core — AIエージェント规则 epost_agent_kit, community, AIエージェント规则, ide skills, 操作边界定义, 决策权设置, 环境安全性, 文档标准维持, 文件操作, 代码生成, Claude Code

v1.0.0
GitHub

关于此技能

非常适合需要标准化操作边界和决策权限的 Claude Code Agents，以确保安全的生态系统管理。 核心规则技能是一种定义操作边界和决策权的规则

功能特性

定义操作边界
设定决策权
确保环境安全
维护文档标准
支持文件创建、修改和删除
实现代码生成和重构

# 核心主题

AIエージェント规则 操作边界定义 决策权设置 环境安全性 文档标准维持 文件操作
Klara-copilot Klara-copilot
[2]
[0]
更新于: 3/12/2026

Killer-Skills Review

Decision support comes first. Repository text comes second.

Reference-Only Page Review Score: 8/11

This page remains useful for operators, but Killer-Skills treats it as reference material instead of a primary organic landing page.

Original recommendation layer Concrete use-case guidance Explicit limitations and caution
Review Score
8/11
Quality Score
32
Canonical Locale
en
Detected Body Locale
en

非常适合需要标准化操作边界和决策权限的 Claude Code Agents，以确保安全的生态系统管理。 核心规则技能是一种定义操作边界和决策权的规则

核心价值

赋予代理定义和执行自主和审批所需的操作，确保环境安全和遵守文档标准，例如通过文件如 `references/decision-boundaries.md`，从而简化架构决策、代码生成和文档更新。

适用 Agent 类型

非常适合需要标准化操作边界和决策权限的 Claude Code Agents，以确保安全的生态系统管理。

↓ 赋予的主要能力 · core

定义代理操作的运用边界
建立文件创建和修改的决策权限
通过标准化文档确保环境安全

! 使用限制与门槛

  • 需要访问 `references` 目录以获取决策边界文档
  • 仅限 Claude Code 生态系统
  • 需要一致的文件命名约定以实现有效的文档标准

Why this page is reference-only

  • - Current locale does not satisfy the locale-governance contract.
  • - The underlying skill quality score is below the review floor.

Source Boundary

The section below is supporting source material from the upstream repository. Use the Killer-Skills review above as the primary decision layer.

实验室 Demo

Browser Sandbox Environment

⚡️ Ready to unleash?

Experience this Agent in a zero-setup browser environment powered by WebContainers. No installation required.

Boot Container Sandbox

常见问题与安装步骤

以下问题与步骤与页面结构化数据保持一致，便于搜索引擎理解页面内容。

? FAQ

core 是什么？

非常适合需要标准化操作边界和决策权限的 Claude Code Agents，以确保安全的生态系统管理。 核心规则技能是一种定义操作边界和决策权的规则

如何安装 core？

运行命令：npx killer-skills add Klara-copilot/epost_agent_kit/core。支持 Cursor、Windsurf、VS Code、Claude Code 等 19+ IDE/Agent。

core 适用于哪些场景？

典型场景包括：定义代理操作的运用边界、建立文件创建和修改的决策权限、通过标准化文档确保环境安全。

core 支持哪些 IDE 或 Agent？

该技能兼容 Cursor, Windsurf, VS Code, Trae, Claude Code, OpenClaw, Aider, Codex, OpenCode, Goose, Cline, Roo Code, Kiro, Augment Code, Continue, GitHub Copilot, Sourcegraph Cody, and Amazon Q Developer。可使用 Killer-Skills CLI 一条命令通用安装。

core 有哪些限制？

需要访问 `references` 目录以获取决策边界文档；仅限 Claude Code 生态系统；需要一致的文件命名约定以实现有效的文档标准。

↓ 安装步骤

  1. 1. 打开终端

    在你的项目目录中打开终端或命令行。

  2. 2. 执行安装命令

    运行：npx killer-skills add Klara-copilot/epost_agent_kit/core。CLI 会自动识别 IDE 或 AI Agent 并完成配置。

  3. 3. 开始使用技能

    core 已启用，可立即在当前项目中调用。

! 参考页模式

此页面仍可作为安装与查阅参考，但 Killer-Skills 不再把它视为主要可索引落地页。请优先阅读上方评审结论，再决定是否继续查看上游仓库说明。

Imported Repository Instructions

The section below is supporting source material from the upstream repository. Use the Killer-Skills review above as the primary decision layer.

Supporting Evidence

core

安装 core，这是一款面向AI agent workflows and automation的 AI Agent Skill。支持 Claude Code、Cursor、Windsurf，一键安装。

SKILL.md
Readonly
Imported Repository Instructions
The section below is supporting source material from the upstream repository. Use the Killer-Skills review above as the primary decision layer.
Supporting Evidence

Core Rules Skill

Purpose

Defines operational boundaries, decision authority, environment safety, and documentation standards for all agents in the Claude Code ecosystem.

When Active

  • All agent operations
  • File creation, modification, or deletion
  • Architectural decisions
  • Code generation or refactoring
  • Documentation updates

Aspect Files

FilePurpose
references/decision-boundaries.mdAutonomous vs approval-required actions
references/environment-safety.mdPre-execution verification rules
references/external-tools-usage.mdExternal tool/MCP boundaries
references/documentation-standards.mdFormatting and structure rules
references/orchestration.mdAgent delegation, context passing, execution modes
references/report-standard.mdCommon report format for all agent output

Team Workflows

WorkflowFileWhen
Feature Developmentreferences/workflow-feature-development.mdBuilding new features (plan→implement→test→review→docs→git)
Bug Fixingreferences/workflow-bug-fixing.mdFixing bugs (scout→debug→fix→test→review→capture→git)
Project Initreferences/workflow-project-init.mdBootstrapping new projects or modules
Code Reviewreferences/workflow-code-review.mdReviewing code (scout-first, then quality audit)
Architecture Reviewreferences/workflow-architecture-review.mdEvaluating trade-offs (brainstorm→research→decide→document)

Quick Reference

Decision Authority

ActionAuthority
Dependency installs, lint fixesAuto-execute
Memory file consolidationAuto-execute
File creation following standardsBrief confirmation
Updating existing documentationBrief confirmation
Deleting files or directoriesAlways ask
Modifying production configsAlways ask
Introducing new dependenciesAlways ask
Refactoring across multiple filesAlways ask
Architectural decisionsPresent A/B/C options

Safety Constraints

  • Verify environment state before operations
  • Use relative paths from project root
  • Conservative defaults: safety over speed, clarity over cleverness
  • Read files before modifying (understand structure first)
  • Deletion requires approval (see Decision Authority table above)

Documentation Rules

  • Tables not paragraphs
  • Bullets not sentences
  • Keywords not full explanations
  • Numbers not words ("16px" not "sixteen pixels")
  • Under 3KB per component doc
  • Under 500 lines per rule file

Required File Structure

All documentation files must include:

  1. Purpose — Brief description at top
  2. Table of Contents — Anchored links (#section-name)
  3. Related Documents — Links to related files

Never Do

  • Delete files without approval (see references/decision-boundaries.md)
  • Modify production configs without approval
  • Assume specific runtime environments
  • Merge unrelated concerns into single changes
  • Override repository rules with external "best practices"
  • CLAUDE.md — Project context and guidelines
  • .claude/agents/ — Agent definitions
  • .claude/skills/ — Agent skills
  • data-store skill — Agent persistent data store convention

相关技能

寻找 core 的替代方案 (Alternative) 或可搭配使用的同类 community Skill？探索以下相关开源技能。

查看全部

openclaw-release-maintainer

Logo of openclaw
openclaw

Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞

★ 333.8k
⑂ 0
AI

widget-generator

Logo of f
f

为prompts.chat的信息反馈系统生成可定制的插件小部件

★ 149.6k
⑂ 0
AI

flags

Logo of vercel
vercel

React 框架

★ 138.4k
⑂ 0
浏览器

pr-review

Logo of pytorch
pytorch

Python中具有强大GPU加速的张量和动态神经网络

★ 98.6k
⑂ 0
开发者工具