Token导航 LogoToken导航TokenDH.com
研究检索需要联网github未标认证来源可访问许可证需确认审计通过

map地图

Agent Skill

map 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

210

周安装

9

GitHub Stars

公开资料未说明

下载量

73
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:map(地图)
来源仓库:https://github.com/cerico/macfair
仓库路径:skills/map
安装命令:
npx skills add https://github.com/cerico/macfair --skill map
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/cerico/macfair --skill map

简介

Map 探索项目代码库并生成 CLAUDE.md 的结构化上下文章节。

  • 适用于新项目初始化或架构梳理,输出包含实体图和领域摘要。
  • 使用 Mermaid 图表展示关键关系,提供术语表和架构概览。
  • 安装前需确认是否会修改现有文件,建议备份后再运行。
  • 输出标记为 <!-- map:start --> / <!-- map:end --> 以实现幂等更新。

SKILL.md

Map

Explore the current project codebase and generate a structured context section in CLAUDE.md. This section includes a domain summary, mermaid entity diagram, key concepts glossary, architecture overview, and gotchas.

Uses <!-- map:start --> / <!-- map:end --> markers for idempotent updates.

Instructions

1. Explore the Codebase

Use a Task agent with subagent_type=Explore (thoroughness: "very thorough") to gather project context. The exploration prompt should cover:

Detect project type from manifest files:

  • package.json (Node/JS/TS) - check for Next.js, Express, Remix, Astro, etc.
  • Cargo.toml (Rust)
  • go.mod (Go)
  • pyproject.toml / requirements.txt (Python)
  • mix.exs (Elixir)
  • Gemfile (Ruby)
  • composer.json (PHP)

Read the data model (if present):

  • Prisma schema (prisma/schema.prisma)
  • SQL migrations (migrations/, drizzle/, db/migrate/)
  • TypeORM/Sequelize entities
  • GraphQL schema files

Map routes and endpoints:

  • Next.js app/ or pages/ directories
  • Express/Fastify route files
  • tRPC routers
  • API route patterns

Identify key directories: Where business logic, components, utilities, services, and config live.

Understand the domain: From schema + routes + README + key files, determine what the app does in business terms.

The Explore agent should return a structured summary with all findings. Keep the exploration focused on understanding, not cataloguing every file.

2. Generate Content

From the exploration results, write the content that goes between markers. Follow this template:

<!-- map:start -->
## Domain

[One paragraph: what this project does, who it's for, and its business domain.]

## Entity Diagram

erDiagram [Core entities and their relationships] [Limit to ~15-20 relationships max - key entities only]


If there's no data model (CLI tools, libraries, config projects), use a flowchart instead:

graph TD [Key components/modules and how they connect]


## Key Concepts

- **[Entity/Concept]**: [One-line description]
- [5-10 entries covering the domain vocabulary]

## Architecture

[Where things live. Key directories and their roles. Data flow summary. Reference actual paths like `src/services/` or `app/api/`.]

## Gotchas

- [Non-obvious patterns, edge cases, or decisions future sessions should know]
- [Unusual config, env requirements, or conventions specific to this project]

3. Write to CLAUDE.md

Determine the CLAUDE.md path: use the project root (current working directory).

Three cases:

No CLAUDE.md exists: Create CLAUDE.md with just the markers and generated content.

CLAUDE.md exists, no markers: Read the file. Append two blank lines, then the markers and content at the end.

Markers already exist: Read the file. Use the Edit tool to replace everything from <!-- map:start --> through <!-- map:end --> (inclusive) with the new markers and content. Content outside markers is never touched.

4. Verify

After writing, read back CLAUDE.md and confirm:

  • Markers are present and properly paired
  • Mermaid diagram syntax is valid (no unclosed blocks)
  • Content is between markers
  • Nothing outside markers was changed (if file pre-existed)

Output

Tell the user:

  • What project type was detected
  • How many entities/routes were found
  • Whether CLAUDE.md was created, appended to, or updated
  • Suggest re-running /map after major schema or architecture changes

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

37.38%
按下载量换算27

Claude

30.1%
按下载量换算22

Cursor

19.9%
按下载量换算15

Gemini CLI

8.89%
按下载量换算6

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills