CommunityArte e Designgithub.com

hamza-ali-shahjahan/hamzaish

Get into Builder Mode — an AI-native startup factory and OS you point Claude Code (or Cursor/Codex/Windsurf) at.

Funciona comClaude CodeCodex CLICursorWindsurf
npx skills add hamza-ali-shahjahan/hamzaish

Ask in your favorite AI

Open a new chat with this agent skill pre-loaded.

Documentação

/tidy — the cleanup stage

Rot accumulates in every repo: links that point at moved/gitignored files, a key that slipped into a commit, an asset nothing references, a dependency that 404s on install. None of it is visible day-to-day — it surfaces at the worst time (a reader's 404, a failing CI, a leaked key). /tidy is the stage where you find and clear it on purpose.

This is a stage, not a gate. The fast guards (check-changelog, check-assets) block on every push; /tidy is the broad, report-first sweep you run when it's time to clean.

How to run it

bun run tidy                      # the current repo → summary + recommended next steps
bun run tidy <path>               # a specific repo
bun run tidy --all ~/Claude       # every git repo under a folder → one aggregate table
bun run tidy --links              # drill into one category (--secrets --files --deps)

It scans four kinds of rot and shows the extent first (a count per category) so you never get a wall of output:

what it catches
🔗 Linksmarkdown/HTML links to missing files, or that resolve only to a gitignored target (works on the author's disk, 404s for everyone else)
🔑 Secretshigh-confidence committed-key patterns (AWS, Stripe, GitHub, Google, Slack, private-key blocks) — flagged for review, not auto-judged
🗑 Dead filescommitted images/assets that nothing references — safe to prune
📦 Depspackage.json dependencies that don't resolve on npm (the dead-@inngest/sdk class)

The flow (report → recommend → ask → fix)

  1. Run it. Present the summary table — extent per category, per repo.
  2. Recommend. Call out which categories matter and why (a broken link is a reader's 404; a flagged secret needs rotating; an unresolvable dep breaks every install).
  3. Ask. Let the operator pick what to clean. Never assume.
  4. Fix with confirmation — plan, then apply. bun run tidy --fix <repo> prints the exact change plan (which broken links get de-linked) without writing anything; re-run with --apply to make those edits. So you always see each change first. Auto-fix is narrow on purpose: it only de-links broken/gitignored text links ([text](bad)text). Secrets, deps, images/HTML links, and file deletions are never auto-applied — they're listed as manual with guidance (rotate a key, pick the right package), because their fix is judgement, not a line-edit. Survey many with --all, then --fix one repo at a time.

Discipline

  • Report-first, always. The value is seeing the rot before touching anything.
  • Don't gate on it. /tidy is slow (network dep checks) and broad — keep it a stage, not a push-blocker. The narrow guards stay the gate.
  • Secrets are review-grade. A pattern match is a prompt to look, not a verdict — and if real, the fix is rotate-the-key, not just delete-the-line.
  • Frozen dirs are skipped (_archive/, node_modules/) — they're not yours to tidy.
  • The engine is portable. scripts/tidy.ts runs against any repo path — that's what makes the 100+-repo sweep one command.

Habilidades Relacionadas

stablyai/orca-cli

Use the public `orca` CLI to operate Orca-managed worktrees, folder contexts, terminals, repos, automations, worktree comments, and the browser embedded inside the Orca app. Use when the user says "$orca-cli", "use orca cli", "Orca worktree", "child worktree", "spawn codex/claude in a worktree", "read/wait/send Orca terminal", "terminal send", "Orca browser", or "control the browser inside Orca". Prefer this over raw `git worktree`, ad hoc PTYs, Playwright, or Computer Use when the task touches Orca-managed state. Use Computer Use for browser windows, webviews, or desktop UI outside Orca's embedded browser.

community

sdoosa123/skills-build-applications-w-copilot-agent-mode

Exercise: Build applications with GitHub Copilot agent mode

community

dliedke/ClaudeCodeExtension

Visual Studio .NET extension that provides a better interface for Claude Code, Codex, Cursor, OpenCode, PI, Windsurf and Antigravity supporting multi-line prompts with image/file attachments for AI assisted high coding productivity

community

batterfried-philosophy172/Agent-Startup-Skills

Build SaaS products using an 8-phase AI agent pipeline with mandatory approval gates for Antigravity, Claude Code, and Codex.

community

haunchen/n8n-skills

Designed specifically for AI assistants, the n8n Workflow Automation Skills Suite.

community

microsoft/hve-core

A refined collection of Hypervelocity Engineering components (instructions, prompts, agents, and skills) to start your project off right, or upgrade your existing projects to get the most out of GitHub Copilot

community