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.mdchecklist. - 💾 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
Support
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
Metadata
Release files for agentic-coder 0.6.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| agentic_coder-0.6.0.tar.gz | 2.2 MB | Without provenance | Details for agentic_coder-0.6.0.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| agentic_coder-0.6.0-py3-none-any.whl | Python 3 | none | any | Details for agentic_coder-0.6.0-py3-none-any.whl |
Total release size: 2.3 MB
Release files / agentic_coder-0.6.0.tar.gz
| Download URL | agentic_coder-0.6.0.tar.gz |
|---|---|
| Size | 2.2 MB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
ad898a0811d6e4ea29e0f03b11fd3fc004605b65cb8757a6c82471bed11957a8
|
|
BLAKE2b-256 checksum How to use checksums |
bc5c650693d6ef25c2c1ffcc11e0815349b946439afdda640b8587875ed995ec
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.10.0
|
Release files / agentic_coder-0.6.0-py3-none-any.whl
| Download URL | agentic_coder-0.6.0-py3-none-any.whl |
|---|---|
| Size | 82.0 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
ae5df8710893949a1b624560b493b97a9d7ce461fe205ae09d8284d8e5a0e85c
|
|
BLAKE2b-256 checksum How to use checksums |
54acee3608cb5478927ddfafd3a4d71cbb2eda69725f88b4d7ba7792c13909ea
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.10.0
|
Security for agentic-coder 0.6.0
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.