Today's Prompt Engineering: Fastest-Growing Projects — April 20, 2026
Today's Prompt Engineering, we're seeing a surge in activity around Claude Code and Happy Horse model-related projects, as well as innovative approaches to spatial prompt engineering. The growth scores reflect this trend, with repowise-dev/claude-code-prompts taking the lead at 54.24, closely followed by ShipWithAI/claude-code-mastery's impressive 14.41 score.
repowise-dev/claude-code-prompts is a repository of independently authored prompt templates for AI coding agents, including system prompts, tool prompts, and agent delegation. With 945 stars and a growth score of 54.24, this project is gaining traction due to its comprehensive coverage of Claude Code's capabilities, demonstrating the community's interest in optimizing AI coding assistants.
ShipWithAI/claude-code-mastery offers a free, comprehensive course on mastering Claude Code, comprising 16 phases and 55 modules. With an impressive 89 commits over the past month, this project is growing rapidly (14.41 growth score) due to its value as a learning resource for developers seeking to improve their prompt engineering skills.
ZeroLu/awesome-happy-horse is a community-curated collection of prompts, news, and benchmarks for the Happy Horse model, featuring text-to-video and image-to-video examples. This repository's moderate growth (8.73 score) can be attributed to its role as a centralized hub for exploring the capabilities of the Happy Horse AI.
TaewoooPark/NODEPROMPT is an innovative project that enables spatial prompt engineering through interactive 3D concept graphs. With a growth score of 5.83 and 31 stars, this repository's unique approach is generating interest among developers looking to experiment with novel prompt design methods.
Josh-blythe/bordair-multimodal offers an open-source cross-modal and multimodal prompt injection test suite, backed by research papers from reputable institutions. Its moderate growth (4.90 score) can be attributed to its value as a testing resource for ensuring the security of AI models against various attack payloads.
ZeroLu/awesome-gpt-image is another curated collection, this time focusing on the best GPT Image 2 prompts and examples from top creators on X. With a growth score of 4.00 and 26 stars, this repository's popularity can be attributed to its curation of high-quality prompts for image generation tasks.
Leonxlnx/agentic-ai-prompt-research is an ongoing investigation into how agentic AI coding assistants work, including reconstructed prompt patterns, agent coordination, and security classification. Despite having a relatively low growth score (3.90), this repository's 2,322 stars indicate its importance as a resource for understanding the underlying mechanics of AI coding agents.
OuterSpacee/claude-emotion-prompting is a research-backed prompting toolkit for Claude, grounded in Anthropic's 2026 emotion vectors paper. With a growth score of 2.47 and 60 stars, this repository's moderate growth can be attributed to its value as a specialized tool for crafting emotionally intelligent prompts.
JonusNattapong/ClaudeCode-Learning provides resources for learning prompt engineering with Claude Code, although its description is somewhat limited. Nevertheless, with 28 stars and a growth score of 1.61, this repository remains relevant for those seeking to improve their Claude Code skills.
repowise-dev/claude-code-prompts is a repository of independently authored prompt templates for AI coding agents, including system prompts, tool prompts, and agent delegation. With 945 stars and a growth score of 54.24, this project is gaining traction due to its comprehensive coverage of Claude Code's capabilities, demonstrating the community's interest in optimizing AI coding assistants.
ShipWithAI/claude-code-mastery offers a free, comprehensive course on mastering Claude Code, comprising 16 phases and 55 modules. With an impressive 89 commits over the past month, this project is growing rapidly (14.41 growth score) due to its value as a learning resource for developers seeking to improve their prompt engineering skills.
ZeroLu/awesome-happy-horse is a community-curated collection of prompts, news, and benchmarks for the Happy Horse model, featuring text-to-video and image-to-video examples. This repository's moderate growth (8.73 score) can be attributed to its role as a centralized hub for exploring the capabilities of the Happy Horse AI.
TaewoooPark/NODEPROMPT is an innovative project that enables spatial prompt engineering through interactive 3D concept graphs. With a growth score of 5.83 and 31 stars, this repository's unique approach is generating interest among developers looking to experiment with novel prompt design methods.
Josh-blythe/bordair-multimodal offers an open-source cross-modal and multimodal prompt injection test suite, backed by research papers from reputable institutions. Its moderate growth (4.90 score) can be attributed to its value as a testing resource for ensuring the security of AI models against various attack payloads.
ZeroLu/awesome-gpt-image is another curated collection, this time focusing on the best GPT Image 2 prompts and examples from top creators on X. With a growth score of 4.00 and 26 stars, this repository's popularity can be attributed to its curation of high-quality prompts for image generation tasks.
Leonxlnx/agentic-ai-prompt-research is an ongoing investigation into how agentic AI coding assistants work, including reconstructed prompt patterns, agent coordination, and security classification. Despite having a relatively low growth score (3.90), this repository's 2,322 stars indicate its importance as a resource for understanding the underlying mechanics of AI coding agents.
OuterSpacee/claude-emotion-prompting is a research-backed prompting toolkit for Claude, grounded in Anthropic's 2026 emotion vectors paper. With a growth score of 2.47 and 60 stars, this repository's moderate growth can be attributed to its value as a specialized tool for crafting emotionally intelligent prompts.
JonusNattapong/ClaudeCode-Learning provides resources for learning prompt engineering with Claude Code, although its description is somewhat limited. Nevertheless, with 28 stars and a growth score of 1.61, this repository remains relevant for those seeking to improve their Claude Code skills.