PullRepo

Daily radar for the fastest-growing AI tools & repos

Today's Prompt Engineering: Fastest-Growing Projects — April 22, 2026

The Prompt Engineering space saw significant activity this week, with several repositories gaining traction among developers and researchers. One notable trend is the increasing focus on Claude Code, a popular AI coding agent, with multiple repositories providing prompts, templates, and courses to help users master its capabilities. Additionally, there's a growing interest in multimodal prompt engineering, with repositories exploring text-to-video and image-to-video examples.

repowise-dev/claude-code-prompts is a standout repository this week, boasting a growth score of 49.31 and 952 stars. This collection of independently authored prompt templates for AI coding agents has gained popularity due to its comprehensive coverage of system prompts, tool prompts, agent delegation, memory management, and multi-agent coordination, all informed by studying Claude Code.

ShipWithAI/claude-code-mastery is another rapidly growing repository, with a growth score of 13.86 and 72 stars. This comprehensive free course on Claude Code mastery has attracted attention for its extensive coverage of 16 phases and 55 modules, making it an invaluable resource for those looking to improve their skills.

ZeroLu/awesome-happy-horse is a community-curated repository with a growth score of 7.65 and 61 stars. This hub for Happy Horse model resources provides prompts, news, benchmarks, and sample outputs, showcasing the potential of this AI model for text-to-video and image-to-video applications.

ZeroLu/awesome-gpt-image has gained popularity with a growth score of 6.60 and 81 stars. This curated collection of top GPT Image 2 prompts and examples from creators on X offers inspiration and guidance for those working with this powerful AI tool.

TaewoooPark/NODEPROMPT is an innovative repository with a growth score of 5.50 and 31 stars, focusing on spatial prompt engineering through interactive 3D concept graphs. This unique approach to prompt engineering has garnered interest among researchers and developers exploring new ways to interact with AI models.

Josh-blythe/bordair-multimodal is a notable repository with a growth score of 4.12 and 33 stars, offering an open-source cross-modal and multimodal prompt injection test suite. With over 50,500 attack payloads across various modalities, this resource has attracted attention from those interested in AI security research.

Leonxlnx/agentic-ai-prompt-research is a well-established repository with a growth score of 3.54 and an impressive 2,330 stars. This comprehensive research into agentic AI coding assistants provides insights into reconstructed prompt patterns, agent coordination, and security classification.

OuterSpacee/claude-emotion-prompting has gained traction with a growth score of 2.24 and 61 stars, offering a research-backed prompting toolkit for Claude grounded in emotion vectors. This repository's focus on emotions and sentiment analysis has attracted interest from those exploring the creative possibilities of AI.

JonusNattapong/ClaudeCode-Learning is another repository gaining attention with a growth score of 1.45 and 28 stars, providing resources for learning prompt engineering specifically tailored to Claude Code.
Back to all reports