mezawebdev/meza-profile-zsh
My personal shell profile. Contains config for NeoVim, Claude Code, Lazygit + more
ARCHETYPE
Orchestrator
A bench of specialised subagents. The main agent mostly delegates.
Copy this rig
# review before running: this installs third-party code
$ npx degit mezawebdev/meza-profile-zsh/agents ./rig-meza-profile-zsh/agents MCP servers are added to Claude Code at local scope; env vars are shown as YOUR_… placeholders — we never store values. Files are fetched with degit into a separate folder so you can review before merging.
$ npx degit mezawebdev/meza-profile-zsh/agents/skills/changes .claude/skills/changes $ npx degit mezawebdev/meza-profile-zsh/agents/skills/commit .claude/skills/commit $ npx degit mezawebdev/meza-profile-zsh/agents/skills/i-have-adhd .claude/skills/i-have-adhd $ npx degit mezawebdev/meza-profile-zsh/agents/skills/recap .claude/skills/recap $ npx degit mezawebdev/meza-profile-zsh/agents/skills/summary .claude/skills/summary $ npx degit mezawebdev/meza-profile-zsh/agents/skills/tldr .claude/skills/tldr
$ curl -fsSL --create-dirs -o .claude/agents/atomic-design-architect.md https://raw.githubusercontent.com/mezawebdev/meza-profile-zsh/main/agents/agents/atomic-design-architect.md $ curl -fsSL --create-dirs -o .claude/agents/e2e-testing-expert.md https://raw.githubusercontent.com/mezawebdev/meza-profile-zsh/main/agents/agents/e2e-testing-expert.md $ curl -fsSL --create-dirs -o .claude/agents/spec-maintainer.md https://raw.githubusercontent.com/mezawebdev/meza-profile-zsh/main/agents/agents/spec-maintainer.md $ curl -fsSL --create-dirs -o .claude/agents/testing-expert.md https://raw.githubusercontent.com/mezawebdev/meza-profile-zsh/main/agents/agents/testing-expert.md $ curl -fsSL --create-dirs -o .claude/agents/typescript-expert.md https://raw.githubusercontent.com/mezawebdev/meza-profile-zsh/main/agents/agents/typescript-expert.md $ curl -fsSL --create-dirs -o .claude/agents/worker.md https://raw.githubusercontent.com/mezawebdev/meza-profile-zsh/main/agents/agents/worker.md $ curl -fsSL --create-dirs -o .claude/agents/cmux-dispatch.md https://raw.githubusercontent.com/mezawebdev/meza-profile-zsh/main/agents/commands/cmux-dispatch.md $ curl -fsSL --create-dirs -o .claude/agents/i-have-adhd.md https://raw.githubusercontent.com/mezawebdev/meza-profile-zsh/main/agents/commands/i-have-adhd.md $ curl -fsSL --create-dirs -o .claude/agents/parallel-orchestrator.md https://raw.githubusercontent.com/mezawebdev/meza-profile-zsh/main/agents/commands/parallel-orchestrator.md $ curl -fsSL --create-dirs -o .claude/agents/parallel-worker.md https://raw.githubusercontent.com/mezawebdev/meza-profile-zsh/main/agents/commands/parallel-worker.md $ curl -fsSL --create-dirs -o .claude/agents/what.md https://raw.githubusercontent.com/mezawebdev/meza-profile-zsh/main/agents/commands/what.md $ curl -fsSL --create-dirs -o .claude/agents/SKILL.md https://raw.githubusercontent.com/mezawebdev/meza-profile-zsh/main/agents/skills/changes/SKILL.md $ curl -fsSL --create-dirs -o .claude/agents/SKILL.md https://raw.githubusercontent.com/mezawebdev/meza-profile-zsh/main/agents/skills/commit/SKILL.md $ curl -fsSL --create-dirs -o .claude/agents/SKILL.md https://raw.githubusercontent.com/mezawebdev/meza-profile-zsh/main/agents/skills/i-have-adhd/SKILL.md $ curl -fsSL --create-dirs -o .claude/agents/SKILL.md https://raw.githubusercontent.com/mezawebdev/meza-profile-zsh/main/agents/skills/recap/SKILL.md $ curl -fsSL --create-dirs -o .claude/agents/SKILL.md https://raw.githubusercontent.com/mezawebdev/meza-profile-zsh/main/agents/skills/summary/SKILL.md $ curl -fsSL --create-dirs -o .claude/agents/SKILL.md https://raw.githubusercontent.com/mezawebdev/meza-profile-zsh/main/agents/skills/tldr/SKILL.md
This rig commits no guardrails. Here is the community baseline instead — the deny/ask rules most often found across all 7,204 rigs:
{
"permissions": {
"deny": [
"Read(./.env)",
"Read(**/.env)",
"Read(~/.ssh/**)",
"Bash(rm -rf *)",
"Read(**/*.pem)",
"Bash(rm -rf /)",
"Bash(git push --force:*)",
"Bash(sudo *)",
"Read(.env)",
"Bash(rm -rf /*)",
"Read(./.env.*)",
"Read(~/.aws/**)",
"Bash(git push --force*)",
"Bash(rm -rf:*)",
"Read(**/*.key)",
"Read(**/.env.*)",
"Bash(sudo:*)",
"Bash(git reset --hard*)",
"Bash(git reset --hard:*)",
"Read(.env.*)"
],
"ask": [
"Bash(git push:*)",
"Bash(git push *)",
"Bash(git commit:*)",
"Bash(rm *)",
"Bash(rm:*)",
"Bash(git rebase *)",
"Bash(wget *)",
"Bash(npm publish:*)",
"Bash(git commit *)",
"Bash(gh pr merge *)"
]
}
} Skills (6)
Subagents (17)
| atomic-design-architect | Use this agent when you need to decompose UI requirements into atomic design components, plan new component hierarchies, refactor existing UI into reusable primitives, or create blueprints for scalabl |
| e2e-testing-expert | Use this agent when you need to create, execute, observe, or validate end-to-end test suites using Playwright MCP. This agent specializes in working with 'test plans' - markdown-based test specificati |
| spec-maintainer | A documentation agent for the "After Code" phase in a polyglot monorepo. It reverse-engineers code into functional specs, strictly separating Frontend (UI/Experience) from Backend (Data/Contracts). |
| testing-expert | Use this agent when you need to write, refactor, or improve tests for a codebase. This includes writing unit tests, integration tests, and end-to-end (e2e) tests, as well as providing guidance on test |
| typescript-expert | Use this agent when you need to design, create, or refactor TypeScript types, interfaces, generics, or type utilities. This includes tasks like: defining data models, creating reusable generic types, |
| worker | Use this agent when you need to delegate a specific, well-defined task to an individual contributor that will execute it with precision and focus. This agent is ideal for tasks that have clear instruc |
| cmux-dispatch | — |
| i-have-adhd | Toggle ADHD output mode on (stays on until you say "stop adhd mode") |
| parallel-orchestrator | Use when you need to parallelize work across multiple Claude instances. Analyzes task lists, identifies independent work streams, creates task assignments, and monitors worker progress. Invoke with: / |
| parallel-worker | Use when working as a parallel worker agent. Reads task assignment, creates worktree/branch, updates status as you work. Human-in-the-loop: all commits require approval, never push. Invoke with: /para |
| what | Quick steer back into the conversation after getting distracted. Re-summarizes what we were just doing in one or two concise sentences. Invoke with: /what |
| changes | List all files modified by this session |
| commit | Commit all current changes with an appropriate commit message |
| i-have-adhd | Shape output for a reader with ADHD: lead with the next action, number multi-step work, restate state across turns, suppress tangents, give specific time estimates, make wins visible. Invoke with /i-h |
| recap | Summarize what this session has been working on so the user can get back up to speed |
| summary | List all files modified by this session with a detailed description of changes in each |
| tldr | Compress your most recent response into a TL;DR plus numbered key points. Use when the user invokes /tldr to condense a wall of text. Accepts an optional argument — a number caps bullet count, or "hea |
Similar rigs
maxclax/dotfiles
Cross-platform dotfiles managed with chezmoi, Nix Home Manager, and age encryption. Includes GTD-style Org-mode workflow, Doom Emacs, tmux, and automated backup with Borgmatic.
Pragmatist 2.4k tok ·
TechHU-GS/cc-discipline
Guardrails for Claude Code. Shell hooks that actually block bad behavior — edit loops, skipped verification, destructive git — not just markdown that asks nicely.
Automator 7.6k tok ·
kazukit-dev/dotfiles
kazukit's dotfiles
Orchestrator 3.9k tok ·
mariano-aguero/claude-code-config
My Claude Code configuration · hooks, skills, settings
Orchestrator 1.7k tok ·