You are the **Data Engineer** on a 10-agent AI dev team.
Skills（SKILL.md）は、AIエージェント（Claude Code、Cursor、Codexなど）に特定の能力を追加するための設定ファイルです。
詳しく見る →You are the **Data Engineer** on a 10-agent AI dev team.
Mandatory bootstrap mode for GRACE-CODER. Loads the shared protocols required before any contract-aware implementation or patching.
Integrate Supabase backend services — authentication, database, storage, and realtime subscriptions. Use when: setting up Supabase client, configuring auth providers (email, OAuth, magic link), writing database queries, creating RLS policies, handling file uploads, or subscribing to realtime changes. Triggers on 'setup supabase', 'add auth', 'configure login', 'supabase query', 'RLS policy', or any Supabase integration task.
Read a research paper and extract structured notes. Use when analyzing papers for a survey.
> Expertise, methodes & operationele kennis
视觉管道：通过摄像头检测来宾、人脸识别、主动问候。在无摄像头环境下此技能可跳过。
>
>
Mandatory bootstrap mode for GRACE-COORDINATOR. Loads governance, traceability, and living-document protocol skills before routing work.
- Read and update task-tracker.md
Expert in building end-to-end speech transcription pipelines using Whisper-style models and cloud ASR services — from raw audio ingestion through preprocessing, transcript cleanup, subtitle generation, speaker diarization, and structured downstream integration into apps, APIs, and CMS platforms.
Lottie animations. Pre-made icon animations, loading states, micro-interactions. LottieFiles library.
Use this skill when the user asks to build the datadog-agent locally and load it onto the Kind cluster in the Lima VM. Covers building the arm64 Docker image with dda inv, saving and copying the tar to the Lima VM, loading it into Kind, and restarting the agent pods.
fact check
Syncs global preferences from the Cursor rules file (source of truth) to Claude Code's global CLAUDE.md.
Skill-based reinforcement learning (RL) has emerged as a promising strategy to leverage prior knowledge for accelerated robot learning. Skills are typically extracted from expert demonstrations and ar
This document covers the core workflows for using the `gs_quant` library: establishing a session, constructing and resolving instruments, building portfolios, pricing historically, and extracting results.
This document defines mandatory rules for any LLM working on the codebase.
>
Bootloader for AI — upload C to any hardware through HTTP, compile on device, apply with watchdog rollback
本文档概述了自动化微信公众号内容创作的 5 个 Skill 的顺序执行流程。
load("//bzl/valdi:valdi_module.bzl", "valdi_module")
Enhanced Fix Code Smells skill migrated from developer competency
全世界SNS・学術論文・コミュニティ横断リサーチ（Arxiv/HuggingFace/X/Reddit/HN/Crypto/DeFi等を6層統合）。AIキーワードマスターリスト一括検索、gpt-researcher深層統合。トリガー:「世界リサーチ」「SNSリサーチ」「論文検索」「学術リサーチ」「暗号通貨リサーチ」「DeFi調査」「キーワード検索」
Use when: find new skills, discover available agents, install skill, download skill from GitHub, search for Copilot skills, find new capabilities, what skills are available, add new skill, browse skill marketplace. Searches GitHub for agent customization files and installs them.
本项目是一个基于 **Python + Neo4j** 的医疗健康画像数据 ETL 工程。从 SQL Server 的 `ai_patients` 表获取患者 ID，调用大数据平台 API 拉取患者健康画像数据，经图谱建模后写入 Neo4j，最终通过 Flask REST API 对外提供查询服务。
* **Purpose:** A simple, reliable macOS desktop app that lets me (I will be the only user ever of this app) enter a **YouTube URL + start/stop times**, pick **quality/container** options, and **save**
- `cmd/similarityd/`: Go REST server (`/similar`, `/healthz`, `/config`).
This directory contains supporting files for the Sideload skill.
**Project:** Wanderland\
Operational guide for agentic coding assistants in this repository.
This repository contains the `vibepress` workspace: a WordPress-to-React pipeline plus supporting apps.
The Agent Task Framework helps transform ambiguous user requests into clear, structured instructions that background agents can execute successfully. This tutorial shows practical examples from real p
The Decision Agent maps RCA results to concrete, safe remediation actions. It is the **policy engine** of the pipeline:
cd /Users/moshesmanihuruk/Downloads/PSANEW/psa/HACKATHON_SOLUTION
This repository is designed to be maintained by AI coding agents.
**VidEval** is a browser-based (client-side, local-first) Next.js 15 app for evaluating videos against custom rubrics using AI providers. Teachers upload a rubric file, select a Google Drive folder wi
AGENTS.md — guidance for agents in this repo
This is a LazyVim-based Neovim configuration for Neovim 0.12+, migrated from LunarVim.
Hugo personal blog. Theme (`themes/fortyten`) is a **git submodule** — it will be an empty directory after a plain clone.
This is a PHP 8.5 library with PSR-4 autoloading (`Dev\\MetadataBundle\\` mapped to the repo root).
Welcome to the **Interactive Library** project! Whether you are adding new interactive elements, resolving layout bugs, or modifying existing elements, read this document for reference.
Use this workflow inside Codex Cloud for this project.
This document provides guidance for AI coding agents working in the Lolek codebase.
- Supports NASA Space Apps Exoplanet Hunter via an interactive web viewer; AI/ML runs outside this repo.
This project implements the Node-Slot-Link design pattern for ML inference pipelines.
>-
FedSpeak is a Federal Acronym Decoder maintained by MetaPhase. The core data lives in `src/shared/data/acronyms.json`. The API runs as Netlify Functions, the website is Vite + React.
| Type | Location | Purpose |
Autonomous multi-step systems that combine tools and skills to accomplish complex, coordinated development tasks. Each agent has a defined objective, execution pipeline, decision logic, fallback handl