benchi/claude-dotfiles-public
Public Claude Code dotfiles: shared settings, hooks, agents, and skills
ARCHETYPE
Fort Knox
Deny lists, pre-tool hooks, sandboxing. Nothing touches prod without a signature.
GUARDRAILS
4/5
Blocks destructive commands · Pre-tool screening hook · No YOLO mode · Sandbox or ask-first rules · details
Copy this rig
# review before running: this installs third-party code $ npx degit benchi/claude-dotfiles-public/agents ./rig-claude-dotfiles-public/agents $ npx degit benchi/claude-dotfiles-public/skills ./rig-claude-dotfiles-public/skills $ npx degit benchi/claude-dotfiles-public/hooks ./rig-claude-dotfiles-public/hooks
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 benchi/claude-dotfiles-public/skills/code-feature .claude/skills/code-feature $ npx degit benchi/claude-dotfiles-public/skills/code-task .claude/skills/code-task $ npx degit benchi/claude-dotfiles-public/skills/define-feature .claude/skills/define-feature $ npx degit benchi/claude-dotfiles-public/skills/refine-feature .claude/skills/refine-feature
$ curl -fsSL --create-dirs -o .claude/agents/coder-advanced.md https://raw.githubusercontent.com/benchi/claude-dotfiles-public/main/agents/coder-advanced.md $ curl -fsSL --create-dirs -o .claude/agents/coder-basic.md https://raw.githubusercontent.com/benchi/claude-dotfiles-public/main/agents/coder-basic.md $ curl -fsSL --create-dirs -o .claude/agents/pm-architect.md https://raw.githubusercontent.com/benchi/claude-dotfiles-public/main/agents/pm-architect.md
Merge into .claude/settings.json (project) or ~/.claude/settings.json (user). Hook commands reference scripts in the source repo — copy those too.
{
"permissions": {
"ask": [
"Bash(rm -rf:*)"
]
},
"hooks": {
"PreToolUse": [
{
"matcher": "Bash|Agent",
"hooks": [
{
"type": "command",
"command": "~/.claude/hooks/audit.sh"
}
]
}
]
}
} Skills (4)
Subagents (3)
| coder-advanced model: opus | | |
| coder-basic model: sonnet | | |
| pm-architect model: opus | | |
Hooks (7)
| event | matcher | runs |
|---|---|---|
| PreCompact | * | bd prime |
| SessionStart | * | bd prime |
| SessionStart | * | ~/.claude/hooks/audit.sh |
| SessionEnd | * | ~/.claude/hooks/audit.sh |
| UserPromptSubmit | * | ~/.claude/hooks/audit.sh |
| PreToolUse | Bash|Agent | ~/.claude/hooks/audit.sh |
| SubagentStart | * | ~/.claude/hooks/audit.sh |
Permissions
deny (0)
—
ask (1)
Bash(rm -rf:*)
allow (15)
Read
Bash(npm:*)
Bash(yarn:*)
Bash(pnpm:*)
Bash(python:*)
Bash(pytest:*)
Bash(go:*)
Bash(cargo:*)
Bash(mvn:*)
Bash(gradle:*)
Bash(git:*)
Bash(ls:*)
Bash(mkdir:*)
Bash(bd:*)
Bash(beads:*)
Similar rigs
fitlab-ai/agent-infra
Collaboration infrastructure for AI coding agents | AI 编程代理的协作基础设施
Orchestrator 2.2k tok ·
ruvnet/ruflo
🌊 The original agent harness. Deploy intelligent multi-player swarms, coordinate autonomous workflows, and build conversational AI systems. Features adaptive memory, self-learning intelligence, federation, vector RAG integration, and native
Fort Knox 35.3k tok ·
Donchitos/Claude-Code-Game-Studios
Turn Claude Code into a full game dev studio — 49 AI agents, 72 workflow skills, and a complete coordination system mirroring real studio hierarchy.
Fort Knox 11.6k tok ·
FlorianBruniaux/claude-code-ultimate-guide
The most comprehensive Claude Code guide: agentic workflows, hooks, skills, MCP servers, quizzes, and production-ready templates. 430K+ lines.
Fort Knox 33.8k tok ·