Browse agent skills

Drop-in voice + text store assistant for React Native — the VoiceBot web widget, native in your app.

Design Intelligence Framework for AI coding assistants.

Fayz SDK — API client, app params, runtime broker helpers, and shared types for generated Fayz projects

Health diagnosis and fix planning for JS, TS, Python, and mixed repositories.

CLI for WaitSpin sponsored wait-state ads: auth, campaigns, checkout, and publisher installs for VS Code, Claude Code, MiMo Code, and OpenCode.

RAIF — a repairable, JSON-alternative wire format for LLM tool-call output.

Claude Code skill to build Memory Toast flashcard decks (卡包) — with optional AI image generation using your own OpenAI/Gemini key — and upload, update, or publish them to your Memory Toast account.

A CLI tool to create an Express.js starter project.

Repo-native guardrails for agentic coding.

n8n community node for browser-use-cdp-orchestrator (agents web browser-use pilotés en CDP) — généré depuis les schémas Zod de l'orchestrateur

AI-powered React components for smart form validation and natural language table filtering

PSR (Pulsar Syntax Runtime) to TypeScript transformer with Registry Pattern code generation

Agent-Based Modeling canteen simulation with 3D voxel visualization

The Node.js SDK for the Witness Application.

Professional multi-agent codebase auditing and automated repair engine

OS + workspace tools for nebula: fs, shell, code, web, headless browser, skills — with a multi-tier execution sandbox

The Mantle on-chain tools for nebula: transfers, Agni + Merchant Moe swaps, Aave V3 lending, DeFiLlama discovery, and ERC-8004 identity — every write policy + simulation gated

Open Knowledge Format (OKF) for Claude Code — author, maintain & validate portable markdown knowledge bundles for AI agents. Dual: Claude Code plugin + skills.sh.

A [pi-coding-agent](https://github.com/earendil-works/pi-coding-agent) extension that registers a **deepseek** provider via a configurable AI proxy.

Claude Code plugin that delegates boring, low-risk tasks (bulk i18n, trivial style tweaks) to the Gemini CLI — a transparent Claude ⇄ Gemini pipeline.

AI-агентная платформа для командной разработки

Measure MLX quantization quality loss — KL divergence, perplexity, top-token agreement for KV cache and weights

Token-efficient context packages for any AI tool. Terminal-first CLI for understanding, tracing, reviewing, breaking down, and proposing — all backed by a small scanner-built context package.

The Universal AI CLI — One tool. Every model. Every platform.

Tracing SDK for agent applications

Daemonless, per-tab-isolated browser automation CLI for Claude Code. Drives one shared headed Chrome (persistent profile + extensions) over per-target CDP — many parallel sessions, no focus stealing, nothing long-lived to crash.

LLM-maintained Markdown wiki: spec + CLI + graph-navigated learning site (Obsidian Publish alternative)

Local VS Code GitHub Copilot usage, cost, and session insights.

AI Native Slidev deck workflow for technical presentations - includes deck creation, research-backed narrative shaping, slide drafting, visual quality review, and build verification tools

Token-efficient agent harness with tool readiness for Codex, Claude Code, and Antigravity