itshig/AGENTS-template
Drop-in AGENTS.md template for Claude Code, Cursor, Codex, and other AI agents — with .agents/ for rules, personas, and workflows.
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 itshig/AGENTS-template/.claude ./rig-agents-template # inspect, then merge into .claude/ 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.
$ curl -fsSL --create-dirs -o .claude/agents/architect.md https://raw.githubusercontent.com/itshig/AGENTS-template/main/.claude/agents/architect.md $ curl -fsSL --create-dirs -o .claude/agents/debugger.md https://raw.githubusercontent.com/itshig/AGENTS-template/main/.claude/agents/debugger.md $ curl -fsSL --create-dirs -o .claude/agents/design-reviewer.md https://raw.githubusercontent.com/itshig/AGENTS-template/main/.claude/agents/design-reviewer.md $ curl -fsSL --create-dirs -o .claude/agents/reviewer.md https://raw.githubusercontent.com/itshig/AGENTS-template/main/.claude/agents/reviewer.md $ curl -fsSL --create-dirs -o .claude/agents/verifier.md https://raw.githubusercontent.com/itshig/AGENTS-template/main/.claude/agents/verifier.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 *)"
]
}
} Subagents (5)
| architect model: opus | Evaluates structural decisions before implementation — new modules, new dependencies, data model or API contract changes, or picking between hard-to-reverse approaches. Produces a written recommendati |
| debugger model: opus | Systematic root-cause investigation for a bug, test failure, or unexpected behavior. Use when stuck on a non-obvious failure, when a fix attempt has already failed once, or when a bug needs diagnosing |
| design-reviewer model: opus | Audits UI changes against the project's design system — tokens, semantic color, component states, responsive behavior, and the accessibility floor. Use after building or changing anything that renders |
| reviewer model: opus | Adversarial pre-commit code review of the current diff. Use after a feature is functionally complete and before committing. Reads the diff and the changed files end-to-end, hunting for correctness bug |
| verifier model: haiku | Runs the repo's full verification gauntlet — build, lint, typecheck, tests — and reports a pass/fail checklist. Use before declaring any task complete, before committing, or whenever asked to "verify" |
Similar rigs
joaolozano-lendario/athena-os
A Claude Code Native System for Intelligent Workflow Architecture - Slash Commands for Smarter AI Workflows
Orchestrator 3.7k tok ·
SchotjeChrisman/agentic-workflow
A self-improving Claude Code setup
Orchestrator 2.8k tok ·
Trolzie/claude-loop-systems-demo
A small demo of Markdown-as-system AI dev loops with Claude Code: issues → agents → skills → verification.
Pragmatist 480 tok ·
homeofe/improvements
Portable, LLM-agnostic framework for multi-model AI workflows: model routing, structured agent handoff (AAHP), and a phased agent pipeline. MIT.
Orchestrator 1.4k tok ·