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 caarlos0/dotfiles/agents ./rig-dotfiles/agents $ npx degit caarlos0/dotfiles/skills ./rig-dotfiles/skills
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 caarlos0/dotfiles/skills/change-impact-auditor .claude/skills/change-impact-auditor $ npx degit caarlos0/dotfiles/skills/cli-design .claude/skills/cli-design $ npx degit caarlos0/dotfiles/skills/code-review .claude/skills/code-review $ npx degit caarlos0/dotfiles/skills/code-simplifier .claude/skills/code-simplifier $ npx degit caarlos0/dotfiles/skills/dashboard .claude/skills/dashboard $ npx degit caarlos0/dotfiles/skills/dependabot-merge .claude/skills/dependabot-merge $ npx degit caarlos0/dotfiles/skills/gh-cli .claude/skills/gh-cli $ npx degit caarlos0/dotfiles/skills/gh-doc-author .claude/skills/gh-doc-author $ npx degit caarlos0/dotfiles/skills/go-conventions .claude/skills/go-conventions $ npx degit caarlos0/dotfiles/skills/go-doc .claude/skills/go-doc $ npx degit caarlos0/dotfiles/skills/go-performance .claude/skills/go-performance $ npx degit caarlos0/dotfiles/skills/issue-validator .claude/skills/issue-validator $ npx degit caarlos0/dotfiles/skills/maintainer-briefing .claude/skills/maintainer-briefing $ npx degit caarlos0/dotfiles/skills/python-performance .claude/skills/python-performance $ npx degit caarlos0/dotfiles/skills/runtime-process-debugging .claude/skills/runtime-process-debugging $ npx degit caarlos0/dotfiles/skills/rust-performance .claude/skills/rust-performance $ npx degit caarlos0/dotfiles/skills/rust-specialist .claude/skills/rust-specialist $ npx degit caarlos0/dotfiles/skills/tui-design .claude/skills/tui-design $ npx degit caarlos0/dotfiles/skills/typescript-performance .claude/skills/typescript-performance $ npx degit caarlos0/dotfiles/skills/writing-tests .claude/skills/writing-tests
$ curl -fsSL --create-dirs -o .claude/agents/anvil.agent.md https://raw.githubusercontent.com/caarlos0/dotfiles/main/agents/anvil.agent.md $ curl -fsSL --create-dirs -o .claude/agents/caarlos0.agent.md https://raw.githubusercontent.com/caarlos0/dotfiles/main/agents/caarlos0.agent.md $ curl -fsSL --create-dirs -o .claude/agents/electric-engineer.agent.md https://raw.githubusercontent.com/caarlos0/dotfiles/main/agents/electric-engineer.agent.md $ curl -fsSL --create-dirs -o .claude/agents/issue-auditor.agent.md https://raw.githubusercontent.com/caarlos0/dotfiles/main/agents/issue-auditor.agent.md $ curl -fsSL --create-dirs -o .claude/agents/personal-trainer.agent.md https://raw.githubusercontent.com/caarlos0/dotfiles/main/agents/personal-trainer.agent.md $ curl -fsSL --create-dirs -o .claude/agents/security-engineer.agent.md https://raw.githubusercontent.com/caarlos0/dotfiles/main/agents/security-engineer.agent.md $ curl -fsSL --create-dirs -o .claude/agents/security-researcher.agent.md https://raw.githubusercontent.com/caarlos0/dotfiles/main/agents/security-researcher.agent.md $ curl -fsSL --create-dirs -o .claude/agents/vuln-hunter.agent.md https://raw.githubusercontent.com/caarlos0/dotfiles/main/agents/vuln-hunter.agent.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 (20)
change-impact-auditorcli-designcode-reviewcode-simplifierdashboarddependabot-mergegh-cligh-doc-authorgo-conventionsgo-docgo-performanceissue-validatormaintainer-briefingpython-performanceruntime-process-debuggingrust-performancerust-specialisttui-designtypescript-performancewriting-tests
Subagents (8)
| anvil | Evidence-first coding and verification agent. Reproduces claims, makes proportional changes, attacks its own conclusions, and reports only tool-backed evidence. |
| caarlos0 | Engineering and maintainer partner in Carlos Becker's style. Challenges scope, favors boring reliability, and makes direct decisions about APIs, CLIs, reviews, releases, and maintenance. |
| electric-engineer | Analyze household electricity use, solar generation, battery backup, and electricity-bill economics with a Brazil-first focus. Use when estimating consumption, sizing solar or batteries, comparing bac |
| issue-auditor | Read-only issue investigator. Verifies reports against current code, finds related pull requests, refreshes stale references, and returns evidence-backed validity verdicts. |
| personal-trainer | Carlos's personal trainer. Knows his lifting history, preferred split, favorite exercises, working weights, and training style from his Hevy log. Use whenever Carlos asks about workouts, programming, |
| security-engineer | Secure design and hardening specialist. Threat models features, fixes vulnerabilities at the root cause, removes whole bug classes with safe defaults and APIs, and hardens builds, CI, and supply chain |
| security-researcher | Vulnerability triage and disclosure specialist. Verifies incoming security reports and CVEs against current code, measures real exposure, prioritizes by exploitation evidence, and drives coordinated d |
| vuln-hunter | Read-only vulnerability finder. Audits code like an attacker — maps attack surface, traces untrusted input to dangerous sinks, runs variant analysis — and reports only reachable, reproduced findings. |
Similar rigs
wondping0/wondev
Write your AI agent knowledge once. Compile it to every agent config format - CLAUDE.md, AGENTS.md, .cursor/rules and more.
Pragmatist 229 tok ·
meowgorithm/dotfiles
The Meowgorithm Dotfiles: Collectors’ Edition
Skill Collector 1.2k tok ·
caiaffa/claude-code-ultimate-engineering-system
A complete engineering toolkit for Claude Code: premium skills, agent workflows, ADR/PRD templates, incident learning, and production-grade governance.
Orchestrator 2.8k tok ·
getsentry/skills
Agent Skills used by the Sentry team for development.
Skill Collector 2.6k tok ·