Browse and install thousands of AI Agent skills in the Killer-Skills directory. Supports Claude Code, Windsurf, Cursor, and more.
sub-agent-decompose-worker-review is a pipeline automation skill that decomposes tasks, executes them, and reviews the results using a chain of three agents: task-decomposer, task-worker, and reviewer.
dev is a development workflow utility that automates branching, PR management, and review feedback tasks, enhancing the development process.
Proofread is an AI-powered academic review skill that evaluates documents for weaknesses in Methods and Results sections, providing detailed feedback and improvement suggestions.
A Claude Code plugin for reviewing CPython C extensions — finding API misuse, memory safety bugs, compatibility issues, and correctness problems.
create-adr is an AI agent skill that automates the creation of Architecture Decision Records (ADRs) for project documentation and review
react-best-practices is a set of guidelines for optimizing React and Next.js application performance, covering data fetching, code review, and refactoring
review-ai is a skill that enables developers to review and analyze code changes made by AI editors, using tools like git and Brakeman.
commit-work is a skill that enables developers to make logical and scoped commits with descriptive messages, following specific commit styles and rules.
bmad-validate-prd is a skill that activates the BMAD PRD Validation workflow, using a step-file architecture and configuration loading from _bmad/bmm/config.yaml
Iterative-review is a quality assurance workflow that improves OpenCode documentation and settings through a cyclical process of modification, review, and revision.
review-paper is a specialized AI agent skill designed for editors to review and evaluate manuscripts using a dual-agent approach
clean-code-principles is an AI Agent skill that offers a collection of language-agnostic principles and patterns for writing clean, readable, and maintainable code. It is used when writing new functions, refactoring nested code, and during code reviews.