Project Submission Template

Provide a brief, high-level summary of the project's purpose, business or personal impact, and what type of agent/workflow it implements.

---

🎯 Goals & Scope

In-Scope

• Detail what features are built.
• List specific tasks this agent handles.

Out-of-Scope

• Identify boundaries to prevent scope creep.

---

🧠 System Architecture

Use this section to outline the cognitive loop or agent orchestration architecture.

Cognitive Flow

• Trigger: Describe the input or event that starts the agent/workflow.
• Planning: How does the system decide what to do? (e.g. ReAct, Plan-and-Solve, hardcoded tree).
• Execution: What actions does it perform?
• Reflection: How does it verify output correctness before delivering?

---

🛠️ Tech Stack & Linux Setup

Environment Requirements

• Python: >=3.10
• Dependencies: List key packages (e.g. langgraph, pydantic, yfinance).
• Database: PostgreSQL (vector search enabled), Redis (caching), etc.

Linux Quickstart

Commands to initialize and run this project:

# 1. Jump to project folder
cd ~/AI_BOOTCAMP/projects/[project_dir]

# 2. Activate Conda environment
conda activate ai_dev

# 3. Install/sync dependencies via UV
uv pip install -r requirements.txt   # or 'uv sync'

# 4. Set up environment variables
cp .env.example .env
nano .env

# 5. Run the entrypoint
python main.py

---

🔑 Configuration & Env Variables

List all required variables to keep in .env (excluding sensitive secrets themselves):

OPENAI_API_KEY=your-api-key
POSTGRES_HOST=localhost
POSTGRES_DATABASE=agent_db

---

📝 Roadmap & Tasks

• [ ] Define system prompts and agent roles.
• [ ] Implement tool definitions and JSON schemas.
• [ ] Connect agent to long-term memory (PostgreSQL Vector).
• [ ] Setup workflow automation trigger (cron or webhook).
• [ ] Add rigorous evaluation and validation tests.