Handles project setup and configuration for assistant-ui, supporting AI SDK, LangGraph, and custom backend integrations for building AI chat interfaces.
`/ar:setup` command for creating an autoresearch experiment by interactively collecting domain, target file, eval command, metric, and direction, then invoking `setup_experiment.py`. Ships with built-in evaluators like `benchmark_speed`, `test_pass_rate`, and `llm_judge_content`.
The Compound Engineering Plugin provides tools that make each unit of engineering work easier than the last, with cross-platform support for Claude Code, Cursor, OpenCode, Codex, Factory Droid, Gemini CLI, Copilot, and Kiro via a Bun/TypeScript converter CLI.
Powertools for Claude Code covering git workflows, GitHub automation, code quality, and project tooling. Features a modular rules architecture with standalone rule files, opinionated audit workflows with severity-graded reports, and built-in GitLeaks secret detection.
Provider-agnostic operations framework for agentic resources, codifying patterns and practices developed across dozens of agentic application projects.
Production-ready skills for SAP development with Claude Code CLI
Official Sanity agent toolkit providing MCP integration, interactive skills, and context rules for building with Sanity across multiple AI editors and agents.
# Analysis Based on the README, "abilityai-trinity" is a Claude plugin marketplace that appears to be a **platform for deploying and managing autonomous AI agents** rather than a traditional skill...
Enforces human oversight in AI coding workflows with specification-driven development, ensuring sustainable software through structured planning and validation.
A Claude Code plugin that uses hooks to associate each session's changes to a single jj (Jujutsu) commit, making it easy to isolate edits from different Claude sessions.
Hyperspell's plugins and skills for Claude Code