time-mcp

What is time-mcp?

time-mcp is a minimal agentic AI system that answers time-related and general questions using a tool-augmented LLM pipeline. It integrates a Flask API for the current timestamp, an MCP agent server for reasoning and tool calling, and a Streamlit chat interface.

How to use time-mcp?

Set the OPENROUTER_API_KEY environment variable, then run three components in separate terminals: python flask_api.py for the time API, python mcp_server.py for the MCP agent server, and streamlit run streamlit_ui.py for the UI. Open the Streamlit UI in your browser (default http://localhost:8501) and ask questions.

Key features of time-mcp

• Flask API provides the current timestamp.
• MCP agent server detects user intent and calls tools.
• Integrates with LLMs via OpenRouter (OpenAI-compatible API).
• Streamlit UI offers a simple chat interface.
• Easily extensible with additional tools via the MCPAgent class.

Use cases of time-mcp

• Ask for the current time and receive a natural language response.
• Ask general knowledge questions that are answered by the LLM.
• Demonstrate a tool-augmented agent pipeline for learning or prototyping.

FAQ from time-mcp

What dependencies are required to run time-mcp?

Python 3.7+ and the packages listed in requirements.txt. Also need an OpenRouter API key set as the OPENROUTER_API_KEY environment variable.

How does authentication work?

The system uses an OpenRouter API key for LLM access. The Flask time API does not require authentication.

Can I add more tools to the agent?

Yes. Implement new methods in the MCPAgent class and update self.tools. You can also improve intent detection by extending detect_intent().

What LLM model does time-mcp use?

The model is defined in the call_llm() method and can be changed by updating the model field. It uses OpenRouter, which provides access to various models.

Is there a web interface?

Yes, a Streamlit UI is provided. Run streamlit run streamlit_ui.py to open the chat interface in your browser.