Skill UI

A comprehensive guide to configuring and maximizing the potential of Claude Code, an advanced agentic coding tool. This resource details best practices, setup templates (CLAUDE.md), advanced prompting techniques (e.g., thinking keywords), and debugging strategies for building reliable, complex software autonomously.

Audits and improves CLAUDE.md files across repositories by discovering every relevant file, scoring its sections, reporting quality issues, and preparing targeted additions before any write actions.

A sophisticated, two-stage system designed to capture and internalize user corrections and learnings during ongoing conversations. It automatically detects patterns (e.g., "use X instead of Y") and queues them. Users can manually review and apply these learnings to a persistent knowledge base (CLAUDE.md), ensuring the AI continuously improves its knowledge and adherence to specific user context for future sessions.

Automates Claude Code plugin audits whenever audit or security keywords appear, checking for vulnerabilities, compliance with CLAUDE.md standards, marketplace/MCP requirements, best practices, documentation, and repository hygiene before emitting a structured report.

Runs AgentShield to audit Claude Code configuration files and MCP components for secrets, injection risks, and overly permissive settings, offering CLI, JSON/Markdown/HTML reports, auto-fix, Opus deep analysis, and CI-ready GitHub Action hooks to harden new or existing projects.

Hierarchical Agent Memory (HAM) scopes Claude Code agent context by generating per-directory CLAUDE.md files, tracking token savings, and surfacing analytics via a dashboard so multi-folder projects avoid repeated full scans.

This skill formalizes successful, recurring patterns discovered in auto-memory (MEMORY.md) and graduates them into permanent, enforced rules within the project's core instruction set (CLAUDE.md or scoped rules). It ensures that learned best practices become mandatory project conventions, significantly improving adherence and consistency across all future AI interactions and development cycles. Use this when a temporary note proves to be a vital, repeatable guideline.

This agent enhances AI's auto-memory by analyzing raw learned patterns (from MEMORY.md). It allows users to promote proven knowledge into structured project rules (CLAUDE.md) or extract them into reusable skills. Use it to audit memory health, enforce best practices, and graduate temporary notes into permanent, actionable standards.

A deep expert system for Anthropic's Claude Code CLI. This skill maximizes productivity by managing advanced features like hooks, persistent context via CLAUDE.md, memory management, sub-agent orchestration, and integrating with local development ecosystems. Ideal for complex coding tasks, debugging, and advanced workflow automation.

This skill manages the secure transition of private internal projects to public open-source releases. It executes a three-stage pipeline: Forking (removing secrets), Sanitizing (verifying clean code), and Packaging (generating required documentation like README, setup.sh, and CLAUDE.md). It ensures code safety and readiness before public deployment.

This skill compresses natural language files like CLAUDE.md or todo lists into a concise "caveman" format to reduce input tokens for AI models. It preserves code blocks, technical terms, and structure while removing filler words. The compressed file overwrites the original, with a backup saved in a secure data directory. Triggered via "/caveman-compress <filepath>".

This skill helps developers create highly focused, folder-specific documentation (`CLAUDE.md` and `AGENTS.md`). It systematically analyzes all files in a target directory to extract project purpose, operational constraints, naming conventions, and critical decisions. This deep context ensures that future AI agents (like Claude or Codex) operate with localized knowledge, preventing them from relying on global, generalized instructions and improving their accuracy in that specific domain.