": {
    "perplexity": {
      "command": "node",
      "args": [
        "/path/to/perplexity-server/build/index.js",
        "--api-key",
        "pplx-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
      ]
    }
  }
}
```

```json
// Option 3: .env file with --cwd (reads PERPLEXITY_API_KEY from .env)
{
  "mcpServers": {
    "perplexity": {
      "command": "node",
      "args": [
        "/path/to/perplexity-server/build/index.js",
        "--cwd",
        "/path/to/your/project"
      ]
    }
  }
}
```

--------------------------------

### Configure Perplexity MCP Server (.env File)

Source: https://github.com/dainfernalcoder/perplexity-mcp/blob/main/README.md

Configure the MCP server to read API keys from an explicit .env file using the `--cwd` argument. This is a fallback method for providing environment variables.

```json
{
  "mcpServers": {
    "perplexity": {
      "command": "node",
      "args": [
        "/path/to/perplexity-server/build/index.js",
        "--cwd",
        "/path/to/your/project"
      ],
      "disabled": false,
      "autoApprove": []
    }
  }
}
```

--------------------------------

### Use Perplexity Reason Tool

Source: https://github.com/dainfernalcoder/perplexity-mcp/blob/main/README.md

Handles complex, multi-step tasks requiring detailed analysis, such as explanations and comparisons. The `force_model` option is optional.

```javascript
const result = await use_mcp_tool({
  server_name: "perplexity",
  tool_name: "reason",
  arguments: {
    query: "Compare and contrast REST and GraphQL APIs, explaining their pros and cons",
    force_model: false // Optional: force using this model even if query seems simple
  }
});
```

--------------------------------

### Analyze Architecture with Reason Tool

Source: https://context7.com/dainfernalcoder/perplexity-mcp/llms.txt

Employ the 'reason' tool for architectural decision analysis. Specify the architectures to compare and the context, such as team size and project type. The `force_model` argument can be set to `false` to allow the model to choose the best approach.

```javascript
// Architectural decision analysis
const architectureResult = await use_mcp_tool({
  server_name: "perplexity",
  tool_name: "reason",
  arguments: {
    query: "What are the advantages and disadvantages of using microservices vs monolithic architecture for a startup with a team of 5 developers building an e-commerce platform?",
    force_model: false
  }
});
```

--------------------------------

### Automatic Query Routing to Reason Model

Source: https://context7.com/dainfernalcoder/perplexity-mcp/llms.txt

Queries containing complexity indicators like 'how', 'explain', or 'difference' are automatically upgraded from 'search' to the 'reason' tool (Sonar Reasoning Pro).

```javascript
// This query will be auto-routed to "reason" (Sonar Reasoning Pro)
// because it contains complexity indicators: "how", "explain", "difference"
await use_mcp_tool({
  server_name: "perplexity",
  tool_name: "search",  // Will be upgraded to "reason" automatically
  arguments: {
    query: "How does React's reconciliation algorithm work and what's the difference between React 17 and React 18?"
  }
});
```

--------------------------------

### Perform Deep Research with deep_research Tool

Source: https://github.com/dainfernalcoder/perplexity-mcp/blob/main/examples/deep-research.md

Use this snippet to initiate a deep research query on a specific topic. Provide a detailed query and optional focus areas for a comprehensive analysis.

```javascript
const result = await use_mcp_tool({
  server_name: "perplexity-ask",
  tool_name: "deep_research",
  arguments: {
    query: "The impact of quantum computing on cryptography",
    focus_areas: [
      "Post-quantum cryptographic algorithms",
      "Timeline for quantum threats to current encryption",
      "Practical mitigation strategies for organizations"
    ]
  }
});

console.log(result);
```

--------------------------------

### Use Perplexity Deep Research Tool

Source: https://github.com/dainfernalcoder/perplexity-mcp/blob/main/README.md

Conducts comprehensive research and generates detailed reports for in-depth analysis. The `force_model` option is optional. You can specify `focus_areas` for more targeted research.

```javascript
const result = await use_mcp_tool({
  server_name: "perplexity",
  tool_name: "deep_research",
  arguments: {
    query: "The impact of quantum computing on cryptography",
    focus_areas: [
      "Post-quantum cryptographic algorithms",
      "Timeline for quantum threats",
      "Practical mitigation strategies"
    ],
    force_model: false // Optional: force using this model even if query seems simple
  }
});
```

--------------------------------

### Query with Specific Details

Source: https://context7.com/dainfernalcoder/perplexity-mcp/llms.txt

Providing specific details such as error messages, code snippets, versions, and environment information in the query leads to more precise and actionable responses, avoiding the 'missing details' note.

```javascript
// Better query with specific details - no missing details note
const specificResult = await use_mcp_tool({
  server_name: "perplexity",
  tool_name: "reason",
  arguments: {
    query: `TypeError: Cannot read property 'map' of undefined
    at UserList (UserList.tsx:15:23)

Code:
\`\`\`typescript
// UserList.tsx
const UserList = ({ users }) => {
  return users.map(user => <div key={user.id}>{user.name}</div>);
};
\`\`\`

React 18.2.0, TypeScript 5.3.3, Next.js 14.1.0 on macOS`
  }
});
```

--------------------------------

### Debug Errors with Reason Tool

Source: https://context7.com/dainfernalcoder/perplexity-mcp/llms.txt

Utilize the 'reason' tool to debug specific errors by providing the error message, relevant code context, library versions, and operating system details.

```javascript
// Debug a specific error with context
const debugResult = await use_mcp_tool({
  server_name: "perplexity",
  tool_name: "reason",
  arguments: {
    query: "I'm getting this error in my Node.js 20.11 application:\n\nError: ECONNREFUSED 127.0.0.1:5432\n    at TCPConnectWrap.afterConnect [as oncomplete] \n\nI'm using pg library version 8.11.3 on macOS Sonoma.\nThe PostgreSQL server is running (checked with 'brew services list').\nMy connection config:\n{\n  host: 'localhost',\n  port: 5432,\n  database: 'myapp_dev'\n}\n\nWhat could be causing this connection failure?"
  }
});
```

--------------------------------

### Contextual Research with Focus Areas and Model Control

Source: https://context7.com/dainfernalcoder/perplexity-mcp/llms.txt

Perform research with specific technical context using the 'deep_research' tool. You can define `focus_areas` and control model behavior with `force_model` set to `false`.

```javascript
// Research with specific technical context
const contextualResearch = await use_mcp_tool({
  server_name: "perplexity",
  tool_name: "deep_research",
  arguments: {
    query: "Best practices for implementing authentication in Next.js 14 App Router applications",
    focus_areas: [
      "NextAuth.js v5 vs Clerk vs Auth0 comparison",
      "Server Components authentication patterns",
      "Middleware-based route protection",
      "Session management strategies"
    ],
    force_model: false
  }
});
```

--------------------------------

### Query Lacking Specific Details

Source: https://context7.com/dainfernalcoder/perplexity-mcp/llms.txt

When a query lacks specific details, the server will note this in the response, prompting the user to provide more context for a better answer. This is useful for debugging vague issues.

```javascript
// Query lacking specifics - server will note missing details in response
const vagueResult = await use_mcp_tool({
  server_name: "perplexity",
  tool_name: "reason",
  arguments: {
    query: "Why is my code not working?"
  }
});
```

--------------------------------

### Automatic Query Routing to Search Model

Source: https://context7.com/dainfernalcoder/perplexity-mcp/llms.txt

This query is automatically routed to the 'search' tool (Sonar Pro) due to its simple factual nature. Ensure the 'server_name' and 'tool_name' are correctly configured.

```javascript
await use_mcp_tool({
  server_name: "perplexity",
  tool_name: "search",
  arguments: {
    query: "What is TypeScript?"
  }
});
```

--------------------------------

### Automatic Query Routing to Deep Research Model

Source: https://context7.com/dainfernalcoder/perplexity-mcp/llms.txt

Queries with research indicators such as 'analyze', 'comprehensive', or 'explore' are automatically routed to the 'deep_research' tool (Sonar Deep Research).

```javascript
// This query will be auto-routed to "deep_research" (Sonar Deep Research)
// because it contains research indicators: "analyze", "comprehensive", "explore"
await use_mcp_tool({
  server_name: "perplexity",
  tool_name: "search",  // Will be upgraded to "deep_research" automatically
  arguments: {
    query: "Analyze the comprehensive impact of AI on software development and explore different perspectives"
  }
});
```

--------------------------------

### Perform Technical Deep Dive Research

Source: https://context7.com/dainfernalcoder/perplexity-mcp/llms.txt

Execute a technical deep dive using the 'deep_research' tool without specifying `focus_areas`. This is suitable for broad queries requiring extensive technical information.

```javascript
// Technical deep dive without focus areas
const techResearch = await use_mcp_tool({
  server_name: "perplexity",
  tool_name: "deep_research",
  arguments: {
    query: "Current state of WebAssembly adoption in production environments, performance benchmarks compared to JavaScript, and real-world use cases in 2024"
  }
});
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.