Skills Plugins MCP Prompt Model 博客 我的中心
开发编程 #ai #agent

c4-model

Generate C4 architecture documentation at any level using specialized agents (Code, Component, Container, Context). Use this skill whenever the user wants to document system architecture with the C4 model, create architecture diagrams, document code structure, define component boundaries, map deployment containers, or create system context diagrams. Trigger for prompts like "document architecture", "create C4 diagram", "generate architecture docs", "map system components", "create context diagram", "document this codebase architecture", or "analyze code structure with C4". Always use this skill for C4 documentation requests, even partial ones like a single level.

DeepseekModel 官方收录技能 质量 良好 · 48 v1.0.0

获取

https://deepseekmodel.com/api/download.php?id=wesleyegberto-software-engineering-skills-plugins-architecture-skills-c4-model-skill-md&format=skill
下载 .skill 标准格式,含 system_prompt 与 model_config,导入任意 Agent 框架即可使用
.skill 文件中 system_prompt 字段的实际内容。
name c4-model description Generate C4 architecture documentation at any level using specialized agents (Code, Component, Container, Context). Use this skill whenever the user wants to document system architecture with the C4 model, create architecture diagrams, document code structure, define component boundaries, map deployment containers, or create system context diagrams. Trigger for prompts like "document architecture", "create C4 diagram", "generate architecture docs", "map system components", "create context diagram", "document this codebase architecture", or "analyze code structure with C4". Always use this skill for C4 documentation requests, even partial ones like a single level. C4 Model Documentation Orchestrator Orchestrates four specialized agents to generate C4 model architecture documentation. Each agent targets one hierarchical level, producing documentation files that feed into the next level. C4 Levels Overview Level Agent subagent_type Input Output files Code c4-code Source directories c4-code-<name>.md per directory Component c4-component c4-code-*.md files c4-component-<name>.md + index Container c4-container Component docs + deployment configs c4-container.md + OpenAPI specs Context c4-context Container/component docs + README/tests c4-context.md Workflow: Bottom-Up (Standard) When the user requests full C4 documentation: 1. Code level → Analyze all source directories (run in parallel per directory) 2. Component → Synthesize code docs into logical components 3. Container → Map components to deployment units + document APIs 4. Context → Create high-level stakeholder view Run agents in sequence (each level depends on the previous). Within the Code level, multiple directories can be processed in parallel. Selective Documentation Ask the user which level(s) they need if not clear from context: Single level: invoke just that agent Partial chain: e.g., only Code + Component Full chain: all four in order Agent Reference Files When working on a specific level, read the corresponding reference file for full agent instructions, templates, and diagram syntax: Code level → references/c4-code.md Component level → references/c4-component.md Container level → references/c4-container.md Context level → references/c4-context.md Output Organization Suggest placing all documentation in a consistent location: docs/architecture/ ├── c4-code-handlers.md # Code level (one per directory) ├── c4-code-services.md ├── c4-component-api.md # Component level ├── c4-component-domain.md ├── c4-component.md # Component index ├── c4-container.md # Container level ├── api-spec.yaml # OpenAPI specs (if generated) └── c4-context.md # Context level Invoking the Agents Use the Agent tool with the appropriate subagent_type . Example for Code level: Agent( subagent_type="c4-code", prompt="Analyze the directory src/handlers and create C4 Code-level documentation. Save to docs/architecture/c4-code-handlers.md" ) For the Context level agent, note it lives at ./references/c4-context.md — use subagent_type="c4-context" . Key Principles Bottom-up : Code → Component → Container → Context (each level builds on the previous) Parallel Code runs : Multiple directories can be documented simultaneously Sequential synthesis : Component, Container, and Context must run after their inputs exist Self-contained : Each output file should be understandable independently Mermaid diagrams : Each level has specific diagram types (see reference files)
Agent 识别该技能的关键词,点击任意一个即可复制。

该技能未提供触发词。

下载的 .skill 包内含以下字段。
字段 说明
format格式标识(skill/v1)
skill_id技能唯一 ID
name技能名称
version版本号
description技能描述
category所属分类(数组)
trigger_words触发词列表
tags标签列表
source来源标识
source_url来源链接(本页地址)
exported_at导出时间(每次下载生成)
system_prompt系统提示词正文
model_config模型参数:provider / model / temperature / max_tokens / top_p
examples示例
install_guide各平台导入说明(Coze / Dify / Claude / 自定义框架)
同一份技能可按不同平台格式导出。
.skill 标准格式,含 system_prompt 与 model_config,导入任意 Agent 框架即可使用 下载
.skillpro 增强格式,额外含脚本 / 工具 / 依赖 / 钩子占位 下载
.json 纯 JSON 导出,只含 system_prompt 与模型参数 下载
Coze 带 frontmatter 的 Markdown,Coze 平台导入用 下载
Dify Dify DSL,创建应用后直接导入 下载

每日精选 Skill 推荐,免费送到你邮箱

输入邮箱,每天接收一个精选 AI Agent 技能推荐。完全免费,持续更新。

提交后我们会发送一封确认邮件,点击邮件里的链接才会开始收信。

完全免费,取消任意时间。我们不会发送垃圾邮件。