Browse the library

This collection of skills guides Claude Code and Codex agents to focus on outcomes, manage scope, verify results, and provide explicit evidence for their work.

Refines product specifications using a multi-LLM debate process where models critique and improve the spec until consensus is reached.

Guides Claude Code through spec-driven development by creating structured research, requirements, design, and task files, then executing them iteratively.

This Claude Code meta-agent interviews you to design a complete software architecture, producing a self-contained blueprint for another Claude Code instance to build.

Provides an opinionated set of plugins, skills, and commands to enhance Claude Code with tools for brainstorming, planning, and code review.

Markdown skills for Claude Code and Codex that run a plan, implement and finalize workflow, including review and shipping a change as a pull request.

This plugin acts as a bounded-autonomy harness for AI agents, channeling user intent to guide implementation across various agent platforms like Claude Code and Copilot CLI.

This plugin provides a structured engineering harness for AI coding agents like Claude Code, adding planning, audit gates, and role-based workflow management to make agent output more reliable.

Han is a suite of AI skills and agents for Claude Code that provides specialists for evidence-based planning, test-driven implementation, deep code review, and documentation maintenance, streamlining product engineering workflows.

Implements a governance framework for AI coding agents, enforcing a gated workflow and verifying completion evidence before shipping code.

Composable skills that make AI coding assistants follow a design-first workflow from requirements to implementation and review, with project standards kept in a .lattice folder.

Claude Code plugin with 100+ skills, agents and hooks that enforce review gates and verifiable completion, with skills-only install for Codex, Cursor and others.

Provides reusable agentic loops as open-standard Agent Skills for tasks like autoresearch, data analysis, and code optimization.

Enables AI coding agents to learn and improve over time by remembering past successes and failures from their trajectories.

A lightweight loop harness for Codex, Claude Code and Antigravity where the agent must leave real evidence artifacts for each piece of work before claiming it is done.

A Claude Code skill that enables any Claude model to exhibit Fable 5-grade work discipline through plan-gating, self-verification, and sub-agent routing mechanisms.

A local plugin that opens a browser UI when the agent finishes a plan so you can strike, replace, insert or comment, then approve or send structured feedback back.

A Claude Code sub-agent and custom commands implementing the RIPER-5 workflow (Research, Innovate, Plan, Execute, Review) for structured development.

Skills for agentic coding CLIs that store project direction, next tasks, and a definition of done in the repo so the agent works against them and verifies results.

This Claude Code plugin acts as an intelligent harness for development workflows, ensuring changes pass through a "Write Gate" and tracking project history to align Claude with architectural preferences.

This plugin provides a structured "Plan, Go, Close" workflow with quality gates for AI coding sessions, ensuring agents follow plans and verify changes.

This Claude Code plugin orchestrates a complete AI-native development workflow, from PRD generation and screen specifications to parallel implementation, quality analysis, and verified commits using 11 specialized agents.

This plugin for OpenAI Codex provides development discipline, structured workflows (PABCD), and multi-model subagents, offering 13 specialized dev skills and orchestration for a disciplined development environment.

This plugin provides a comprehensive Outline-Driven Development methodology for Claude Code, featuring a large skill library and AST-based editing.