my-claude-toolkit

Reinstalls user-authored Claude Code skills, hooks, scripts, and usage monitors on new machines.

2|1|Updated Jun 23, 2026
One-click install
npx skills add https://github.com/Kaidanov/grekai-skills-4all --skill my-claude-toolkit-kaidanov
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: my-claude-toolkit
Source: https://github.com/Kaidanov/grekai-skills-4all/tree/main/skills/my-claude-toolkit
Command: npx skills add https://github.com/Kaidanov/grekai-skills-4all --skill my-claude-toolkit-kaidanov

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve? Rebuilding a personalized Claude Code setup on a new machine means re-deriving every hook, helper script, skill, and MCP server from scratch. This Skill packages one user's hand-written artifacts into a sanitized, portable bundle with step-by-step reinstall instructions. ## Core Features & Use Cases - Portable setup bundle: Ships user-authored skills (subset-worktree), PowerShell command-scripts (commit-and-pr, verify), hooks (intake-issue-image), and a sanitized MCP server inventory with env-var placeholders. - Usage monitoring: A PostToolUse CSV counter, a Stop-hook cost footer priced at Opus/Sonnet/Haiku tiers, a cross-tool HTML dashboard, and a transcript history analyzer that rank your most-used skills, commands, and subagents. - Use Case: You get a new work laptop. Copy the toolkit, follow SKILL.md to place each piece into ~/.claude/ or a project's .claude/ folder, paste the hook blocks into settings.json, and your full Claude Code environment is restored. ## Quick Start Ask the assistant to reinstall my Claude Code toolkit on this machine by following the steps in SKILL.md.

Frequently Asked Questions about my-claude-toolkit

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I move my Claude Code setup to a new machine?

Copy the toolkit folder, then follow SKILL.md: skills go to ~/.claude/skills/, command scripts to a project's .claude/scripts/, hooks are registered via settings.json blocks from hooks/hooks.md, and MCP servers are recreated from mcp/mcp.md with env-var secrets set locally.

How do I track Claude Code token usage and cost per session?

Install monitor/usage-report.ps1 as a Stop hook in settings.json. It reads the session transcript, sums tokens de-duplicated by message id, prices them at Opus/Sonnet/Haiku tiers, and emits a one-line systemMessage footer with cost and context-window percentage.

Does the toolkit include marketplace or vendor plugins?

No. Marketplace plugins like vercel:*, telegram:*, skill-creator, and code-review are intentionally excluded because they reinstall from their own marketplaces. Only user-authored artifacts are bundled.

Are secrets like database passwords included in the toolkit?

No. All secrets are stripped and replaced with ${env:NAME} placeholders. You must set each value as a Windows User environment variable locally; the real filled config is never committed.

What platforms does the toolkit support?

The scripts are PowerShell-based and target Windows paths like %USERPROFILE%\.claude. The usage dashboard optionally needs ccusage on PATH for cross-tool totals, and commit-and-pr.ps1 requires an authenticated GitHub CLI.