Browse the library

Open-source Claude Code skills providing six cognitive firewalls to block AI hallucination, bias, and sloppy reasoning, enhancing agent reliability.

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

This Claude Code plugin automatically offloads large command outputs to the filesystem, keeping the agent's context clean and optimizing token usage.

Boffin is a staff-engineer control layer that routes architectural constraints to an AI agent for specific edits and requires verification of the results.

This collection provides portable AI agent skills and plugins, including the MSW Kernel for Minimum Sufficient Work, to optimize agent workflows.

dotpals

Rikinshah787

This plugin helps users see what their coding agent actually did, verifies test passes, and makes agents fix failing tests.

This Claude Code plugin helps users stay present in AI-driven workflows by ensuring human decision-making is integrated into the process.

This Claude Code plugin suite offers development tools, prompt optimization, and agent workflow management for complex AI tasks.

This skill runs parallel implementations of a feature, reviews them independently, and selects or synthesizes the best result for quality control.

Provides a collection of battle-tested agent skills and Cursor rules to enhance agent workflows and persistent context across Cursor, Claude Code, and other agents.

This Claude Code plugin allows agents to build and manage code-native integrations and custom connectors for the Prismatic.io platform through conversational commands.

This Claude Code plugin enables autonomous software development through a six-phase workflow (Explore, Research, Plan, Act, Validate, Compound) with classifier-driven routing and a self-improving memory system.

This project is a curated collection of professional Claude Code plugins, subagents, and skills, organized into tiers for focused, token-conscious, and modular AI-assisted development.

This Claude Code plugin provides an autonomous, self-learning agent system with 36 specialized subagents and a 4-tier architecture for analysis, planning, execution, and validation of code, focusing on privacy and security.

Provides an installable AI coding workflow for agents like Claude Code and Cursor, focusing on risk-based routing, scoped sub-agents, and evidence-backed completion.

This Claude Code plugin orchestrates AI agents, using a top-tier model as a tech lead for high-level decisions and cheaper subagents for routine tasks, optimizing token usage.

Generates and modifies Metin2 client UI code, including `uiscript` dicts and `ui*.py` classes, from descriptions or screenshots for various AI coding assistants.

Synchronizes configurations, rules, skills, and MCP servers across various AI coding tools and enables agents to learn from past mistakes.

Turns architectural decisions and ADRs into deterministic guardrails that keep coding agents and CI from drifting away from your architecture.

This resource provides Chinese versions of Andrej Karpathy's coding guidelines as rules and prompts for Claude Code, Cursor, and other AI agents to improve code quality.

This plugin marketplace provides a curated collection of commands, specialized agents, and coding rules for Claude Code, Cursor, and other AI coding tools, including todo persistence.

Updates AI coding agents with current technology knowledge, syntax, and APIs to overcome LLM knowledge cutoff limitations, ensuring accurate and modern code generation.

Provides an agent-neutral operating system with 23 skill categories, 28 agents, and 35 commands for 32+ AI coding tools to build and deploy products using one-line prompts.

Enables AI agents to deploy code to any cloud provider using plain English prompts, automatically configuring infrastructure like servers, DNS, and SSL certificates.