What is the MCP Server?

The MCP Server is an advanced sequential thinking process that utilizes a Multi-Agent System (MAS) built with the Agno framework and served via MCP (Model Context Protocol).

How does the MCP Server work?

It uses a team of specialized AI agents (Planner, Researcher, Analyzer, Critic, Synthesizer) coordinated by a central ‘Team’ agent to analyze and solve complex problems sequentially.

What are the key benefits of using the MCP Server?

Key benefits include advanced reasoning capabilities, comprehensive problem-solving, dynamic information gathering, and seamless integration with the UBOS platform.

What is the Agno framework?

Agno is a framework used to build the Multi-Agent System (MAS) that powers the MCP Server.

What is MCP (Model Context Protocol)?

MCP is an open protocol that standardizes how applications provide context to LLMs, allowing AI models to access and interact with external data sources and tools.

What kind of problems is the MCP Server suitable for?

It is suitable for complex problem-solving scenarios that require in-depth analysis, critical evaluation, and creative synthesis, such as strategic planning, risk assessment, scientific research, and creative writing.

How does the MCP Server integrate with UBOS?

The MCP Server seamlessly integrates with the UBOS platform, providing users with a comprehensive AI development environment, including AI Agent orchestration, enterprise data connectivity, and custom AI Agent development.

What is the token consumption like for the MCP Server?

Due to its Multi-Agent System architecture, the MCP Server consumes significantly more tokens than single-agent alternatives. Plan your budget accordingly.

What do I need to use the MCP Server?

You need access to a compatible LLM API (configured for Agno), such as Groq, DeepSeek, or OpenRouter, and an Exa API key if using the Researcher agent’s capabilities.

How do I install and configure the MCP Server?

You can install it via Smithery or manually by cloning the repository, setting environment variables, and installing dependencies. Refer to the installation instructions for detailed steps.

What is sequential thinking?

Sequential thinking is an advanced AI process where problems are solved through a sequence of coordinated thoughts and actions, much like human reasoning.

How does the MCP Server handle revisions and branches in the thinking process?

The MCP Server supports revisions of previous steps and branching to explore alternative paths, allowing for a more flexible and nuanced problem-solving approach.

Can I use custom models with the MCP Server?

Yes, you can specify different models for the Team Coordinator and Specialist Agents by setting the TEAM_MODEL_ID and AGENT_MODEL_ID environment variables.

How do I troubleshoot issues with the MCP Server?

Refer to the detailed logging, which tracks every step of the process, including agent interactions. This helps in identifying and debugging any issues.

Where are the logs stored?

Logs are written to ~/.sequential_thinking/logs/sequential_thinking.log by default.