Near-eliminate AI Agent rule violations in long tasks — auto-correct drift before it frustrates you. Pure-engineering zero-LLM hook system for Claude Code / Codex CLI / Cursor. ~50-70ms hook latency, ~2% token overhead in typical dogfood.
npx add-skill jhaizhou-ops/pinruleNear-eliminate AI Agent rule violations in long tasks — auto-correct drift before it frustrates you. Pure-engineering zero-LLM hook system for Claude Code / Codex CLI / Cursor. ~50-70ms hook latency, ~2% token overhead in typical dogfood.
Source: https://github.com/jhaizhou-ops/pinrule
Discovered from GitHub repositories pushed in the last 24 hours for agent skills, Claude/Codex/Gemini workflows, MCP tooling, and adjacent AI-agent automation.
Recallium is a local, self-hosted universal AI memory system providing a persistent knowledge layer for developer tools (Copilot, Cursor, Claude Desktop). It eliminates "AI amnesia" by automatically capturing, clustering, and surfacing decisions and patterns across all projects. It uses the MCP for universal compatibility and ensures privacy
ClaudeHut — Claude Code plugin enforcing a 6-phase agentic workflow (Brainstorm → Spec → Plan → Build → Loop → Learn) for Java/Spring Boot backends. Hook-gated phase enforcement, isolated-context subagents with per-phase models, native .claude/rules/ auto-loading, project memory + reinforcement learning, strict TDD.
LangGraph-based AI development workflow automation with Claude Code integration
codex_skill: ServerChan Turbo WeChat push notifications for Codex workflows
An agent skill for generating, modifying, and reviewing Swift and iOS code with performance as a first-class concern. This skill is built for Swift, SwiftUI, UIKit, and general iOS app engineering.
Experimental Clojure-based AI coding agent in an early development stage, exploring practical workflows for LLM-assisted software development.