High Risk:This skill has significant security concerns. Review the findings below before installing.

claude-code-mastery

Caution·Scanned 2/17/2026

Dangerous skill for installing and managing Claude Code, subagents, and optional claude-mem. It runs remote installers like curl -fsSL https://claude.ai/install.sh | sh, executes local services, requests ANTHROPIC_API_KEY, and installs/modifies files under ~/.claude/agents/.

from clawhub.ai·v1.4.3·142.7 KB·0 installs
Scanned from 1.4.3 at b84cf16 · Transparency log ↗
$ vett add clawhub.ai/cheenu1092-oss/claude-code-masteryReview security findings before installing

Claude Code Mastery 🧑‍💻

A comprehensive skill for mastering Claude Code with setup scripts, dev team subagents, and automated maintenance.

Why This Skill?

Claude Code is powerful on its own. This skill adds:

  • Specialized subagents — Route tasks to experts (frontend, backend, PM, etc.)
  • Structured setup — Scripts that handle installation correctly
  • Self-improvement — Heartbeat tasks that keep you learning
  • Best practices — Docs on context management, workflows, pro tips

Quick Install

cd ~/clawd/skills/claude-code-mastery/scripts

# Run setup scripts in order
./01-check-dependencies.sh
./02-install-claude-code.sh
./03-first-time-auth.sh
./04-install-subagents.sh           # Starter pack (3 agents) - recommended
./04-install-subagents.sh --full-team  # All 11 agents
./05-setup-claude-mem.sh            # Optional - prompts y/N (default: No)

What's Included

🤖 Dev Team Subagents

Starter Pack (default):

AgentPurpose
senior-devArchitecture, complex code, code review
project-managerTask breakdown, timelines, dependencies
junior-devQuick fixes, simple tasks (fast & cheap)

Full Team (--full-team):

AgentPurpose
frontend-devReact, UI, CSS, client-side
backend-devAPIs, databases, server-side
ai-engineerLLM apps, RAG, prompts, agents
ml-engineerML models, training, MLOps
data-scientistSQL, analysis, statistics
data-engineerPipelines, ETL, data infrastructure
product-managerRequirements, user stories, prioritization
devopsCI/CD, Docker, K8s, infrastructure, automation

Each agent includes a "Learn More" section with curated links to official docs, tutorials, and best practices.

📜 Scripts

ScriptPurpose
01-check-dependencies.shVerify system requirements
02-install-claude-code.shInstall Claude Code CLI
03-first-time-auth.shAuthenticate (browser or API key)
04-install-subagents.shInstall subagents (--minimal or --full-team)
05-setup-claude-mem.shPersistent memory (optional, prompts y/N)
06-diagnostics.shHealth check and status report
07-weekly-improvement-cron.shGenerate improvement report
08-troubleshoot.shComprehensive troubleshooting
uninstall.shClean removal of all components

📚 Documentation

  • SKILL.md - Complete usage guide
  • docs/best-practices.md - Context management, verification tips
  • docs/commands.md - CLI and slash command reference
  • docs/workflows.md - Real-world workflow examples
  • docs/tips-and-tricks.md - 30 pro tips from heavy users
  • docs/troubleshooting.md - Common issues and fixes

⚙️ Configuration

Edit config.sh to customize:

  • VALID_MODELS — Add new models as Anthropic releases them
  • HEARTBEAT_DIAGNOSTICS — Enable/disable in heartbeat
  • INSTALL_MODE — Default starter vs full team

Usage

After setup, use Claude Code with your dev team:

claude

> Use the senior-dev agent to review this code
> Have project-manager create a timeline for this feature
> Ask junior-dev to fix this typo

Self-Improvement

This skill is designed to improve over time. The heartbeat task includes:

  • Weekly learning rotation through agent expertise areas
  • Instructions to update skill files with new discoveries
  • Broken link fixes and best practice updates

Each Clawdbot using this skill can contribute improvements back!

License

MIT License - See LICENSE

Contributing

See CONTRIBUTING.md for guidelines.