Agentic AI Prompt Research
StaleDescription
Research into how agentic AI coding assistants work, reconstructing and analyzing agent prompt engineering.
Key Features
- Prompt reconstruction - reverse-engineers system prompts of major AI coding assistants
- Mechanism analysis - parses agent loops, tool calling, and decision logic
- Comparative study - compares implementation differences across agent harnesses
- Open data - provides reproducible research data and methodology
- Continuously updated - tracks prompt changes in latest agent tools
Use Cases
💡 Understand the internal workings of AI coding assistants
💡 Optimize your own agent prompt strategies
💡 Learn best practices in agent system design
Categories
Quick Start
# Clone the research repo
git clone https://github.com/Leonxlnx/agentic-ai-prompt-research
cd agentic-ai-prompt-research
# Read README for analysis results