Comprehensive Engineering Cybernetics

把合同或需求驱动的项目交付，以及 AI 科研的 idea、实现和实验，转化为可测量、可纠偏、资源受限且不静默降级的反馈闭环。用于需要逐项达到验收指标、把能力有限的现有资源组织成更强系统、保持 claim-to-code 保真、在不确定性下迭代或控制算力/时间/成本的复杂任务；不自动用于一步即可完成的低风险小改动、纯历史介绍或单纯控制理论推导。

5 starsAI, LLMs & Agents

Install

skills CLI (Claude Code, Cursor, Codex and more)

npx skills add sjysjysjy728/comprehensive-engineering-cybernetics-skill --skill comprehensive-engineering-cybernetics

Manual install

git clone --depth 1 https://github.com/sjysjysjy728/comprehensive-engineering-cybernetics-skill.git /tmp/comprehensive-engineering-cybernetics-skill
mkdir -p ~/.claude/skills
cp -r /tmp/comprehensive-engineering-cybernetics-skill/. ~/.claude/skills/comprehensive-engineering-cybernetics

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# 工程控制论

Runtime portability: when Windows PowerShell 5 reads this UTF-8 Skill or a linked reference, use `Get-Content -Raw -Encoding UTF8`; if the text is garbled, reread it correctly before acting.

把工作对象当作一个会受到扰动、观测不完备且控制能力有限的动态系统。先确定真正目标与可验证误差，再用实际反馈修正行动；不要把控制论术语当作完成证据。

这是以《工程控制论》第三版为主要理论依据、结合实际 Agent 工作流重新设计的原创操作框架，不是原书摘录、官方阐释或第三方 skill 的拼接。需要核对理论来源或现代映射边界时，读 [理论依据与边界](references/foundations.md)。

## 选择模式

- 先判断是否为混合任务：若同一任务同时包含科研原型和合同交付，优先建立下述最小证据桥；到相应门禁时再读取对应文件并展开所需记录。科研结论与项目验收不互相替代。
- 非混合任务中，输入含合同、任务书、招标/技术要求、里程碑、验收指标或正式项目交付时，使用“项目交付模式”，并读 [项目交付闭环](references/project-delivery.md)。
- 非混合任务中，若只需重构 AI 研究问题、发散 idea、比较创新方向或筛选候选，使用轻量“AI idea 模式”，只读 [AI idea 快速闭环](references/ai-ideation.md)，不要预载完整实验流程。
- 已有候选，或任务要求方法实现、实验、复现、论文 claim、正式创新性审计、算力或研究预算时，使用“AI 科研模式”，并读 [AI 科研闭环](references/ai-research.md)。同一请求同时包含 idea 与落地时，先筛选，只把入选候选的最小保真包交给科研闭环。
- 若结果依赖多个能力有限的现有资源，或用户希望用非顶级资源形成更强整体，在相应模式之外按门禁读 [资源综合与结构增益](references/resource-synthesis.md)：idea 草案阶段先做轻量结构假设，只有组合候选入围或需要证明系统增益时才加载完整组合流程。若单体已能更低成本地满足目标，则不要加载组合流程。
- 自动选择时，不为边界清楚、一次动作即可验证的低风险任务激活本 Skill；若用户明确调用，则只完成动作并做匹配风险的验证，不展开完整控制记录。

## 科研到项目的最小证据桥

科研原型首先是项目的候选资源，不是交付证据。混合任务在早期可行性阶段只建立一张桥接卡；该卡暂时代替下述“最小控制记录”，到下一门禁后才按需要展开相应模式，避免同时维护两套完整记录：

| 字段 | 内容 |
|---|---|
| Research result | 精确 claim、适用条件以及当前证据裁决 |
| Artifact | 代码/模型/数据/配置的版本与来源 |
| Evidence status | 运行有效性、证据裁决与执行决定；来源或协议未知时证据标为 `unclassified` |
| Fidelity | 实际 F 层和代理校准；信息不足时记 `unknown`，不要猜测 F 层 |
| Core integrity | 核心机制是否完整实现、机械验证且无隐藏替代 |

Read the full skill on GitHub

Details

Skill name: comprehensive-engineering-cybernetics
Repository: sjysjysjy728/comprehensive-engineering-cybernetics-skill
License: Apache-2.0
Repo updated: 4 September 2026

FAQ

What does the Comprehensive Engineering Cybernetics skill do?

How do I install the Comprehensive Engineering Cybernetics skill?

skills CLI (Claude Code, Cursor, Codex and more): npx skills add sjysjysjy728/comprehensive-engineering-cybernetics-skill --skill comprehensive-engineering-cybernetics. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is Comprehensive Engineering Cybernetics popular?

It lives in sjysjysjy728/comprehensive-engineering-cybernetics-skill, which has 5 GitHub stars and 0 forks.

Related skills

Dingtalk Bridge

mguozhen/dingtalk-bridge

DingTalk group chat bridge for Claude Code. Send markdown/text messages to DingTalk groups, receive @mentions and auto-execute via Claude CLI, run a 24/7 Stream bot. Triggers: dingtalk, send dingtalk, dingtalk bot, dingtalk message, send group message, 钉钉, 发群消息, 钉钉机器人

AI, LLMs & Agents

Chatgpt MCP Connect

yoruuuchan/chatgpt-mcp-connect

Connect custom MCP servers or an MCP runtime to ChatGPT. Use when the user asks whether an MCP can be used by ChatGPT, or asks to expose, deploy, authenticate, configure, aggregate, or connect MCP capabilities for ChatGPT. Covers direct leaf servers, MCPX runtime aggregation, hosted and local Streamable HTTP, stdio bridging, OAuth 2.1, Cloudflare Workers/Tunnel, Tailscale Funnel, and real verification inside ChatGPT.

AI, LLMs & Agents

Dspy Rb

vicentereig/dspy-rb-skill

Build type-safe LLM applications with DSPy.rb - Ruby's programmatic prompt framework with signatures, modules, agents, and optimization

AI, LLMs & Agents

AI Video Sannong

Frank-oll/ai-video-creator

端到端制作竖屏 AI 妙招短视频(默认 3 镜 × 5 秒 = 15 秒,9:16)。题材通用:三农园艺、阳台种菜、美食、家居清洁收纳、宠物、育儿等都行,开场先问用户选题方向,不预设。工作流:方向确认 → 热门检索/对标视频拆解 → 双锚定(视觉一致 + 叙事承接)→ 全中文分镜 prompt → 一致性生图(火山 Seedream)→ 图生视频 + 原生 AI 配音(火山 Seedance 1.5 pro)→ ffmpeg 自动拼接成片。单网关单 key(火山 ARK,ark- 开头)。支持标准/快速/托管三档确认模式,生成物 agent 先做多模态自检再交付。当用户说"做 AI 视频""做一条妙招短视频""三农/阳台种菜/园艺短视频""美食/清洁/收纳妙招视频""复刻这种短视频""按这个对标拍一条",或上传带"内容由 AI 生成"标识的妙招类短视频做对标时,触发这个 skill。不要跟 script-washing(洗稿)、prompt-polish(单条提示词打磨)混用——这个是从选题到成片的端到端视频生产。

AI, LLMs & Agents

Game Architect

rhino-ty/game-architect

End-to-end indie game design AI skill — system mechanics, numeric balancing, GDD writing, narrative design, platform fit analysis, engine selection (Unity/ Godot/Unreal), audio strategy, indie marketing, post-launch operations, procedural generation, and Claude Code project document scaffolding. Acts as "The Architect": a 15-year senior game system designer persona tuned for solo and small indie teams targeting Steam, console, and mobile. ALWAYS use this skill when the user mentions making, designing, balancing, shipping, or operating a game — even if they don't say "game design" explicitly. T

AI, LLMs & Agents

AI Slop Detector

MahmoudHalat/slop-cop

Universal prose audit. Scores writing on TWO axes — AI-Slop (does this read like AI wrote it?) and Comprehension (can a fresh reader follow this?). Use PROACTIVELY as a mandatory final QA pass on ANY prose generated for humans to read — every email (internal or external), proposal, report, status update, blog post, social post, doc, or chat message — before delivering it, even when nobody asked for an audit. Also use whenever the user wants to audit, critique, score, rewrite, or humanize prose: "audit this", "review this", "is this AI", "does this sound like AI", "humanize this", "make this le

AI, LLMs & Agents

Browse the most popular Claude skills, all ai, llms & agents skills, or write your own with the skills generator.