Talk to your call center data - no SQL, no dashboards, just questions.
While Large Language Models excel at reasoning and summarization, they deliver real business value only when connected to structured, governed enterprise data.
In this post, we demonstrate how a Google Agent built with the Agent Development Kit (ADK) and Gemini can query 3CX Call Detail Records (CDRs) stored in BigQuery, allowing supervisors and analysts to talk to their call center data without writing SQL or opening reports.
The result: a conversational interface to your 3CX CDRs that runs locally from a CLI, yet respects Enterprise Security and Access Controls.
Read on!
Why Google ADK is a Natural Fit for AI Agents Development
Google’s Agent Development Kit (ADK) provides a structured framework for building production-grade AI agents that can query a database, pull information from it, and present it in a human readable way without prior knowledge of the database schema, or SQL programming experience. Gemini handles reasoning and natural language understanding, while the ADK ensures all database access remains controlled and observable.
How the Agent-BigQuery Flow Works
At a high level, the system follows this flow:
- User asks a question via CLI
- ADK Runner forwards the request to the agent
- LlmAgent sends the prompt and tool definitions to Gemini
- Gemini decides which tool to call (e.g., execute SQL)
- BigQueryToolset runs the query in BigQuery
- Results are returned to Gemini for formatting
- User receives a clear, structured response
This separation ensures that the LLM always has restricted database access making the solution suitable for real Enterprise environments.
Configuration & Setup
Prerequisites
- Working Python installation
- 3CX Phone System connected to a Google BigQuery Data source
- Gemini API Access
Installation
-
- Get the Python sources from here.
- Install Dependencies
pip install -r requirements.txt - Create a .env file based on .env.example, enter your values:
GOOGLE_CLOUD_PROJECT=your-project-id
GOOGLE_API_KEY=your-gemini-api-key
- Authentication - You can authenticate using either:
- Service Account JSON
- Application Default Credentials
- gcloud auth application-default login
Run the Agent
python main.py
Practical Use Cases
From a simple command-line interface, PBX owners, administrators and call center supervisors can ask questions such as:
- “What datasets are available?”
- “Show me the schema of the cdroutput table”
- “How many calls did …. answer in the last 30 days?”
- “Which of those calls were spam-related or abusive?”
- “Is there any call involving impersonation of the Tax Department?”
- “Generate a performance summary and coaching suggestions for …. .”
- “List all questions I asked during this session.”
The agent behaves like an on demand data-analyst, backed by real CDR data.
Follow Us For Updates
Join the conversation in our dedicated 3CX AI Forum. Follow us on X and LinkedIn to stay-up-to date on latest news and feature releases.
