Frequently Asked Questions about the Perplexity MCP Server

Q: What is the Perplexity MCP Server?

A: The Perplexity MCP Server is a Python-based interface that allows you to easily integrate the Perplexity AI API with the UBOS AI Agent Development Platform. It simplifies the process of querying Perplexity for responses, managing chat histories, and configuring interactions.

Q: What is MCP?

A: MCP stands for Model Context Protocol. It’s an open protocol that standardizes how applications provide context to Language Learning Models (LLMs), allowing different AI tools and services to communicate effectively.

Q: What are the key features of the Perplexity MCP Server?

A: Key features include streamlined integration with Perplexity AI, persistent chat history, flexible model configuration via environment variables, a comprehensive toolset (ask_perplexity, chat_perplexity, list_chats_perplexity, read_chat_perplexity), a web UI for easy management, and open-source customizability.

Q: What are some use cases for the Perplexity MCP Server?

A: Use cases include enhanced customer support, automated content creation, improved research and analysis, personalized learning experiences, code generation and debugging, and AI-powered knowledge bases.

Q: How do I configure the Perplexity MCP Server?

A: You configure the server by setting environment variables such as PERPLEXITY_API_KEY, PERPLEXITY_MODEL, PERPLEXITY_MODEL_ASK, PERPLEXITY_MODEL_CHAT, and DB_PATH.

Q: What is the ask_perplexity tool used for?

A: The ask_perplexity tool is used for asking specific questions and receiving expert programming assistance from Perplexity AI. It focuses on coding solutions, error debugging, and technical explanations.

Q: What is the chat_perplexity tool used for?

A: The chat_perplexity tool is used for engaging in ongoing conversations with Perplexity AI. It maintains chat history and allows you to create new chats or continue existing ones.

Q: What is the list_chats_perplexity tool used for?

A: The list_chats_perplexity tool lists all available chat conversations with Perplexity AI. It returns a paginated list of chats with chat IDs, titles, and creation times.

Q: What is the read_chat_perplexity tool used for?

A: The read_chat_perplexity tool retrieves the complete conversation history for a specific chat, reading directly from the local database without making any API calls to Perplexity.

Q: How do I access the Web UI?

A: When WEB_UI_ENABLED is set to true, the web UI is available at http://WEB_UI_HOST:WEB_UI_PORT (by default, http://127.0.0.1:8050).

Q: What are the prerequisites for using the Perplexity MCP Server?

A: The prerequisites are Python 3.10 or higher and the uvx package manager installed.

Q: Is the Perplexity MCP Server open source?

A: Yes, the Perplexity MCP Server is an open-source project licensed under the MIT License.

Q: How does the Perplexity MCP Server integrate with UBOS?

A: The Perplexity MCP Server acts as a bridge, allowing AI models within the UBOS platform to access and interact with Perplexity AI. This simplifies the process of incorporating Perplexity’s powerful language models into your AI Agent workflows.

Q: Can I use different Perplexity models for the ask_perplexity and chat_perplexity tools?

A: Yes, you can specify different models for these tools by setting the PERPLEXITY_MODEL_ASK and PERPLEXITY_MODEL_CHAT environment variables. If these are not set, the PERPLEXITY_MODEL variable will be used as the default.