Browse agent skills

Agent memory infrastructure — Volume-based knowledge management for opencode

Agent-work verification, not code review — a deterministic, cross-agent Report Card of what your coding agent actually did.

Arc-code - AI-powered coding CLI with Arc 3.4 Ultra

Official Wam Pay SDK — payment intents, saved cards, subscriptions, refunds, hosted checkout, webhook verification, and POS terminal events.

Pi extension for configuring multiple local LLM inference engine connections.

GemX-specific analytic components, hooks, and types — extracted from `gemx-shopify-app` so they can be shared across repos (dev / staging apps, and future consumers).

Multi-agent collaboration template for OpenCode

Cron scheduling system for pi — port of Claude Code's internal scheduler

Plug-and-play headless React hooks to add multi-platform restreaming (connect socials, go live, live chat + viewers, collaborative split-screen) with zero client backend.

Provider-neutral Stackline AI contracts, server core, RAG, memory, and adapter interfaces.

Generate and sync CLAUDE.md and AGENTS.md for any repo — npx contextmd and your AI assistant finally knows your project

COS Glasses — self-hosted AI heads-up-display server for Even G2 smart glasses, powered by your local Claude Code or Codex CLI

Vectorize architectural floor plans into clean, interactive, measured SVG. A Claude skill.

Practical AI management toolkit: weekly planning, meeting agendas, feedback scripts, performance reviews, and decision briefs.

Plug local-model cartridges into apps with a permissive activation layer. Vercel-AI-SDK-shaped drop-in API (createMachine, generateText, streamText, generateObject, tool) over a backend-agnostic ActivationRuntime adapter.

Jarvis — an agentic coding CLI with pluggable model backends (Anthropic, OpenAI, Ollama, HuggingFace, custom GGUF).

Cosmonapse is the open protocol for building distributed multi-agent/model harnesses

OpenCode parity adapter for katana plugins — enables the same hook implementations to work on both Claude Code and OpenCode

Discord bot bridge for a self-hosted opencode instance

top, but for your running coding agents — a live terminal dashboard of every Claude Code, OpenAI Codex, xAI Grok, Google Gemini, and Antigravity CLI session on your machine.

Browser automation + multi-provider web search for your Pi agent — with automatic screenshot audit trail (optional).

Read the web as clean AIDocuments through Lyrenth, with a Vercel AI SDK tool.

An LM Studio provider plugin for OpenCode — dynamic model discovery, streaming auto-load, and full LM Studio REST API integration

Pi extension package for browser-agnostic automation with Puppeteer-core, Chromium-first and Firefox-ready architecture.

The AI Technical Debt Scanner — diagnostic-only CLI that scores your repo's AI harness across Copilot, Cursor, Codex, Antigravity, Claude Code, and Gemini. No writes, no telemetry.

Thin shared database client for the WebHouse fleet — one DB SDK, every project. Turso/libSQL today; Postgres/Supabase transports later. Pilot: Buddy Cloud (F067).

Shared executor factory and model registry for the AgentMark Vercel AI SDK v4 and v5 adapters. Consumed by those adapters as a regular dependency.

Stealth Chromium build with a drop-in Playwright harness — fetches and launches the ChromiumFish browser.

Clean, highly optimized AI coding agent using spec-first loop

SaQura post-quantum cryptography SDK for JavaScript/TypeScript — byte-compatible with the .NET/Kotlin/Swift SDKs.