Perplexity AI MCP Server for AutoGen 14 tools — connect in under 2 minutes
Microsoft AutoGen enables multi-agent conversations where agents negotiate, delegate, and execute tasks collaboratively. Add Perplexity AI as an MCP tool provider through the Vinkius and every agent in the group can access live data and take action.
ASK AI ABOUT THIS MCP SERVER
Vinkius supports streamable HTTP and SSE.
import asyncio
from autogen_agentchat.agents import AssistantAgent
from autogen_ext.tools.mcp import McpWorkbench
async def main():
# Your Vinkius token — get it at cloud.vinkius.com
async with McpWorkbench(
server_params={"url": "https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"},
transport="streamable_http",
) as workbench:
tools = await workbench.list_tools()
agent = AssistantAgent(
name="perplexity_ai_agent",
tools=tools,
system_message=(
"You help users with Perplexity AI. "
"14 tools available."
),
)
print(f"Agent ready with {len(tools)} tools")
asyncio.run(main())
* Every MCP server runs on Vinkius-managed infrastructure inside AWS - a purpose-built runtime with per-request V8 isolates, Ed25519 signed audit chains, and sub-40ms cold starts optimized for native MCP execution. See our infrastructure
About Perplexity AI MCP Server
Connect your Perplexity AI API key to any AI agent and harness the power of real-time web search with AI-generated answers, citations, and related questions through natural conversation.
AutoGen enables multi-agent conversations where agents negotiate, delegate, and collaboratively use Perplexity AI tools. Connect 14 tools through the Vinkius and assign role-based access — a data analyst queries while a reviewer validates, with optional human-in-the-loop approval for sensitive operations.
What you can do
- Answer Questions — Ask any question and get grounded answers with real-time web search and source citations
- Deep Research — Perform exhaustive research on complex topics with comprehensive reports and thorough citations
- Logical Reasoning — Solve complex problems requiring step-by-step analysis and chain-of-thought reasoning
- Domain-Filtered Search — Restrict search results to specific domains for academic, technical, or trusted-source queries
- Recency Filtering — Get answers based on recent information only (hour, day, week, month, or year)
- Multi-Turn Conversations — Maintain context across multiple questions for iterative research sessions
- Structured Output — Get responses in JSON format following a defined schema for programmatic integration
- Visual Results — Include relevant images and related questions in search results
The Perplexity AI MCP Server exposes 14 tools through the Vinkius. Connect it to AutoGen in under two minutes — no API keys to rotate, no infrastructure to provision, no vendor lock-in. Your configuration, your data, your control.
How to Connect Perplexity AI to AutoGen via MCP
Follow these steps to integrate the Perplexity AI MCP Server with AutoGen.
Install AutoGen
Run pip install "autogen-ext[mcp]"
Replace the token
Replace [YOUR_TOKEN_HERE] with your Vinkius token
Integrate into workflow
Use the agent in your AutoGen multi-agent orchestration
Explore tools
The workbench discovers 14 tools from Perplexity AI automatically
Why Use AutoGen with the Perplexity AI MCP Server
AutoGen provides unique advantages when paired with Perplexity AI through the Model Context Protocol.
Multi-agent conversations: multiple AutoGen agents discuss, delegate, and collaboratively use Perplexity AI tools to solve complex tasks
Role-based architecture lets you assign Perplexity AI tool access to specific agents — a data analyst queries while a reviewer validates
Human-in-the-loop support: agents can pause for human approval before executing sensitive Perplexity AI tool calls
Code execution sandbox: AutoGen agents can write and run code that processes Perplexity AI tool responses in an isolated environment
Perplexity AI + AutoGen Use Cases
Practical scenarios where AutoGen combined with the Perplexity AI MCP Server delivers measurable value.
Collaborative analysis: one agent queries Perplexity AI while another validates results and a third generates the final report
Automated review pipelines: a researcher agent fetches data from Perplexity AI, a critic agent evaluates quality, and a writer produces the output
Interactive planning: agents negotiate task allocation using Perplexity AI data to make informed decisions about resource distribution
Code generation with live data: an AutoGen coder agent writes scripts that process Perplexity AI responses in a sandboxed execution environment
Perplexity AI MCP Tools for AutoGen (14)
These 14 tools become available when you connect Perplexity AI to AutoGen via MCP:
chat_completion
The Sonar model searches the web, synthesizes information, and provides a concise answer. This is the basic query tool for factual questions, summaries, and general knowledge. Use this for quick lookups where you need accurate, up-to-date information. Ask Perplexity AI a question and get a grounded, cited answer
chat_with_citations
Each claim or fact in the response is linked to its original source. This is essential for research, fact-checking, and academic work where sources matter. The response includes a citations array with URLs of all referenced sources. Ask Perplexity AI and get answers with source citations
chat_with_domain_filter
Provide domains as a comma-separated list (e.g., "arxiv.org,nih.gov,github.com"). Only sources from the specified domains will be used in generating the answer. Use this for domain-specific research, academic papers, or trusted sources only. Citations are automatically included to verify sources. Ask Perplexity AI restricting search to specific domains
chat_with_history
Provide messages as a JSON array of {role: "user"|"assistant"|"system", content: "text"} objects. This enables follow-up questions where the model understands previous context. Use this for complex queries that build on previous answers or require contextual understanding. Example: [{ "role": "user", "content": "What is quantum computing?" }, { "role": "assistant", "content": "Quantum computing uses quantum bits..." }, { "role": "user", "content": "How does it differ from classical computing?" }] Ask Perplexity AI with multi-turn conversation history
chat_with_images
The response includes an images array with URLs to related images found during the search. Use this for visual topics, product searches, or when you need images to accompany the answer. Ask Perplexity AI and get relevant images with the answer
chat_with_recency_filter
Available recency filters: "hour", "day", "week", "month", "year". This ensures the answer is based on recent information only. Use this for news, recent events, or time-sensitive queries where outdated info is not useful. Ask Perplexity AI with results filtered by time recency
chat_with_related_questions
The response includes a related_questions array with suggested questions for further exploration. Use this for research, learning, and discovering related topics you might want to explore. Ask Perplexity AI and get related follow-up questions
deep_research
This model performs extensive web searches and generates detailed reports with thorough citations. It takes longer than regular queries but provides much more depth and breadth. Use this for complex topics, literature reviews, competitive analysis, or thorough investigations. Maximum tokens default to 4096 for comprehensive responses. Perform deep research with exhaustive web search and comprehensive report
follow_up
Provide the conversation history as a JSON array of messages and the follow-up question. This maintains context from previous turns in the conversation. Use this for multi-turn research sessions where each question builds on previous answers. Ask a follow-up question in an ongoing conversation with Perplexity AI
list_models
Use this to discover what models are available before choosing which one to use for your queries. List all available Perplexity AI models
reasoning
This model excels at multi-step reasoning, mathematical problems, code analysis, and chain-of-thought tasks. Use this for problems requiring step-by-step analysis, mathematical proofs, code reviews, or logical deductions. Citations are included where external information is referenced. Ask Perplexity AI for complex logical reasoning and step-by-step analysis
search_query
This combines all search features: cited sources, relevant images, and follow-up questions. Use this when you want the fullest possible search result with all supplementary information. The response includes content, citations array, images array, and related_questions array. Perform a comprehensive web search with citations, images, and related questions
structured_query
The model will return the answer as JSON matching your schema definition. Provide the JSON schema as a string. This is useful for programmatic data extraction, API integrations, and when you need consistent, parseable responses. Example schema: { "type": "object", "properties": { "name": { "type": "string" }, "age": { "type": "number" } } } Ask Perplexity AI and get a structured JSON response following a schema
system_prompt_query
The system prompt defines how the model should respond (e.g., "You are a medical expert...", "Answer in bullet points..."). Use this for specialized queries, role-playing, formatting requirements, or domain-specific expertise. Example system prompt: "You are a senior software architect. Explain concepts with code examples." Ask Perplexity AI with a custom system prompt to set behavior and context
Example Prompts for Perplexity AI in AutoGen
Ready-to-use prompts you can give your AutoGen agent to start working with Perplexity AI immediately.
"What are the latest developments in quantum computing as of this week?"
"Do deep research on the competitive landscape of electric vehicle manufacturers in Southeast Asia, including market share, pricing strategies, and government incentives."
"Search for news about AI regulation in the European Union from the last month, restricted to europa.eu and reuters.com domains."
Troubleshooting Perplexity AI MCP Server with AutoGen
Common issues when connecting Perplexity AI to AutoGen through the Vinkius, and how to resolve them.
McpWorkbench not found
pip install "autogen-ext[mcp]"Perplexity AI + AutoGen FAQ
Common questions about integrating Perplexity AI MCP Server with AutoGen.
How does AutoGen connect to MCP servers?
Can different agents have different MCP tool access?
Does AutoGen support human approval for tool calls?
Connect Perplexity AI with your favorite client
Step-by-step setup guides for every MCP-compatible client and framework:
Anthropic's native desktop app for Claude with built-in MCP support.
AI-first code editor with integrated LLM-powered coding assistance.
GitHub Copilot in VS Code with Agent mode and MCP support.
Purpose-built IDE for agentic AI coding workflows.
Autonomous AI coding agent that runs inside VS Code.
Anthropic's agentic CLI for terminal-first development.
Python SDK for building production-grade OpenAI agent workflows.
Google's framework for building production AI agents.
Type-safe agent development for Python with first-class MCP support.
TypeScript toolkit for building AI-powered web applications.
TypeScript-native agent framework for modern web stacks.
Python framework for orchestrating collaborative AI agent crews.
Leading Python framework for composable LLM applications.
Data-aware AI agent framework for structured and unstructured sources.
Microsoft's framework for multi-agent collaborative conversations.
Connect Perplexity AI to AutoGen
Get your token, paste the configuration, and start using 14 tools in under 2 minutes. No API key management needed.