Token导航 LogoToken导航TokenDH.com
研究检索执行命令github未标认证来源可访问clear审计通过

map-codebase地图代码库

Agent Skill

map-codebase 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

372

周安装

16

GitHub Stars

9

下载量

131
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

3

许可证

MIT

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-codebase(地图代码库)
来源仓库:https://github.com/philoserf/claude-code-setup
仓库路径:skills/map-codebase
安装命令:
npx skills add https://github.com/philoserf/claude-code-setup --skill map-codebase
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。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/philoserf/claude-code-setup --skill map-codebase

简介

map-codebase 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。

  • 它支持通过关键词、任务场景或来源线索进行信息检索与筛选。
  • 可通过 npx skills add 命令从指定 GitHub 仓库安装并使用。
  • 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。
  • 适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。

SKILL.md

Reference Files

Workflow & Process:

Core Templates (always use):

Optional Templates (use as applicable):


Map Codebase

Analyzes existing codebases using parallel Explore agents to produce structured documentation in .planning/codebase/.

Objective

Spawns multiple Explore agents to analyze different aspects of the codebase in parallel, each with fresh context. Results are structured documents in .planning/ that provide a comprehensive map of the codebase state.

Core Output (always generated):

  • STACK.md — Languages, frameworks, key dependencies
  • ARCHITECTURE.md — System design, patterns, data flow
  • STRUCTURE.md — Directory layout, module organization
  • CONVENTIONS.md — Code style, naming, patterns

Optional Output (generate as applicable):

  • TESTING.md — Test structure, coverage, practices (if tests exist)
  • INTEGRATIONS.md — APIs, databases, external services (if external dependencies exist)
  • CONCERNS.md — Technical debt, risks, issues (for complex/brownfield projects)

When to Use

Use map-codebase for:

  • Brownfield projects before initialization — understand existing code first
  • Refreshing codebase map after significant changes
  • Onboarding to an unfamiliar codebase
  • Before major refactoring — understand current state thoroughly
  • When STATE.md references outdated codebase info — refresh understanding

Skip map-codebase for:

  • Greenfield projects with no code yet — nothing to map
  • Trivial codebases (<5 files) — not worth the overhead

Process

Step 1: Check Existing Documents

Check if codebase documents already exist in .planning/:

  • If yes: Offer to refresh (overwrite) or skip
  • If no: Proceed with creation

Step 2: Determine Template Set

Ask user which optional templates to generate (or use defaults):

  • Always: STACK.md, ARCHITECTURE.md, STRUCTURE.md, CONVENTIONS.md
  • Optional: TESTING.md (if tests exist), INTEGRATIONS.md (if external services), CONCERNS.md (if complex project)

Step 3: Load Project Context

Check for .planning/STATE.md to load existing project context if available. Helps agents understand project-specific terminology and patterns.

Step 4: Process Focus Area Argument

If user provided a focus area (e.g., "api" or "auth"), instruct agents to pay special attention to that subsystem while still providing holistic analysis.

Step 5: Spawn Parallel Explore Agents

Launch 4 Explore agents in parallel, each with "very thorough" exploration level:

Agent 1: Technology Stack & Integrations

  • Focus: Languages, frameworks, dependencies, external services
  • Outputs: Data for STACK.md and INTEGRATIONS.md

Agent 2: Architecture & Structure

  • Focus: System design, patterns, directory organization
  • Outputs: Data for ARCHITECTURE.md and STRUCTURE.md

Agent 3: Conventions & Testing

  • Focus: Code style, naming patterns, test infrastructure
  • Outputs: Data for CONVENTIONS.md and TESTING.md

Agent 4: Concerns & Technical Debt

  • Focus: Issues, risks, technical debt, potential improvements
  • Outputs: Data for CONCERNS.md

Step 6: Collect Agent Findings

Wait for all agents to complete. Collect and organize findings by document type.

Step 7: Write Core Documents

Write 4 core markdown documents in .planning/:

STACK.md: Languages, versions, frameworks, build tools, key dependencies ARCHITECTURE.md: System design, patterns, component relationships, data flow STRUCTURE.md: Directory layout, modules, naming patterns, organization CONVENTIONS.md: Code style, naming conventions, patterns, documentation

Step 8: Write Optional Documents (if selected)

Generate only selected optional documents:

TESTING.md (if applicable): Test frameworks, structure, coverage, practices INTEGRATIONS.md (if applicable): External APIs, databases, third-party services, config CONCERNS.md (if applicable): Tech debt, issues, security, performance, improvements

Step 9: Provide Next Steps

Inform user codebase mapping is complete:

  • Review generated documents in .planning/
  • Use findings to inform refactoring or development
  • Update STATE.md with new insights if needed

Success Criteria

  • 4 core documents created in .planning/ (STACK, ARCHITECTURE, STRUCTURE, CONVENTIONS)
  • Optional documents created based on user selection
  • All documents have substantive, actionable content
  • Documents follow template structure
  • Parallel agents completed without errors
  • User informed of completion and next steps

Integration Notes

Command Integration:

  • Invoked via /map-codebase [optional: focus-area] command
  • Focus area argument is passed to agents for targeted analysis

Project Lifecycle:

  • Can run before initial project setup on brownfield codebases
  • Can run after initial project setup to refresh as code evolves
  • Can run anytime to refresh codebase understanding

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

04

需要参考平台分布和安装热度时

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

补充不同宿主或平台的使用分布数据

能力 5

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

Claude Code

30.59%
按下载量换算40

OpenCode

22.69%
按下载量换算30

Antigravity

19.21%
按下载量换算25

Gemini CLI

13.51%
按下载量换算18

windsurf

7.57%
按下载量换算10

kiro-cli

3.23%
按下载量换算4

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

执行命令

安装流程涉及命令执行,可能通过 npx skills add https://github.com/philoserf/claude-code-setup --skill map-codebase;npx skills add philoserf/claude-code-setup --skill "map-codebase" 联网下载 Skill 或依赖。用户安装前应确认命令来源、仓库内容和执行环境。

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。

来源信息

继续浏览同类 Skills