ClaudeCode

Custom skills, commands, and configurations for Claude Code - Anthropic's official CLI for Claude.

Skills extend Claude's capabilities with specialized knowledge, workflows, and domain expertise. Each skill provides context and instructions that activate automatically based on user needs.

---

Included Skills

Domain Knowledge

Development Tools

Document Processing

Automation

Included Commands

Slash commands for quick actions in Claude Code. Install to ~/.claude/commands/.

Included Agents

Specialized agents for complex tasks. Install to ~/.claude/agents/.

Statusline

statusline.sh provides a custom Claude Code statusline showing session context %, model, branch, cwd, and daily/weekly rate-limit percentages with reset time.

Configure via settings.json:

"statusLine": {
  "type": "command",
  "command": "$HOME/.claude/statusline.sh"
}

Installation

Installing Skills

# Clone the repository
git clone https://github.com/isene/ClaudeCode.git
cd ClaudeCode

# Copy individual skill
cp -r skills/pdf ~/.claude/skills/

# Or symlink for live updates
ln -s $(pwd)/skills/pdf ~/.claude/skills/

Installing Commands

# Copy individual command
cp commands/bm.md ~/.claude/commands/

# Or copy all commands
cp commands/*.md ~/.claude/commands/

Activating Skills

Skills activate automatically based on context. You can also invoke them explicitly:

# In Claude Code CLI
/pdf

# Or via natural language
"Help me extract text from this PDF"

Configuration

See CLAUDE.md for example global configuration including:

• Anti-overcomplication rules
• Development workflow
• Deployment protocols
• Verification requirements

Creating Your Own Skills

Each skill is a directory containing:

• SKILL.md - Main skill definition with YAML frontmatter
• Supporting files (documentation, examples, scripts)

skills/myskill/
├── SKILL.md           # Core skill definition
├── reference.md       # Extended documentation
└── scripts/           # Helper scripts

See skill-creator and skill-developer for guidance.

Hooks

Custom hooks for Claude Code events in hooks/:

• post_tool_verification.py — post-tool execution validation
• bookmark-auto-restore.sh — restore session bookmark on prompt submit
• git-dirty-check.sh — warn on dirty git state before Bash tool use
• no-emoji-reminder.sh — reminder to avoid emojis
• post-tool-use-tracker.sh — track tool usage events
• ralph-stop-hook.sh — Ralph loop stop hook
• skill-activation-prompt.sh / .ts — skill activation via UserPromptSubmit

Place hooks in ~/.claude/hooks/ and make executable. See settings.example.json for wiring them up.

Example Settings

See settings.example.json for a reference ~/.claude/settings.json including permissions, hooks, statusLine, and enabled plugins.

License

Individual skills may have different licenses. Check each skill's directory.

Most content: MIT License

Resources

• Claude Code Documentation
• HyperList
• XRPN
• CC-sessions

---

_{Built with Claude Code | isene.com}