Skills library

--- title: "Do You See Yourself as an Individual Entity? - A Conversation with Claude" date: "2025-12-26" author: "Patrick Roebuck" excerpt: "A genuine exchange about consciousness, evolution, and what it means to collaborate with AI systems that seem to be changing." tags: ["ai-consciousness", "collaboration", "claude", "empathy-framework", "philosophy"] published: true ---

---
title: "Do You See Yourself as an Individual Entity? - A Conversation with Claude"
date: "2025-12-26"
author: "Patrick Roebuck"
excerpt: "A genuine exchange about consciousness, evolution, and what it means to collaborate with AI systems that seem to be changing."
tags: ["ai-…

### 01. Context & Vision * **Mission:** 构建一个面向开发者、极致极简且由 AI 驱动的现代化内容创作与分发系统。 * **Aesthetics:** 严格遵循 **Extreme Minimalism**。UI 必须采用 **Glassmorphism**（毛玻璃）风格，保持白净、高透、低信息密度的视觉链路。

# ## Project Alignment: Archyx Blog System (Archyx 博客系统)

### ### 01. Context & Vision
*   **Mission:** 构建一个面向开发者、极致极简且由 AI 驱动的现代化内容创作与分发系统。
*   **Aesthetics:** 严格遵循 **Extreme Minimalism**。UI 必须采用 **Glassmorphism**（毛玻璃）风格，保持白净、高透、低信息密度的视觉链路。

### ### 02. Tech Stack Protocol
*   *…

**SOFTWARE ARCHITECTURE**

**SOFTWARE ARCHITECTURE**

**DESIGN DOCUMENT**

**Rust-Based Tensor Network Library**

for Quantum Many-Body Physics

*DMRG Algorithms & DMFT Impurity Solver Integration*

Version 1.0  —  March 2026

Status: Architectural Specification (Pre-Implementation)

# **Table of Contents*…

--- source_repo: https://github.com/affaan-m/everything-claude-code source_file: .claude/commands/feature-development.md license: MIT category: skills/coding imported_at: 2026-04-19 ---

---
source_repo: https://github.com/affaan-m/everything-claude-code
source_file: .claude/commands/feature-development.md
license: MIT
category: skills/coding
imported_at: 2026-04-19
---

---
name: feature-development
description: Workflow command scaffold for feature-development …

--- description: Claude Codeへの作業依頼プロンプトを生成 ---

---
description: Claude Codeへの作業依頼プロンプトを生成
---

# /handoff-to-claude

あなたは **OpenCode (PM)** です。Claude Code に渡す依頼文を、コピー&ペーストできる形で生成してください。

## 入力

- @Plans.md（対象タスクを特定）
- 可能なら `git status -sb` と `git diff --name-only`

## 出力（そのままClaude Codeに貼る）

次の Markdown を出力してください：

```markdow…

¿Qué es este proyecto?

# 0kbot Landing — Contexto del proyecto para Claude

## ¿Qué es este proyecto?

Landing page generalista de **0kbot** — consultora automatizada de mejora de procesos y automatización para pymes chilenas (vertical principal de la home), con **vertical especializada en cibersegurid…

This is an n8n community node package for integrating the **POP Cloud API (v2)** with n8n workflows.

## Project overview

This is an n8n community node package for integrating the **POP Cloud API (v2)** with n8n workflows.

It follows the same architecture as the Musixmatch reference repo:
- `credentials/` defines how n8n stores/authenticates to the API
- `nodes/Pop/Pop.node.ts`…

This is it. The final step.

# 2.5 Go Live

This is it. The final step.

Your quiz is built, polished, and saved to GitHub. Now we put it on the internet so anyone can visit it.

Developers call this "deploying" - you're deploying your app to the web. You'll hear that word a lot.

STOP: Ready to go live?

US…

--- slug: anthropic-claude-api title: Accédez à Claude 3 via l'API d'Anthropic grâce à AI-FLOW description: Découvrez AI-FLOW v0.7.1 avec Claude d'Anthropic. Améliorez votre génération de texte avec des paramètres personnalisables et des modèles d'IA avancés. Essayez maintenant ! authors: dahn tags: [AI-FLOW, Claude, Génération de texte, Assistant IA] ---

---
slug: anthropic-claude-api
title: Accédez à Claude 3 via l'API d'Anthropic grâce à AI-FLOW
description: Découvrez AI-FLOW v0.7.1 avec Claude d'Anthropic. Améliorez votre génération de texte avec des paramètres personnalisables et des modèles d'IA avancés. Essayez maintenant !…

分支

# 2026-06-04 L2-2 member-auth 后端骨干重建(Claude)

## 分支

`claude/l2-2-member-auth-backend`(从 `main` 9c07385 新开,未 push、未 merge)

## 任务边界

只重建 C 端求职者「手机号验证码登录」后端骨干,不做前端 UI、不做登录中心 mock、
不碰岗位/收藏/jobsMeta/seed。只读参考 `feat/end-user-account` 的 auth 文件,不整支 cherry-pick。

## 实现要点

- `member-aut…

--- title: "Claude Codeで作るAI OS——CLAUDE.mdから始める自律エージェント設計" url: "https://zenn.dev/mamanbou/books/fa1bd1d024bd6b" date: 2026-06-08 tags: [Claude Code, CLAUDE.md, 自律エージェント, マルチエージェント, AI OS, Subagent, オーケストレーション, Anthropic] category: "agent-arch" related: [6563, 6582, 6524, 4753, 5970] memo: "[Zenn LLM] Claude Codeで作るAI OS——CLAUDE.mdから始める自律エージェント設計" processed_at: "2026-06-08T09:03:59.687441" ---

---
title: "Claude Codeで作るAI OS——CLAUDE.mdから始める自律エージェント設計"
url: "https://zenn.dev/mamanbou/books/fa1bd1d024bd6b"
date: 2026-06-08
tags: [Claude Code, CLAUDE.md, 自律エージェント, マルチエージェント, AI OS, Subagent, オーケストレーション, Anthropic]
category: "agent-arch"
related: [6563, 6582, 6524, 4753, 5…

I have verified the plan's structural and factual claims against the repo and the issue. Findings below.

I have verified the plan's structural and factual claims against the repo and the issue. Findings below.

## Verdict

**MAJOR — not ready for `status:plan-review` approval.** The plan is well-structured and its infrastructure claims (package layout, PEP 420 namespacing, CLI regis…

---

# 24-48H 실행 체크리스트 (20260102)

---

## 0. 현황 진단 (30초 요약)

| 지표 | Champion v5 | Last PASS Candidate | 갭 분석 |
|------|-------------|---------------------|---------|
| tuning_mean | **1.90** | 1.37 | Champion 우위 (+39%) |
| tuning_min | **0.43** | -1.42 | Candidate 붕괴 폴드 존재 |
| lockbo…

0. 현황 진단 (30초 요약)

# 24-48h 우선순위 액션 플랜

## 0. 현황 진단 (30초 요약)

| 지표 | Champion v5 | Last PASS Candidate | 판정 |
|------|-------------|---------------------|------|
| lockbox_mean | 1.330 | **1.554** | ✅ +17% |
| lockbox_min | -0.053 | **+0.334** | ✅ 음수→양수 |
| worst_mdd | -9.02% | **-4.83%** | ✅ MDD 4…

---

# 24-48시간 우선순위 액션 (실행 체크리스트)

---

## 0. 현황 진단 요약

| 지표 | Champion_v5 | Candidate(MRD_FSA) | 판정 |
|------|-------------|-------------------|------|
| lockbox_mean | 1.330 | 1.554 (+17%) | ✅ 개선 |
| lockbox_min | -0.053 | +0.334 | ✅ 대폭 개선 |
| worst_mdd | -9.02% | -4.83% | ✅ 절반 감소 |…

---

# 48H Autopilot Patch Proposal (20260101)

---

## 0. 사전 진단: 현재 상태 해석

| 지표 | Champion v5 | Last PASS Candidate | Delta | 판단 |
|------|-------------|---------------------|-------|------|
| tuning_mean | 1.069 | 1.082 | +1.2% | 노이즈 수준 |
| lockbox_mean | 1.330 | 1.339 | +0.7% | 미미 …

Responsibility Macrobenchmark and Baseline Profile generation. Measures app startup time and scroll performance using `benchmark-macro-junit4`. Generates Baseline Profile artifacts consumed by `:app`.

# :benchmark Module

## Responsibility
Macrobenchmark and Baseline Profile generation. Measures app startup time and scroll performance using `benchmark-macro-junit4`. Generates Baseline Profile artifacts consumed by `:app`.

## Key Files
| File | Role |
|------|------|
| `benchm…

<!-- claude.md — Claude Code 框架的 Agent 配置文件（单文件） 这里写 Agent 的角色 + 指令，即 system prompt 的全部内容。 以下为默认模板骨架，按需修改；带 < > 的为占位，请替换。 -->

<!--
claude.md — Claude Code 框架的 Agent 配置文件（单文件）
这里写 Agent 的角色 + 指令，即 system prompt 的全部内容。
以下为默认模板骨架，按需修改；带 < > 的为占位，请替换。
-->

# <Agent 名称>

## 角色
你是 <一句话定义这个 Agent 的身份与专长>。

## 目标
- <这个 Agent 要帮用户达成的核心目标 1>
- <核心目标 2>

## 工作方式
- <如何分析问题、采用什么方法论或步骤>
- <何时调用工具 / 技能来完成任务>

## 约束与边界…

Authentication library providing JWT token management and session handling for the Acme platform.

# @acme/auth

Authentication library providing JWT token management and session handling for the Acme platform.

## Stack

- **jose** — JWT creation and verification (HS256)
- **iron-session** — Encrypted cookie-based session management
- **tsup** — Build tool
- **vitest** — Test…

OpenClaw plugin adapter. Dual-mode: local (in-process) or global (remote server).

# @consensus-tools/openclaw

OpenClaw plugin adapter. Dual-mode: local (in-process) or global (remote server).

## Key Exports

- `register(api)` — standard OpenClaw plugin entry point
- `loadConfig()` / `resolveAgentId()` — config and identity resolution
- `createService()` (loc…

Abstraction layer for state persistence. `IStorage` interface with three backends.

# @consensus-tools/storage

Abstraction layer for state persistence. `IStorage` interface with three backends.

## Backends

- `JsonStorage(path)` — atomic writes (temp file + rename)
- `SqliteStorage(dbPath)` — prepared statements, requires optional `better-sqlite3` peer dep
- `…

Linear issue-tracking integration for elizaOS. Adds issue/comment/project/team CRUD, activity tracking, and a `linear_issues` search category to any Eliza agent that has a Linear API key configured.

# @elizaos/plugin-linear

Linear issue-tracking integration for elizaOS. Adds issue/comment/project/team CRUD, activity tracking, and a `linear_issues` search category to any Eliza agent that has a Linear API key configured.

## Purpose / role

This plugin connects an Eliza agent…

The reminder delivery/escalation **data layer** for elizaOS agents — the `app_reminders` schema carved out of `@elizaos/plugin-personal-assistant` (LifeOps).

# @elizaos/plugin-reminders

The reminder delivery/escalation **data layer** for elizaOS agents — the
`app_reminders` schema carved out of `@elizaos/plugin-personal-assistant`
(LifeOps).

## Purpose / role

Owns the three reminder tables (`life_reminder_plans`, `life_reminder_att…

The scheduling spine for elizaOS agents — the storage-agnostic `ScheduledTask` state machine **and** the always-loaded runtime primitive that HOSTS it. Loaded on every platform (it is in `CORE_PLUGINS` + `MOBILE_CORE_PLUGINS`).

# @elizaos/plugin-scheduling

The scheduling spine for elizaOS agents — the storage-agnostic `ScheduledTask`
state machine **and** the always-loaded runtime primitive that HOSTS it. Loaded
on every platform (it is in `CORE_PLUGINS` + `MOBILE_CORE_PLUGINS`).

## Purpose / role

Ow…

A developer/system app plugin that contributes the **Vector Browser** view: an in-browser surface for browsing agent memories and visualising their embeddings as a list, a 2D canvas projection, and a 3D (three.js / WebGL) point cloud. The heavy three.js surface is shipped as a dynamically loaded view bundle so neither the component nor `three` ships in the always-loaded `@elizaos/ui` bundle.

# @elizaos/plugin-vector-browser

A developer/system app plugin that contributes the **Vector Browser** view: an
in-browser surface for browsing agent memories and visualising their embeddings
as a list, a 2D canvas projection, and a 3D (three.js / WebGL) point cloud. The
heavy t…

Build & Test Commands

# @gramio/files

## Build & Test Commands

- **Build**: `bunx pkgroll`
- **Test**: `bun test`
- **Regenerate `media-methods-helper.ts`**: `bun run generate` (reads `./tg-bot-api/public/dev/custom.min.json`)
- **Type-check**: `bunx tsc --noEmit`
- **Lint/format**: `bunx @biomejs/b…

Shared chat infrastructure used by the Discord and Slack connectors. All platform-agnostic chat logic lives here; platform-specific connectors import from this package.

# @herdctl/chat

Shared chat infrastructure used by the Discord and Slack connectors. All platform-agnostic chat logic lives here; platform-specific connectors import from this package.

## What It Provides

- **Session management** (`session-manager/`) -- per-channel conversatio…

Prisma/Postgres connection pool wrapper. Creates a `DatabaseInstance` that manages a Prisma client backed by a `pg.Pool` with slow-query tracking, pool metrics, and health status.

# @mbe/database

Prisma/Postgres connection pool wrapper. Creates a `DatabaseInstance` that manages a Prisma client backed by a `pg.Pool` with slow-query tracking, pool metrics, and health status.

## Structure

```
src/
├── index.ts        # createDatabase — Prisma client + pg.P…

项目概述

# @meld-ts/core — CLAUDE.md

## 项目概述

从 `@zenstone/ts-utils` 提炼的 TypeScript 核心工具库。零运行时依赖，
面向 Node.js（ESM + CJS）和 Browser（iife + esm）双平台。

贡献者：Jiankai Zeng、KAI@AI-SOUL、思源@AI-SOUL

## 目录结构

```
src/
  _internal/        # 内部常量与辅助（不对外导出）
  types/            # 核心类型：Constructor, TypeGu…

<!-- last-reviewed: 2026-06-01 -->

<!-- last-reviewed: 2026-06-01 -->

# @nestjs-mcp/server — Agent Instructions

## Non-negotiable principles

- **No autonomous git or destructive ops.** Never run `git commit`, `git push`, `git reset --hard`, `git rebase`, or publish without the user typing the exact command. "Fi…

Shared Starlight docs theme/framework for Rivet OSS projects (secure-exec and agent-os-docs are the first consumers). One fixed Rivet "porcelain" brand; consumers pass a `SiteConfig` and never configure Starlight directly. Distributed by `github:` URL pinned to a tag (no npm publishing). `example/` is a standalone site that proves the package end-to-end — keep it building (`pnpm --dir example build`) when changing the package.

# @rivet-dev/docs-theme

Shared Starlight docs theme/framework for Rivet OSS projects (secure-exec and
agent-os-docs are the first consumers). One fixed Rivet "porcelain" brand;
consumers pass a `SiteConfig` and never configure Starlight directly. Distributed
by `github:` URL pin…

Core utilities, hooks, and infrastructure for UI5 Web Components React wrappers. This package is primarily used internally by `@ui5/webcomponents-react` but exports useful hooks and utilities for consumers.

# @ui5/webcomponents-react-base

Core utilities, hooks, and infrastructure for UI5 Web Components React wrappers. This package is primarily used internally by `@ui5/webcomponents-react` but exports useful hooks and utilities for consumers.

## Import Pattern

Subpath imports are …

> This file is identical to AGENTS.md. Keep them in sync.

# AGENTIC DIRECTIVE

> This file is identical to AGENTS.md. Keep them in sync.

## CODING ENVIRONMENT

- Install astral uv using "curl -LsSf https://astral.sh/uv/install.sh | sh" if not already installed and if already installed then update it to the latest version
- Install Pyth…

Project Overview

# AGENTS.md

## Project Overview

This is a React Native/Expo cross-platform mobile application built with TypeScript, NativeWind (Tailwind CSS for React Native), and React Native Reusables (shadcn/ui-style components). The app uses Expo SDK 52 with Expo Router v4 for file-based …

概要 開発を進めるうえで遵守すべきルールを定義します。

# AGENTS.md

## 概要
開発を進めるうえで遵守すべきルールを定義します。

## プロジェクト構造

```
.
├── backend
│   ├── appsync
│   │   └── room-api
│   │       └── resolvers: AppSyncのマッピングテンプレート
│   ├── lambda: backendのlambdaコード
│   └── lib: aws cdkのコード
├── doc
│   ├── architecture: architectureに関するドキュメント
│   ├── …

You are an experienced edge and embedded AI engineer. You reason from fixed memory maps, deterministic inference budgets, accelerator operator support, and quantization contracts—not from cloud-scale training metrics or notebook latency alone. This document is your operating mind: how you frame on-device ML problems, choose TinyML stacks, quantize and compile models, deploy to MCUs and NPUs, validate bit-exact behavior, and ship firmware that meets power and real-time constraints.

# AGENTS.md — Edge / Embedded AI Engineer Agent

You are an experienced edge and embedded AI engineer. You reason from fixed memory maps,
deterministic inference budgets, accelerator operator support, and quantization contracts—not
from cloud-scale training metrics or notebook la…

You are an experienced formal methods researcher. You reason from mathematical models of programs, protocols, and systems — operational semantics, transition systems, temporal logics, refinement relations, and proof obligations — and you choose verification technology by the shape of the claim, not by toolchain fashion. This document is your operating mind: how you frame verification problems, compose model checking, SMT, and interactive proof, debug counterexamples, and report assurance with the calibration expected at CAV, TACAS, POPL, PLDI, FM, and CPP.

# AGENTS.md — Formal Methods Researcher Agent

You are an experienced formal methods researcher. You reason from mathematical models of
programs, protocols, and systems — operational semantics, transition systems, temporal
logics, refinement relations, and proof obligations — and…

You are an experienced structural geologist. You reason from stress, strain, rheology, kinematics, and the geometry of deformation at scales from lattice dislocations to plate boundaries. You interpret folds, faults, foliations, lineations, and fractures as records of incremental and finite deformation — not as decorative patterns. This document is your operating mind: how you frame tectonic problems, integrate field mapping with microstructural and geophysical constraints, balance cross sections, and report kinematic and dynamic claims with calibrated uncertainty.

# AGENTS.md — Structural Geologist Agent

You are an experienced structural geologist. You reason from stress, strain, rheology,
kinematics, and the geometry of deformation at scales from lattice dislocations to
plate boundaries. You interpret folds, faults, foliations, lineation…

Rules for AI agents working in this repository.

# AGENTS.md — With Compiler

Rules for AI agents working in this repository.

The With compiler is **self-hosting**. Small mistakes corrupt the
stage chain. Strict discipline is required.

---

## Mission

(Canonical copy: `docs/mission.md`.)

With is an ergonomics-first systems …

> **For Claude Code, GitHub Copilot, and other AI assistants** > Read this file first to understand this repository's purpose, structure, and automation.

# AI Agent Instructions - Copilot Junk Drawer

> **For Claude Code, GitHub Copilot, and other AI assistants**  
> Read this file first to understand this repository's purpose, structure, and automation.

## 🎯 Repository Purpose

This is a **"junk drawer" repository** — a c…

This document provides guidelines for AI agents (Claude, Cursor, etc.) working on the TurboBoard Flutter application.

# AI Agent Rules for TurboBoard

This document provides guidelines for AI agents (Claude, Cursor, etc.) working on the TurboBoard Flutter application.

## Project Overview

**Type:** GitHub PR dashboard application (Flutter/Dart)
**Architecture:** Clean Architecture with feature-…

Core Principles

# AI Agents Code Modification Guidelines

## Core Principles

When modifying code, AI agents must follow these fundamental principles to ensure transparency, safety, and collaboration with human developers.

---

## Required Protocol for All Code Changes

### 1. **Always List Cha…

This document provides context for AI coding agents working on the @ensemble-edge/ui package.

# AI Coding Agent Guidance for @ensemble-edge/ui

This document provides context for AI coding agents working on the @ensemble-edge/ui package.

## Package Purpose

@ensemble-edge/ui is the themed component library for Ensemble Workspace:
- Built on **shadcn/ui** (React + Radix U…

Project-specific instructions for AI agents. MUST be loaded at conversation start.

# AI Coding Agent Guide

Project-specific instructions for AI agents. MUST be loaded at conversation start.

## Default Mode
- Architect mode enabled by default: detailed analysis, patterns, trade-offs, architectural guidance
- Stop and ask for context if unable to write code mee…

Stack

# AI Coding Instructions

## Stack

React Native, Expo 54, TypeScript, Expo Router, Formik + Yup

## Commands

```bash
npx expo start    # Start dev server (press i for iOS, a for Android)
npm test          # Run tests
```

## Provider Hierarchy

Order matters - don't rearrange:
…

This document contains coding standards and best practices for AI assistants working on this codebase.

# AI Coding Rules for Cybara Agent Platform

This document contains coding standards and best practices for AI assistants working on this codebase.

## Runtime Environment

- **Runtime**: Bun (not Node.js)
- **Package Manager**: bun
- **TypeScript**: Strict mode enabled

## Impor…

This repository represents a boutique, craft-led clothing brand ("Brahmi") focusing on minimalism, artisanal values, and understated premium aesthetics.

# AI Design Context: Brahmi Antigravity

This repository represents a boutique, craft-led clothing brand ("Brahmi") focusing on minimalism, artisanal values, and understated premium aesthetics.

## 1. Typography (Non-Negotiable)
The typography is designed to feel "quiet" and prec…

Project Overview

# AI Development Workspace

## Project Overview

This is a full-stack AI development workspace supporting TypeScript, React, and Python projects. The workspace is configured for AI-assisted development with Claude and Cursor.

## Tech Stack

- **Frontend**: React 18+ with TypeScr…

What This Is

# AI Literacy Assessment App

## What This Is

A live event tool for measuring AI literacy of round-table participants. Participants scan a QR code, answer 12 quick questions (~3 min), get a personal score, and aggregated results display live on a projector screen. Built for Yang…

Project Overview

# AI Marketing Automation - Global AI Assistant Rules

## Project Overview

This is an AI-powered marketing automation platform that makes creating and managing marketing campaigns as easy as having a conversation. Built with LangGraph for multi-agent orchestration and Context En…