terezakohutova/claude-code-architecture-showcase
A real Claude Code workspace with the business taken out: CLAUDE.md index, one rule per file, hooks that block, skills and scoped subagents. Tested in CI on Linux, macOS and Windows.
ARCHETYPE
Fort Knox
Deny lists, pre-tool hooks, sandboxing. Nothing touches prod without a signature.
GUARDRAILS
4/5
Blocks destructive commands · Protects secrets · Pre-tool screening hook · No YOLO mode · details
Quality 54/100 Guardrails v2
Copy this rig
# review before running: this installs third-party code
$ npx degit terezakohutova/claude-code-architecture-showcase/.claude ./rig-claude-code-architecture-showcase # 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.
$ npx degit terezakohutova/claude-code-architecture-showcase/.claude/skills/commitment-tracker .claude/skills/commitment-tracker $ npx degit terezakohutova/claude-code-architecture-showcase/.claude/skills/product-listing .claude/skills/product-listing
$ curl -fsSL --create-dirs -o .claude/agents/brand-voice-reviewer.md https://raw.githubusercontent.com/terezakohutova/claude-code-architecture-showcase/master/.claude/agents/brand-voice-reviewer.md $ curl -fsSL --create-dirs -o .claude/agents/inventory-researcher.md https://raw.githubusercontent.com/terezakohutova/claude-code-architecture-showcase/master/.claude/agents/inventory-researcher.md $ curl -fsSL --create-dirs -o .claude/agents/procedure-tuner.md https://raw.githubusercontent.com/terezakohutova/claude-code-architecture-showcase/master/.claude/agents/procedure-tuner.md
Merge into .claude/settings.json (project) or ~/.claude/settings.json (user). Hook commands reference scripts in the source repo — copy those too.
{
"permissions": {
"deny": [
"Bash(curl:*)",
"Bash(wget:*)",
"Bash(ssh:*)",
"Bash(scp:*)",
"Bash(rsync:*)",
"Bash(nc:*)",
"Bash(git push:*)",
"Bash(rm -rf:*)",
"Bash(rm -fr:*)",
"Bash(sudo:*)",
"Read(./.env)",
"Read(./.env.*)",
"Read(~/.ssh/**)",
"Read(~/.aws/**)",
"Edit(./.claude/settings.json)",
"Edit(./.claude/settings.local.json)",
"Edit(./.claude/hooks/**)",
"Write(./.claude/settings.json)",
"Write(./.claude/settings.local.json)",
"Write(./.claude/hooks/**)",
"MultiEdit(./.claude/settings.json)",
"MultiEdit(./.claude/settings.local.json)",
"MultiEdit(./.claude/hooks/**)"
]
},
"hooks": {
"PreToolUse": [
{
"matcher": "Read|Edit|Write|MultiEdit|NotebookEdit|Glob|Grep",
"hooks": [
{
"type": "command",
"command": "node \"$CLAUDE_PROJECT_DIR/.claude/hooks/scope-guard.js\""
}
]
},
{
"matcher": "Edit|Write|MultiEdit|NotebookEdit",
"hooks": [
{
"type": "command",
"command": "node \"$CLAUDE_PROJECT_DIR/.claude/hooks/style-guard.js\""
},
{
"type": "command",
"command": "node \"$CLAUDE_PROJECT_DIR/.claude/hooks/version-snapshot.js\""
}
]
},
{
"matcher": "Bash",
"hooks": [
{
"type": "command",
"command": "node \"$CLAUDE_PROJECT_DIR/.claude/hooks/bash-guard.js\""
}
]
}
]
}
} Skills (2)
Subagents (3)
| brand-voice-reviewer model: sonnet | Checks a draft (listing, buyer reply, social post) against the shop's published voice in Brand/voice-examples.md and runs the same mechanical style check as the style-guard hook. Persona: Karel Jaromí |
| inventory-researcher model: sonnet | Sourced research on an item's maker, material, era, or provenance before it goes into a listing - fans out web/reference lookups and returns a clean writeup instead of filling the main conversation wi |
| procedure-tuner model: sonnet | Reviews .claude/skills/*/SKILL.md and .claude/agents/*.md definitions for dead tool grants, stale file references, unclear triggers, and structural gaps, and returns proposed fixes as diffs for the us |
Hooks (5)
| event | matcher | runs |
|---|---|---|
| PreToolUse | Read|Edit|Write|MultiEdit|NotebookEdit|Glob|Grep | node "$CLAUDE_PROJECT_DIR/.claude/hooks/scope-guard.js" |
| PreToolUse | Edit|Write|MultiEdit|NotebookEdit | node "$CLAUDE_PROJECT_DIR/.claude/hooks/style-guard.js" |
| PreToolUse | Edit|Write|MultiEdit|NotebookEdit | node "$CLAUDE_PROJECT_DIR/.claude/hooks/version-snapshot.js" |
| PreToolUse | Bash | node "$CLAUDE_PROJECT_DIR/.claude/hooks/bash-guard.js" |
| SessionStart | startup | node "$CLAUDE_PROJECT_DIR/.claude/hooks/tuning-ledger-check.js" |
Permissions
deny (23)
Bash(curl:*)
Bash(wget:*)
Bash(ssh:*)
Bash(scp:*)
Bash(rsync:*)
Bash(nc:*)
Bash(git push:*)
Bash(rm -rf:*)
Bash(rm -fr:*)
Bash(sudo:*)
Read(./.env)
Read(./.env.*)
Read(~/.ssh/**)
Read(~/.aws/**)
Edit(./.claude/settings.json)
Edit(./.claude/settings.local.json)
Edit(./.claude/hooks/**)
Write(./.claude/settings.json)
Write(./.claude/settings.local.json)
Write(./.claude/hooks/**)
MultiEdit(./.claude/settings.json)
MultiEdit(./.claude/settings.local.json)
MultiEdit(./.claude/hooks/**)
ask (0)
—
allow (7)
Bash(node scripts/style-check.js:*)
Bash(node scripts/check-docs.js)
Bash(node --test tests/hooks.test.js)
Bash(npm test)
Bash(git status:*)
Bash(git diff:*)
Bash(git log:*)
Similar rigs
migavel508/wisprnote-skills
Agent skills that give Claude Code, Codex, Cursor and Copilot access to meeting context: trace code back to the decision that produced it, catch up after time away, audit action items that were agreed and never tracked, and write ADRs. Plai
Pragmatist 402 tok · · Q0
sansernonline/SQT-Marketplace
Pre-Define AI Agents
Skill Collector 4.7k tok · · Q0
ruvnet/RuView
π RuView turns commodity WiFi signals into real-time spatial intelligence, vital sign monitoring, and presence detection — all without a single pixel of video.
Fort Knox 14.1k tok · · Q20
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 · · Q49