AI tools¶

{
  "model": "opus",
  "effortLevel": "medium",
  "permissions": {
    "allow": [
      "Bash(ls *)",
      "Bash(echo *)",
    ],
    "deny": [
      "Bash(rm -rf *)",
      "Bash(git push *)",
      "Bash(git reset --hard *)"
    ]
  },
  "attribution": {
    "commit": "",
    "pr": ""
  },
  // "statusLine": {
  //   "type": "command",
  //   "command": "~/.claude/statusline.sh"
  // }
}

Claude tools¶

Tools reference - Claude Code Docs

Claude skills¶

skills/skills/skill-creator/SKILL.md at main · anthropics/skills

SKILL.md

---
name: your-skill-name
  # Maximum 64 characters
  # Must contain only lowercase letters, numbers, and hyphens
  # Cannot contain XML tags
  # Cannot contain reserved words: "anthropic", "claude"
description: Brief description of (1) what this Skill does and (2) when to use it
  # Must be non-empty
  # Maximum 1024 characters
  # Cannot contain XML tags
  # Should describe what the Skill does and when to use it
---

# Keep SKILL.md body under 500 lines for optimal performance
# Split content into separate files when approaching this limit

Extend Claude with skills - Claude Code Docs
- Custom commands have been merged into skills. A file at .claude/commands/deploy.md and a skill at .claude/skills/deploy/SKILL.md both create /deploy and work the same way. Your existing .claude/commands/ files keep working. Skills add optional features: a directory for supporting files, frontmatter to control whether you or Claude invokes them, and the ability for Claude to load them automatically when relevant.
- Keep SKILL.md under 500 lines. Move detailed reference material to separate files.
Agent Skills - Claude API Docs

Best practices

Skill authoring best practices - Claude API Docs
Principles
- Set appropriate degrees of freedom: low / medium / high freedom
- Test with all models you plan to use: Haiku / Sonnet / Opus
Skill structure
- Naming conventions: gerund form / noun phrases / action-oriented
- Progressive disclosure patterns: references / domain-specific / conditional details
- Avoid deeply nested references: one level deep from SKILL.md
- Structure longer reference files with table of contents: for reference files > 100 lines
Evaluation and iteration
Checklist for effective Skills

Skill demos and libraries

anthropics/skills: Public repository for Agent Skills

Open issues:

[FEATURE] Add support for subdirectories in skills · Issue #10238 · anthropics/claude-code

AI tools¶

Overview¶

Resources¶

Writing and proofreading¶

Model comparisons¶

AI Chatbots¶

1. Claude Code ¶

Introduction¶

Claude configuration¶

Claude tools¶

Claude skills¶

Built-in commands¶

Claude agents¶

MCP¶

Claude plugins¶

2. Codex ¶

3. Gemini Spark ¶