agentic-coder

Agentic Coder 🤖

Your AI-Powered Project Creator & Iterative Developer

---

What is Agentic Coder?

Agentic Coder creates complete, production-ready projects from natural language descriptions. It thinks, plans, codes, tests, and iteratively improves your projects.

✨ Features

• 🤖 Intelligent Modes:
  • Auto Mode: Automatically detects if you need a full project or a simple script.
  • Direct Mode: Quick, single-file code generation and modification.
  • Autonomous Mode: Full project orchestration with planning, execution, and task tracking.
• 📦 Project Templates: Fast-track creation with standard stacks (react-vite, python-fastapi, nextjs).
• 📊 Live Analytics: Track LLM token usage, costs, and agent actions in PostgreSQL.
• 🧠 Intelligent Planning: Break down complex requirements into actionable steps.
• 📋 Live Task Tracking: Real-time progress updates via tasks.md checklist.
• 💾 Organized Workspace: Projects are stored cleanly with hidden artifacts (.agentic/) keeping your code clutter-free.
• 🔌 Extensible: Use as a CLI tool or import as a Python package.
• 🛠️ Configurable: Customize storage location, LLM providers, and more.

🚀 Installation

pip install agentic-coder

⚙️ Configuration

Create a .env file or set environment variables:

# Required
LLM_API_KEY=sk-...                          # API Key for your LLM provider

# Optional
AGENTIC_PROJECTS_DIR=~/.agentic-coder/projects  # Custom storage location
LLM_MODEL=gpt-4o                                # Default model (e.g., gpt-4o, claude-3, etc.)
LLM_MAX_TOKEN=4096                              # Max context window (auto-truncated)

# Analytics (Optional)
ENABLE_ANALYTICS=true
POSTGRES_DB=postgresql+asyncpg://user:pass@localhost:5432/coding_agent

Usage

1. Initialize Environment:

agentic-coder init

2. Create a Project (Auto Mode):

agentic-coder create "Create a React dashboard" 
# -> Detects intent, uses 'react-vite' template, and starts Autonomous Mode.

3. Use a Template Explicitly:

agentic-coder create --template python-fastapi --project my-api

3. Run the Agent:

agentic-coder project run "Build a FastAPI backend with JWT auth"

4. Manage Projects:

agentic-coder project list
agentic-coder project switch my-app
agentic-coder project info

Quick Start (Legacy):

agentic-coder create "FastAPI backend" --mode autonomous

Examples

Create a Full-Stack Project

agentic-coder create "FastAPI backend with:
- User authentication (JWT)
- CRUD operations for todos
- PostgreSQL database
- Comprehensive tests" --mode autonomous

Iteratively Improve

cd projects/my_app/
agentic-coder improve "add rate limiting"
agentic-coder improve "add logging"
agentic-coder improve "optimize database queries"

📦 Python Package

Integrate agentic-coder into your own tools:

import asyncio
from coding_agent_plugin.managers import ProjectManager
from coding_agent_plugin.agents import OrchestratorAgent, CodingAgent

async def main():
    pm = ProjectManager()
    
    # 1. Direct Mode (Single File Generation)
    # Best for quick scripts or specific file creation
    project = pm.create_project("MyScript", "Simple Script")
    coder = CodingAgent()
    await coder.execute({
        "user_prompt": "Create hello.py",
        "project_id": project['id']
    })
    
    # 2. Autonomous Mode (Full Project Orchestration)
    # Best for complex applications requiring planning and multiple files
    auto_project = pm.create_project("MyApp", "Complex App")
    orchestrator = OrchestratorAgent()
    await orchestrator.run_project(
        user_prompt="Build a calculator app with tests",
        project_id=auto_project['id']
    })

if __name__ == "__main__":
    asyncio.run(main())

Supported LLM Providers

OpenAI:

LLM_API_KEY=sk-...
LLM_MODEL=gpt-4o

NVIDIA:

LLM_BASE_URL=https://integrate.api.nvidia.com/v1
LLM_API_KEY=nvapi-...
LLM_MODEL=qwen/qwen3-next-80b-a3b-instruct

Groq:

LLM_BASE_URL=https://api.groq.com/openai/v1
LLM_API_KEY=gsk_...
LLM_MODEL=llama-3.3-70b-versatile

Documentation

• User Guide - Complete walkthrough
• Roadmap - Upcoming features
• Contributing - How to contribute
• Publishing Guide - PyPI deployment

Roadmap

Upcoming Features

• 🔌 MCP Server Integration - Connect custom tools and context
• 📦 Project Templates - Pre-built templates (FastAPI, Next.js, React, etc.)
• 📊 Cost Tracking - Monitor and control LLM costs
• 🔄 Undo/Redo - Easy change rollback

See full roadmap

Support

• 🐛 Report a bug
• 💡 Request a feature
• 💬 Discussions

License

This project is licensed under the MIT License.

What this means:

• ✅ Commercial use allowed
• ✅ Modification allowed
• ✅ Distribution allowed
• ✅ Private use allowed
• ⚠️ Provided "as is" without warranty

See the LICENSE file for full details.

---

Made with ❤️ by Mohamed Abu Basith

⭐ Star us on GitHub • 📦 Install from PyPI