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 pin15/dotfile/agents ./rig-dotfile/agents $ npx degit pin15/dotfile/skills ./rig-dotfile/skills $ npx degit pin15/dotfile/hooks ./rig-dotfile/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 pin15/dotfile/skills/code-study .claude/skills/code-study $ npx degit pin15/dotfile/skills/prompt .claude/skills/prompt $ npx degit pin15/dotfile/skills/think .claude/skills/think $ npx degit pin15/dotfile/skills/work .claude/skills/work
$ curl -fsSL --create-dirs -o .claude/agents/icd-analyse.md https://raw.githubusercontent.com/pin15/dotfile/main/agents/icd-analyse.md $ curl -fsSL --create-dirs -o .claude/agents/icd-audit.md https://raw.githubusercontent.com/pin15/dotfile/main/agents/icd-audit.md $ curl -fsSL --create-dirs -o .claude/agents/icd-deepdive.md https://raw.githubusercontent.com/pin15/dotfile/main/agents/icd-deepdive.md $ curl -fsSL --create-dirs -o .claude/agents/icd-research.md https://raw.githubusercontent.com/pin15/dotfile/main/agents/icd-research.md $ curl -fsSL --create-dirs -o .claude/agents/verifier.md https://raw.githubusercontent.com/pin15/dotfile/main/agents/verifier.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(git push *)",
"Bash(git reset --hard *)",
"Bash(rm -rf *)"
]
},
"hooks": {
"PreToolUse": [
{
"matcher": "*",
"hooks": [
{
"type": "command",
"command": "__HOME__/.claude/hooks/skill-gate.sh"
}
]
},
{
"matcher": "Read|Grep|Glob|Bash|WebSearch|WebFetch|Write",
"hooks": [
{
"type": "command",
"command": "__HOME__/.claude/hooks/subagent-budget-tracker.sh"
}
]
}
]
}
} Skills (4)
Subagents (5)
| icd-analyse model: opus | ICD investigation agent for general-purpose analytical reasoning. Default for problems that need thinking. |
| icd-audit model: opus | ICD investigation agent for systematic rubric-driven evaluation. Full 6-step loop. |
| icd-deepdive model: opus | ICD investigation agent for deep codebase investigation. Full 6-step loop. |
| icd-research model: opus | ICD investigation agent for external research. Full 6-step loop with source credibility tracking. |
| verifier model: sonnet | > |
Hooks (11)
| event | matcher | runs |
|---|---|---|
| SessionStart | * | __HOME__/.claude/hooks/session-protocol.sh |
| SessionEnd | * | __HOME__/.claude/hooks/session-end-capture.sh |
| PreCompact | * | __HOME__/.claude/hooks/pre-compact-capture.sh |
| Stop | * | You are a verification checker. Given: $ARGUMENTS RULE 1: If stop_hook_active is true in the input, respond {"ok": true} immediately. RULE 2: Check if code changes were made (Edit, Write, or file modification mentions in the conversation) |
| PostToolUseFailure | Bash | __HOME__/.claude/hooks/iteration-guard.sh |
| UserPromptSubmit | * | __HOME__/.claude/hooks/skill-protocol.sh |
| PreToolUse | * | __HOME__/.claude/hooks/skill-gate.sh |
| PreToolUse | Read|Grep|Glob|Bash|WebSearch|WebFetch|Write | __HOME__/.claude/hooks/subagent-budget-tracker.sh |
| PostToolUse | Edit | __HOME__/.claude/hooks/claudemd-sync-check.sh |
| PostToolUse | Write | __HOME__/.claude/hooks/claudemd-sync-check.sh |
| PostToolUse | Agent | __HOME__/.claude/hooks/icd-agent-check.sh |
Plugins (5)
code-review@claude-plugins-officialcode-simplifier@claude-plugins-officialralph-loop@claude-plugins-officialtypescript-lsp@claude-plugins-officiallearning-output-style@claude-plugins-official
Permissions
deny (0)
—
ask (3)
Bash(git push *)
Bash(git reset --hard *)
Bash(rm -rf *)
allow (3)
Bash(.claude/scripts/*)
Bash(claude mcp:*)
Skill
Similar rigs
cjhowe-us/workflow-plugin
Claude Code plugin: workflow, coding standards, templates, and agents
Pragmatist 254 tok ·
Remus3/Legion-Wallpaper
Self-auditing AI image restoration pipeline (super-resolution, LaMa inpainting, face repair, metric-plus-vision gate ladder), built by a multi-agent Claude Code system with independent verifiers, git-hook gates and headless background lanes
Automator 5.5k tok ·
leavemagic-cyber/earned-confidence
A behavior contract for AI coding agents — every rule added after a real failure. Ships with a verifier agent that cannot edit what it judges.
Pragmatist 3.6k tok ·
vitaliikapliuk/modelharness
Make every model cheaper or better. Zero-config behavioral harness for Claude Code, with a reproducible 408-run benchmark
Pragmatist 236 tok ·