muratcankoylan/Agent-Skills-for-Context-Engineering

A comprehensive collection of Agent Skills for context engineering, multi-agent architectures, and production agent systems. Use when building, optimizing, or debugging agent systems that require effective context management.

What it solves

This project provides a comprehensive collection of "Agent Skills" designed to help developers build production-grade AI agent systems. It specifically addresses the challenge of context engineering—the holistic management of a model's limited attention budget (including system prompts, tool definitions, and message history) to prevent performance degradation like the "lost-in-the-middle" phenomenon.

How it works

The project is organized as a library of modular, platform-agnostic skills. Each skill is a directory containing principles, patterns, and practical examples (often using Python pseudocode) that can be integrated into various agent platforms.

It is designed for progressive disclosure, meaning agents only load the skill names and descriptions initially, loading full content only when a skill is activated for a specific task. It is compatible with tools like Claude Code, Cursor, and other Open Plugins hosts via a specific directory layout.

Who it’s for

AI engineers and developers building autonomous agents who need to move beyond simple prompt engineering to architectural patterns for memory, tool design, multi-agent orchestration, and evaluation frameworks.

Highlights

  • Comprehensive Skill Set: Covers foundational context management, architectural patterns (multi-agent, memory systems), and operational skills (evaluation, harness engineering).
  • library of Practical Examples: Includes complete system designs such as a "Digital Brain" personal OS, an X-to-book synthesis system, and a style-transfer SFT pipeline.
  • Platform Agnostic: Works across different agent platforms (Claude Code, Cursor, Codex) using a standardized plugin format.
  • Benchmarked Performance: Includes a "Researcher Operating System" with measured router-benchmark results to optimize how skills are activated by models.

Related

  • Project
    jeremylongshore/tons-of-skills-marketplace

    A model-agnostic agent-skills platform that provides a marketplace of thousands of plugins and skills to extend the capabilities of AI agents.

  • Project
    chujianyun/skills

    A Chinese‑language repo that bundles dozens of reusable “Agent Skills” – small, documented plugins for Claude Code and similar LLM‑agent platforms. Skills cover wiki look‑ups, prompt optimisation, career assessments, diagram generation, document interpretation, audio‑to‑text, image watermarking, web‑scraping, and service run‑books. They are organized under `skills/`, registered in `config/skills.json`, and installed via Claude Code’s marketplace, an installer agent, or the `openskills` CLI.

  • Project
    kangarooking/kangarooking-skills

    A collection of structured, open-source Agent Skills compatible with the Agent Skills standard, providing reusable workflows for content creation, image generation, and AI project engineering.

  • Project
    Pluviobyte/rnskill

    A comprehensive library of modular AI Agent skills for Codex and Claude Code, specializing in automated content creation pipelines for video, social media, and business analysis.

  • Project
    Neeeophytee/finding-unknowns-skills

    A collection of 14 installable skills for AI coding agents that help developers identify requirements gaps and technical unknowns before implementation.