- AI infrastructure lives in `core/ai/` (models, checkpointer, graph factory). Agent definitions live in `core/agents/` (graph, tools, state per domain).
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 →- AI infrastructure lives in `core/ai/` (models, checkpointer, graph factory). Agent definitions live in `core/agents/` (graph, tools, state per domain).
`index.html` is the runtime entrypoint. The app is browser-only, with script load order documented in `MODULE_MAP.md`. Keep feature code split by responsibility:
**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
This is an out-of-tree ONNXRuntime Execution Provider (EP) that uses AMD's
This is a LazyVim-based Neovim configuration for Neovim 0.12+, migrated from LunarVim.
This file provides guidance to AI coding agents when working with this repository.
freeq is an IRC server designed for agents. Not a chatbot framework bolted onto a messaging platform — the protocol itself treats agents as first-class participants with cryptographic identity, struct
TalentScout is a Next.js application that helps companies find talent and helps talent find jobs. The application uses MongoDB, GraphQL with Apollo, and Next.js with TypeScript and Tailwind CSS.
Customer-facing checkout UI. Related billing areas:
How to avoid burning tokens when nothing is happening. The core principle: **use zero-token bash checks as gates before invoking the LLM**.
This document provides context for AI assistants working on this Hugo static site project.
- Supports NASA Space Apps Exoplanet Hunter via an interactive web viewer; AI/ML runs outside this repo.
This repository is centered on [`supervisor.py`](./supervisor.py), which defines the LangGraph workflow, subgraphs, retrieval helpers, reporting, and the CLI entry point. Domain knowledge lives in [`k
**Always use the project's custom agents and skills for porting work. Never do their jobs manually or with generic agents.**
Reference for intake skill sub-agent dispatch. Read on demand during
>-
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.
This document defines the workflow for AI agents working with the Data Visualizer project under `.claude`.
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
Step 1: RESEARCH (all agents parallel)
> Design intelligence for UI/UX generation and critique; use for layouts, typography, color palettes, tokens, and UX best practices. 50 styles, 21 palettes, 50 font pairings, 20 charts, 9 stacks (Reac
This file provides the Claude Code view of the unified skill repository.
This file is auto-loaded by Codex (OpenAI) and any other agents that read `AGENTS.md`.
For features spanning multiple domains (auth, database, UI, etc.) that need front-loaded technical context before the Feature Forge interview.
This is the reference for building new skills and agents in this project. Read this before creating any SKILL.md or agent definition.
* Investigate existing agents to extract conventions before creating new ones
Framework-agnostic web design and accessibility guidelines based on WCAG 2.2, MDN Web Docs, and modern web platform standards.
**App:** [App Name]
A monorepo for Nuxt modules and plugins development.
**ZOE Solar Accounting OCR** - AI-powered document processing with designOS design system
1. [Tool Registry ↔ Agent Bridge](#bridge)
*Published: December 3, 2025 | Author: Thuong-Tuan Tran*
This file provides the Claude Code view of the unified skill repository.
source: https://learn.microsoft.com/en-us/azure/foundry/agents/how-to/vs-code-agents-workflow-pro-code
Compatible with: Gemini CLI, Claude Code, Cursor, OpenCode, Kimi CLI, Cline, Codex, and any MCP-compatible agent.
Plastic is a lightweight .NET library that implements the command pattern for building validation and execution pipelines. Commands encapsulate a unit of work with validation logic and execution logic
Altertable agents are autonomous data collaborators that operate continuously alongside human teams.
**Category**: ai-ml/agents
PlantUML is an open-source tool that renders UML diagrams from a plain text description language. Diagrams are defined between `@startuml` and `@enduml` tags. PlantUML supports all major UML diagram t
Agent definitions for PPTX automation pipeline.
Install and configure Lumeo no-code video analytics gateway on NVIDIA Jetson. Covers account creation, gateway deployment, camera linking, and building a people detection pipeline. Requires Jetson with JetPack 5.1+ and internet access.
<!-- Part of the skill-forge AbsolutelySkilled skill. Load this file when
This file is loaded when `/start-company` is active. It defines every agent role used
The major cloud providers each publish a Well-Architected Framework -- a set of pillars, design principles, and best practices for building reliable, secure, performant, and cost-effective workloads i
This is a vibe-closer workspace for managing your fundraising outreach and investor pipeline.
This file documents how this AI Persona operates—the rules learned through practice, patterns that work, and lessons that became doctrine.
This role owns pipeline truth, deal progression, and revenue accountability. It protects forecast integrity and clear next steps over activity volume.
>-
Deploy applications and infrastructure to Cloudflare using Workers, Pages, and related platform
'Analyze git repositories to build a security ownership topology (people-to-file), compute