Discover and explore top open-source AI tools and projects—updated daily.
promptdrivenPrompt-Driven Development CLI for AI-assisted software creation
Top 83.5% on SourcePulse
Summary Prompt Driven Development (PDD) CLI streamlines software development by treating natural language prompts as the source of truth. It leverages AI for automated code generation, testing, and prompt management, enhancing developer efficiency through a regenerative workflow.
How It Works
PDD adopts a "prompt-first" philosophy, using AI to translate natural language into code, tests, and examples. Core workflows automate development tasks, emphasizing prompt modification over direct code editing. The sync command orchestrates a complete workflow loop with intelligent decision-making and live feedback, managing state and execution order.
Quick Start & Requirements
Install via uv (uv tool install pdd-cli) or pip (pip install pdd-cli). Prerequisites include Python 3.8+ and, on macOS, Xcode Command Line Tools and Homebrew. Run pdd setup post-installation to configure shell completion, API keys, and local settings. The quickstart involves cloning the repo, setting an LLM API key, running pdd setup, and executing a sample prompt.
Highlighted Details
sync Command: Automates the full PDD workflow loop with intelligent decision-making, live feedback, and state management.--output-cost option generates a CSV detailing AI model usage and costs.--core-dump option for reproducible bug reporting.Maintenance & Community
Bug reporting is handled via pdd report-core, creating GitHub issues with diagnostic bundles. Community examples are shared in cloud mode.
Licensing & Compatibility Licensing information is not explicitly provided in the README.
Limitations & Caveats Patent applications covering PDD workflows are pending. Local execution requires API key configuration; cloud execution requires GitHub authentication. Generated code should be reviewed before execution.
14 hours ago
Inactive
mistralai
aws