Klaro

  From Code to Clarity. Instantly.

   

🛑 Project Halted Temporarily

This project has been temporarily halted due to private reasons.

Status Update

• Development is currently on hold.
• We plan to resume work on this project at a future date.
• No support or feature development will take place during this pause.

Thank you for your understanding.

---

Overview

Klaro is an autonomous AI agent designed to solve the problem of neglected technical documentation. It works by autonomously navigating your codebase, performing deep logic analysis, and generating clear, professional, and up-to-date documentation (such as README files and API references) with minimal human intervention.

💡 The Problem

Writing and maintaining documentation is a time-consuming, tedious, and often neglected task. This common issue leads to accumulated technical debt, slow project onboarding for new team members, and overall project inefficiency.

🚀 The Solution

Klaro automates the entire documentation process. It employs an advanced Pure Python ReAct loop coupled with custom analysis tools and a Retrieval-Augmented Generation (RAG) system to ensure the generated documentation is not only accurate but also adheres to specified project style guides.

✨ Core Features (Completed & Planned)

• Autonomous Code Analysis: Implemented using Python's AST (Abstract Syntax Tree) to read the entire file tree, identify key logic (classes, functions), and extract structured data for analysis.
• Style Guide Integration (RAG): (Completed in Stage 3) Learns from your existing project style guides or reference documents, retrieving relevant context from a vector database to match the tone, structure, and style of the final documentation.
• Multi-Tool Orchestration: Uses a Pure Python ReAct (Reasoning and Acting) loop to dynamically plan steps, execute custom tools (list_files, analyze_code, web_search), and self-correct errors during the documentation process.
• Multi-Format Output (Planned): Generates professional Markdown (README.md), API references, and more.
• Smart Model Steering (Planned): Will utilize LangGraph for dynamic model routing (e.g., using GPT-4o mini for file listing and GPT-4o for complex analysis) to optimize API costs.

🛠 Technology Stack

• Core: Python 3.10+ (3.11+ recommended)
• AI Framework: LangChain (Tools & Prompts) & LangGraph (Agent Orchestration)
• Agent Core: LangGraph State Machine (Stage 4 Completed)
• Vector DB/RAG: ChromaDB, OpenAI Embeddings
• Code Analysis: ast (Abstract Syntax Tree)

📦 Installation

Prerequisites

Before installing Klaro, ensure you have the following:

• Python 3.10+ (3.11+ recommended) - Download Python
• OpenAI API Key - Get your API key from OpenAI Platform
• Git - For cloning the repository
• pip - Python package manager (included with Python)

Step 1: Clone the Repository

git clone https://github.com/aethrox/klaro.git
cd klaro

Step 2: Create a Virtual Environment

Creating a virtual environment is recommended to isolate project dependencies.

On Windows:

python -m venv klaro-env
klaro-env\Scripts\activate

On macOS/Linux:

python3 -m venv klaro-env
source klaro-env/bin/activate

Step 3: Install Dependencies

Install all required packages from requirements.txt:

pip install -r requirements.txt

Step 4: Configure Environment Variables

Copy the example environment file and configure your API keys:

cp .env.example .env

Edit .env and add your OpenAI API key:

OPENAI_API_KEY=sk-your-api-key-here

# Optional: LangSmith tracing for debugging (requires LangSmith account)
LANGSMITH_TRACING=false
LANGSMITH_ENDPOINT=https://api.smith.langchain.com
LANGSMITH_API_KEY=your-langsmith-key-here
LANGSMITH_PROJECT=klaro

# Optional: Override default recursion limit (default: 50)
KLARO_RECURSION_LIMIT=50

Important: Never commit your .env file to version control. It's already included in .gitignore.

Step 5: Initialize the Knowledge Base

The knowledge base (ChromaDB vector database) is automatically initialized on first run. No manual setup required.

Step 6: Verify Installation

Run Klaro to verify everything is working:

python main.py

You should see output indicating the agent is analyzing the codebase and generating documentation.

⚙️ Configuration

Environment Variables

Variable	Required	Default	Description
OPENAI_API_KEY	Yes	-	Your OpenAI API key for LLM and embeddings
KLARO_RECURSION_LIMIT	No	50	Maximum agent iterations before timeout
LANGSMITH_TRACING	No	false	Enable LangSmith tracing for debugging
LANGSMITH_API_KEY	No	-	LangSmith API key (if tracing enabled)

For a complete list and detailed documentation of all environment variables, see the Configuration Guide.

Model Selection

By default, Klaro uses gpt-4o for high-quality, accurate outputs. To use a different model (e.g., gpt-4o-mini for cost-effectiveness), edit the LLM_MODEL variable in main.py line 78.

For a detailed comparison of models, cost analysis, and advanced guidance, see the Configuration Guide.

🚀 Usage

Basic Usage

Run Klaro on the current directory:

python main.py

Analyze a specific project:

python main.py /path/to/project

The agent will explore the codebase, analyze code structure, and generate documentation based on your style guidelines.

For detailed usage examples, programmatic usage, integration patterns, and troubleshooting, see the Usage Examples Guide.

License

This project is licensed under the MIT License - see the LICENSE file for details.