Provides a time API endpoint that returns the current timestamp, which the MCP agent can query when handling time-related questions
Connects to OpenRouter (an OpenAI-compatible API) to access language models for generating responses to user queries
Offers a chat interface for users to interact with the MCP agent, allowing them to ask time-related and general questions
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@Time-MCP what time is it in Tokyo right now?"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
A minimal agentic AI system that answers time-related and general questions using a tool-augmented LLM pipeline.
Flask API: Provides the current timestamp.
MCP Agent Server: Reasoning agent that detects user intent, calls tools (like the time API), engineers prompts, and interacts with an LLM via OpenRouter (OpenAI-compatible API).
Streamlit UI: Simple chat interface to talk to the AI agent.
Related MCP server: MCP: Multi-Agent Control Point
pip install -r requirements.txtSet your OpenRouter API key (get one from https://openrouter.ai):
export OPENROUTER_API_KEY=sk-...your-key...Open three terminals (or use background processes):
python flask_api.pypython mcp_server.pystreamlit run streamlit_ui.pyThe Streamlit UI will open in your browser (default: http://localhost:8501)
Ask the agent any question in the Streamlit UI.
If you ask about the time (e.g., "What is the time?"), the agent will call the Flask API, fetch the current time, and craft a beautiful, natural response using the LLM.
For other questions, the agent will answer using the LLM only.
[Streamlit UI] → [MCP Agent Server] → [Tools (e.g., Time API)]
↓
[LLM via OpenRouter]The MCP agent detects intent, calls tools as needed, engineers prompts, and sends them to the LLM.
Easily extensible to add more tools (just add to the MCPAgent class).
Add more tools: Implement new methods in MCPAgent and update self.tools.
Improve intent detection: Extend detect_intent() in MCPAgent.
Change LLM model: Update the model field in call_llm().
Python 3.7+
See requirements.txt for dependencies.
Built using Flask, Streamlit, OpenRouter, and Python.
Inspired by agentic LLM design patterns.
This server cannot be deployed
Wall-clock awareness for LLM agents. Two tools: elapsed-time-between-turns + day rollover detection.
Current time, timezone conversion & date math for AI agents. On Cloudflare Workers.
A real clock for AI agents: current time, timezone conversion, and DST facts from the IANA tzdb.
Minimal chat-first app with durable threads, actions, and the app-agent loop
