agentic-context-engine

A Chinese-language AI agent library containing 215 ready-to-use expert personas across 18 departments (engineering, design, marketing, product, etc.), including 50 China-market originals for Xiaohongshu, Douyin, WeChat, Feishu, and DingTalk. Supports 17 AI coding tools including Claude Code, Cursor, and OpenClaw.

Gives AI agents the ability to read and search across Twitter, Reddit, YouTube, GitHub, Bilibili, and XiaoHongShu without paid APIs. Orchestrates upstream tools like yt-dlp, twitter-cli, and Jina Reader to provide a unified web scraping interface for agents.

A curated, auto-generated collection of installable agent skills derived from the fastest-growing AI agent open-source projects on GitHub. A bot scans trending repos every 30 minutes, ranks by stars/day, and produces ready-to-install SKILL.md files.

An auto-generated skill based on the AgentMemory project (rohitg00/agentmemory), providing guidance on implementing persistent memory systems for AI agents from the AI Agent Skills collection by ara.so.

An auto-generated skill for Oh My OpenAgent, an AI agent orchestration framework sourced from the code-yeongyu/oh-my-openagent project within the AI Agent Skills collection.

A framework of 23 production-grade engineering skills for AI coding agents, encoding lifecycle commands, workflow automation, and quality gates for the complete software development lifecycle.

A fast native Rust CLI for browser automation designed for AI agents, featuring accessibility tree snapshots with stable element refs, semantic locators, batch command execution, network interception, and Chrome DevTools Protocol streaming.

A Codex-first 2D game asset workflow that uses AI agents to generate sprite sheets, animated GIFs, transparent PNGs, layered maps, and engine-ready exports for Godot and Unity.

A minimal self-evolving autonomous agent framework that grants LLMs system-level control over browser, terminal, filesystem, keyboard/mouse, and mobile devices, with automatic skill crystallization that turns solved tasks into reusable skills consuming 6x fewer tokens.

Provides fully local long-term memory for AI agents through a 4-tier progressive pipeline (Conversation, Atom, Scenario, Persona), combining symbolic short-term memory with layered long-term recall to cut token usage by up to 61%.