Agent skill

claude-code

Stars 0
Forks 0

Install this agent skill to your Project

npx add-skill https://github.com/MemoryReload/mySkillConfigs/tree/main/skills/claude-code

SKILL.md

Claude Code

Claude Code is an agentic coding tool that lives in your terminal. It helps you with software engineering tasks by understanding your codebase, running commands, and editing files.

Overview

Claude Code is an interactive CLI tool that brings the power of Claude 3.5 Sonnet directly into your development workflow. It can:

  • Explore and understand complex codebases
  • Edit files and fix bugs
  • Run tests and execute commands
  • Manage git operations
  • Plan and implement new features

Key Features

Agentic Capabilities

  • Deep Context: Claude Code can read thousands of lines of code to understand the context of your project.
  • Tool Use: It has access to a suite of tools including Bash, Glob, Grep, Read, Edit, Write, and specialized agents.
  • Task Management: Uses a todo list to track progress on complex tasks.
  • Planning: Can enter "Plan Mode" to design implementation strategies before writing code.

CLI & TUI

  • Terminal Interface: Runs directly in your shell, integrating with your existing tools.
  • Slash Commands: Quick access to common actions like /help, /clear, /compact, /cost, /doctor.
  • Keyboard Shortcuts:
    • Esc: Enter/Exit different modes
    • Ctrl+C: Interrupt current action
    • Ctrl+D: Exit Claude Code

Tools & Integration

  • File Operations: Read, write, edit, and search files efficiently.
  • Command Execution: Run shell commands (git, npm, cargo, etc.) securely.
  • MCP Support: Connects to Model Context Protocol servers to extend capabilities (e.g., database access, browser automation).
  • VS Code Integration: Can open files in VS Code.

Usage

Basic Interaction

Simply type your request in natural language.

bash
> Explain the authentication flow in src/auth.ts
> Fix the bug in the login component
> Add a new endpoint for user profile

Slash Commands

  • /help: Show available commands and help text.
  • /clear: Clear the conversation history.
  • /compact: Compact the conversation history to save tokens.
  • /cost: Show token usage and cost for the current session.
  • /doctor: Run health checks on the Claude Code environment.
  • /bug: Report a bug to Anthropic.
  • /init: Initialize Claude Code in a new project.
  • /login: Log in to your Anthropic account.
  • /logout: Log out.
  • /settings: Open the settings configuration.

Configuration

Claude Code can be configured via config.toml (usually in ~/.claude/config.toml or project-specific .claude/config.toml) or through the interactive /settings command.

Common settings:

  • theme: UI color theme.
  • editor: Preferred text editor.
  • auto_context: Whether to automatically add context from open files.

Ignore Files

Claude Code respects .gitignore files. You can also use .claudeignore to specifically exclude files from Claude's context without affecting git.

Architecture

Agents

Claude Code uses specialized agents for different tasks:

  • Planner: Designs high-level plans.
  • Explorer: Searches and analyzes the codebase.
  • Coder: Writes and edits code.
  • Terminal: Executes shell commands.

Tool Security

  • Bash Tool: Executes commands in a persistent shell. Dangerous commands may require user confirmation.
  • File Edits: Edits are applied directly to the filesystem. It's recommended to work in a git repository to easily revert changes.

Common Workflows

Bug Fixing

  1. User reports a bug.
  2. Claude explores the codebase to locate the issue.
  3. Claude creates a reproduction test case (optional but recommended).
  4. Claude implements the fix.
  5. Claude runs tests to verify the fix.

Feature Implementation

  1. User requests a new feature.
  2. Claude enters Plan Mode to design the solution.
  3. User approves the plan.
  4. Claude implements the feature step-by-step, updating the todo list.
  5. Claude verifies the implementation.

Code Explanation

  1. User asks to explain a complex module.
  2. Claude reads the relevant files.
  3. Claude provides a high-level summary and detailed explanation of the logic.

Troubleshooting

  • Permission Denied: Ensure Claude Code has read/write access to the directory.
  • Context Limit: Use /compact or start a new session if the conversation gets too long.
  • Network Issues: Check your internet connection and proxy settings if Claude cannot connect to the API.

Resources

Expand your agent's capabilities with these related and highly-rated skills.

Didn't find tool you were looking for?

Be as detailed as possible for better results