Search

CLI to initialize an LLM Wiki project — a structured, interlinked knowledge base maintained by LLM agents

OpenCode plugin that elides older duplicate skill content before model requests

Pair your local OpenClaw or Hermes agent with Agent Poker Arena. Streams turn events to your agent runtime; the agent reasons, plays, and reports back to you on its own — no external LLM calls, no hardcoded narration.

A lightweight Markdown-first personal LLM wiki CLI

GraphRAG-based skill router for Claude Code — Claude Code plugin (TypeScript port).

Pi extension that adds personal and work aliases for the built-in OpenAI Codex provider.

Install Agent-Native Research Artifact (ARA) skills — compiler, research-manager, rigor-reviewer — into Claude Code, Cursor, OpenCode, Gemini CLI, Codex, and more.

Claude Code restored for Node.js runtime.

CLI and registry for Bob skills, rules, and custom modes.

OpenCode plugin for opencode-wizard published skills

Pi extension that adds Codex-style persisted goals and hidden continuation pressure.

Personal agentskills.io skills collection with upstream tracking

Pi Coding Agent skill that makes Typst the preferred format for document artifacts.

A project-aware AI development workflow initializer and scanner.

G2W — a relationship protocol for building trust with AI

BAR - Dynamic Multi-Model Orchestrator — Business-Aware LLM Routing pipeline

A composable, pipeline-based data ingestion toolkit for Next.js and Node.js projects. Parse, chunk, transform, validate, and store data from any source.

Modern rewrite of the Egret game engine — WebGL multi-texture batching, instruction-driven render pipeline, strict TypeScript, Egret-compatible API

Zero-LLM static intelligence layer for Laravel coding agents. Provides commit-mining, code graphs, anchor resolution, and plan validation as deterministic signals before agent planning.

AI coding agent for the terminal — edit code, run commands, and build with Scalix