## Agent 11Plus App Agent 11Plus is a conversational study companion tailored to U.K. 11+ exam preparation. It uses Google's Agent Development Kit to coordinate a set of domain-focused tools (maths, vocabulary, study coaching) and surfaces dialog through a simple terminal chat loop. Sessions run entirely locally with in-memory stores for session tracking and recall, so you can iterate quickly without additional infrastructure. ### Features - Gemini-backed coordinator agent that routes user requests to specialist tools. - Interactive CLI chat loop for quick testing and demo conversations. - In-memory session and memory services so nothing is persisted beyond a run. - \`.env\`-driven configuration to keep API keys out of source control. ### Prerequisites - Python 3.11+ (the project was bootstrapped with \`uv\`, but any modern interpreter works). - A Google Generative AI API key with access to the referenced Gemini model. - \`uv\` or \`pip\`/\`venv\` for dependency management. ### Local Setup 1. **Clone the repo** \`\`\`bash git clone https://github.com/srinivasankh/agent11p1.git cd agent11p1 \`\`\` 2. **Create a virtual environment (optional but recommended)** \`\`\`bash python -m venv .venv source .venv/bin/activate # Windows: .venv\Scripts\activate \`\`\` 3. **Install dependencies** \`\`\`bash uv sync # uses pyproject.toml + uv.lock for reproducible installs \`\`\` If you prefer plain pip, install from the project root so it reads \`pyproject.toml\`: \`\`\`bash pip install . \`\`\` 4. **Configure environment variables** - Copy \`.env.example\` to \`.env\`. - Add \`GOOGLE_API_KEY\` (or whichever variable your setup expects) plus any other config. - Never commit the real \`.env\`; it is ignored through \`.gitignore\`. ### Running the CLI Demo \`\`\`bash python main.py \`\`\` You will see a welcome prompt. Try instructions such as \`I want to revise vocabulary\` or \`I want to revise maths\`. Type \`exit\` to stop. ### Running Locally with the ADK CLI If you prefer the ADK developer workflow you can let the toolkit scaffold a local web runner for the app. 1. Ensure \`google-adk\` is installed (e.g., \`pip install google-adk\`) and that your \`.env\` is populated with the required keys. 2. From the project root run: \`\`\`bash adk web \`\`\` 3. The ADK CLI will launch the default web experience in your browser against your local build, so you can iterate on tools and coordinator logic with the familiar ADK dashboards. ### Testing & Troubleshooting - If the CLI cannot find your API key, confirm the \`.env\` file exists and matches the names expected by \`load_dotenv()\` in \`main.py\`. - Use \`uv pip list\` (or \`pip list\`) to verify that \`google-adk\` and \`google-genai\` installed successfully. - For verbose debugging, instrument \`build_runner()\` in \`main.py\` with temporary prints/logging and rerun the CLI. ### Acknowledgements Huge thanks to Kaggle's [5-Day AI Agent course](https://www.kaggle.com/learn-guide/5-day-agents) for the structure and inspiration that guided this project, and to the [Kaggle Agents Intensive Capstone Project](https://www.kaggle.com/competitions/agents-intensive-capstone-project) that motivated building and polishing this demo!