Kaia is a unified command-line tool that generates Product Requirements Documents (PRD), Technical Specifications, Action Plans, Milestone Specifications, and Go-To-Market Plans from your product idea using OpenAI's GPT-4.
Kaia is designed as part of a broader automation pipeline that transforms product ideas into actionable development tasks:
- Product Idea → PRD (business requirements and user needs)
- PRD → Technical Specification (system architecture and technical requirements)
- Technical Specification → Action Plan (implementation strategy and setup tasks)
- Technical Specification + Action Plan → Milestone Specifications (detailed technical specs for each development phase)
- PRD → Go-To-Market Plan (marketing strategy and launch planning)
This pipeline ensures that your product vision flows seamlessly from high-level requirements down to specific development tasks, with each step building on the previous one to maintain context and alignment.
kaia/
├── scripts/ # Python automation scripts
│ ├── prd_auto.py # Generate PRD from text input
│ ├── spec_auto.py # Generate Technical Spec from PRD
│ ├── action_plan_auto.py # Generate Action Plan from Technical Spec
│ ├── milestones_auto.py # Generate Milestone Specifications
│ ├── gtm_auto.py # Generate Go-To-Market Plan
│ └── master_auto.py # Master script to run all five
├── templates/ # Templates and instructions
│ ├── prd_instructions.csv # PRD generation template
│ ├── spec_instructions.csv # Technical spec generation template
│ └── gtm_instructions.csv # GTM plan generation template
├── tests/ # Test files
│ └── test_action_plan.py # Action plan script tests
├── output/ # Generated documents (auto-created)
│ └── archive/ # Historical versions of generated documents
├── old/ # Legacy files and previous versions
├── kaia # Unified CLI entry point
├── run.sh # Alternative execution script
├── setup_alias.sh # CLI alias setup script
├── requirements.txt # Python dependencies
├── tuning.txt # AI model tuning parameters and notes
├── forager_ig_input.txt # Example input file for Forager project
├── .env # Environment variables (create this)
├── .gitignore # Git ignore rules
├── LICENSE # Project license
└── README.md # This file
# Clone the repository
git clone <your-repo-url>
cd kaia
# Install dependencies
pip install -r requirements.txt
# Create .env file with your OpenAI API key
echo "OPENAI_API_KEY=your_api_key_here" > .env
# Set up CLI alias (optional)
./setup_alias.shKaia provides a single command with subcommands for each step of the workflow:
./kaia "FridgeFlow: People waste food because they forget what’s in their fridge or when it expires. This mobile app will use image recognition and barcode scanning to track ingredients and suggest recipes. The app looks like a clean grocery dashboard with a “cook now” button that filters by what’s expiring soon."
- Input: Product idea as text OR you may link a txt file.
- Output: PRD, Technical Spec, Action Plan, Milestone Specs, and GTM Plan in the
output/folder
./kaia prd "Build XYZ"- Input: Product idea as text
- Output: PRD markdown file in
output/
./kaia techspec output/prd_v1.md- Input: PRD markdown file (generated from previous step)
- Output: Technical Specification markdown file in
output/
./kaia actionplan output/tech_spec_v1.md --prd-file output/prd_v1.md- Input: Technical Specification markdown file (required), PRD markdown file (optional for extra context)
- Output: Action Plan markdown file in
output/
./kaia milestones output/tech_spec_v1.md --action-plan-file output/action_plan_v1.md- Input: Technical Specification markdown file (required), Action Plan markdown file (optional)
- Output: Comprehensive milestone specifications markdown file in
output/
./kaia gtm output/prd_v1.md- Input: PRD markdown file
- Output: Go-To-Market Plan markdown file in
output/
--version 1— Add a custom version suffix to output files--skip-prd,--skip-spec,--skip-action-plan,--skip-milestones,--skip-gtm— Skip steps (forallpipeline only)
All generated files are saved in the output/ directory with versioned naming:
prd_v1.md- Product Requirements Documenttech_spec_v1.md- Technical Specificationaction_plan_v1.md- Action Plan with setup checklist and milestonesmilestone_specs_v1.md- Detailed milestone specifications for developersgtm_plan_v1.md- Go-To-Market Plan
The output/archive/ directory automatically stores historical versions of generated documents, allowing you to track changes and iterations over time. Each run creates timestamped versions for easy reference.
- Unified CLI: One command for the full pipeline or any step
- Automated Pipeline: Generate all documents in sequence
- Version Control: Automatic timestamp-based versioning with archive system
- Flexible Input: Support for any text-formatted product idea
- Modular Design: Run individual steps or the complete pipeline
- Action Planning: Includes setup checklists and milestone breakdown
- Milestone Specifications: Detailed technical specs for each development phase
- Go-To-Market Planning: Marketing strategy and launch planning
- Clean Output: Professional markdown formatting
- Historical Tracking: Automatic archiving of all generated documents
The tool uses templates located in the templates/ folder:
- templates/prd_instructions.csv - Defines PRD sections and generation prompts
- templates/spec_instructions.csv - Defines technical specification sections and prompts
- templates/gtm_instructions.csv - Defines go-to-market plan sections and prompts
- templates/action_plan_template.md - Defines the action plan generation prompt with structured milestone markers You can customize these templates to match your specific needs.
Create a .env file in the project root:
OPENAI_API_KEY=your_openai_api_key_here
The tuning.txt file contains AI model parameters and notes for optimizing generation quality. You can modify these settings to adjust the output style and detail level.
- Python 3.7+
- OpenAI Python client
- pandas
- python-dotenv
Install with: pip install -r requirements.txt
- API Key Issues: Ensure your
.envfile contains a valid OpenAI API key - Token Limits: The scripts are optimized to handle large documents efficiently
- File Not Found: Make sure input files exist and paths are correct
- Permission Issues: Ensure the CLI script is executable (
chmod +x kaia) - Archive Issues: Check that the
output/archive/directory exists and is writable
- Fork the repository
- Create a feature branch
- Make your changes
- Add tests if applicable
- Submit a pull request
This project is licensed under the MIT License - see the LICENSE file for details.