Reception Aigent
This example demonstrates a simple, polite receptionist agent built with the Aigency framework. The Reception Aigent acts as a front-desk assistant: it greets users, asks brief clarifying questions, and conceptually routes them to the appropriate specialist in a detective agency. This agent runs without MCP services to remain lightweight and fast to deploy.
Use Cases
The Reception Aigent is ideal for intake, triage, and routing.
Greeting & Triage
Welcome visitors and understand their needs in one or two sentences. Example Prompt:
"Hello! I need help with a case involving stolen documents."Agent’s Likely Response: “Welcome! I’m here to help. To route you properly, do you already have evidence (e.g., documents, fingerprints, video)? If yes, the Case Agent is best. If the situation spans multiple teams, the Detective Manager can coordinate a broader investigation. Next steps: 1) Share key evidence, 2) Provide timeline, 3) Identify known witnesses.”
Clarifying Question
Ask just one concise question when the request is ambiguous. Example Prompt:
"We suspect someone inside the company leaked information."Agent’s Likely Response: “Understood. One quick question: do you need contacts or credibility checks from an informant network? If so, the Informant Agent is ideal. If you need evidence analysis and suspect profiling, the Case Agent will take the lead.”
Provide Next Steps
Offer a short checklist after routing. Example Prompt:
"Who should handle a disappearance case with a few witnesses?"Agent’s Likely Response: “Given there are witnesses and multiple leads, the Detective Manager can coordinate the case. Next steps: 1) Share a brief timeline, 2) List witnesses and locations, 3) Attach any media or documents you have.”
System Architecture
A single core agent with no MCP (Model Context Protocol) services.
Reception Aigent
This agent serves as the front desk, ensuring users are welcomed and routed efficiently.
Key Responsibilities:
- Greet users warmly and maintain a professional, friendly tone
- Respond in the same language the user is writing in
- Ask one concise clarifying question if needed
- Explain which specialist best fits the case (Case Agent, Informant Agent, Detective Manager)
- Provide brief and clear next steps
Greet and Route, Provide Next Steps.Diagram of the System
How to Run
Prerequisites
Install Docker and Docker Compose
Follow the instructions for your operating system to install Docker and Docker Compose.
Download the Repository
git clone https://github.com/aigency-project/demo-reception-agentConfigure your environment variables in a .env file in the project root
Visit Google AI Studio to generate your Gemini API key, since this agent uses the gemini-2.0-flash model.
GEMINI_API_KEY=your_gemini_api_key
GOOGLE_GENAI_USE_VERTEXAI=FALSEExecution
Run the following command from the reception_agent directory:
docker-compose up --buildAccess Ports
Once running, the services are available at these ports:
Reception Agent: http://localhost:8082 (Main entry point)A2A Inspector: http://localhost:6007
Monitoring and Observability
- A2A Inspector: Agent inspection tools at http://localhost:6007.
- Logs: The agent generates logs for tracking and debugging.
Extensibility
The system is designed to be easily expanded:
- Add MCPs later: Evidence database lookup, appointment scheduling, intake forms
- New Skills: Intake checklist builder, case summary generator
- New Integrations: Connect to the detective agency’s case management system