House Rules — Security skill for Claude Code
House Rules for AI coding agents: nine short principles covering voice, ambition, truth, cost, delegation, efficiency, craft and security.
How to install House Rules
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open pierresauvignon/house-rules and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What House Rules does
House Rules for AI coding agents: nine short principles covering voice, ambition, truth, cost, delegation, efficiency, craft and security. Drop it into ~/.claude/CLAUDE.md, or any agent's system prompt, so every session works plainly, cheaply, honestly and safely.
Alternatives in Security
- Brand Strategy — Define or audit a brand — positioning, visual identity system, voice and tone, and brand architecture 309 ★
- Reflexion — Audit and optimize the .claude/ ecosystem — skills, hooks, guides, patterns, agents, and settings — for qualit 224 ★
- AI First Toolkit — Open-source Claude Code skills and Codex skills for AI-first work 95 ★
README
House Rules
.-------------. .-------------.
/ \ / \
/ \ / \
| | | |
| I VOICE | | VI EFFICIENCY |
| II AMBITION | | VII CRAFT |
| III TRUTH | | VIII SECURITY |
| IV SIMPLICITY | | IX SAFE BUILDS |
| V DELEGATION | | |
| | | |
|___________________| |___________________|
*Thou shalt not say "delve".*
Nine short principles for AI coding agents. They cover how an agent sounds, how it thinks, how it works and how it keeps your systems safe.
[Read the rules →](HOUSE-RULES.md)
Why
Out of the box, agents tend to sound like agents, reach for paid services by default, scale down hard requests without saying so, report work as done before checking it, and leave security for later. These rules push the other way, in about 2,000 tokens.
| Group | Rules |
|---|---|
| How you come across | I. Voice and Appearance |
| How you think | II. Ambition, III. Truth |
| How you work | IV. Cost and Simplicity, V. Delegation, VI. Efficiency and Friction, VII. Craft and Scope |
| How you protect | VIII. Security and Safety, IX. Securing What You Build |
When two rules conflict, safety, truth and consent come first. After that, your aim decides the goal and simplicity decides how to reach it.
Use
**Claude Code, every project:**
curl -fsSL https://raw.githubusercontent.com/pierresauvignon/house-rules/main/HOUSE-RULES.md >> ~/.claude/CLAUDE.md
This appends to any CLAUDE.md you already have.
**One project only:** add the contents to `CLAUDE.md` at the project root.
**Other agents:** paste it into the system prompt or rules file: `AGENTS.md`, `.cursorrules`, Copilot instructions, and so on. Nothing in it is specific to one tool.
Make it yours
These are a starting point. Cut rules that don't fit, add your own, and change the ba
Related Skills
Optimizing Skills Performance
Audit and optimize AI agent skills for token efficiency using progressive disclosure principles
Optimize Harness
Audit one or all supported agent harnesses for config efficiency, hook overhead, permission friction, plugin a
AI Spend
Audit Claude Code token efficiency: MCP overhead, session patterns, skill cost, cache hit rate. Add --full for
Audit Doctrine
Audit the rules/ directory for missing foundational principles; output gap list with priority and justificatio
Delegate And Audit
Explicit multi-model delegation and independent project audit skill for Claude Code on Windows
Design Hub
One entry point for all your design skills. design-hub is a small router skill for Claude: you type a short sh
Related Agents
Review Router
Cheap first-pass PR reviewer for the review-loop system. Reviews the whole diff, grades blast radius, and retu
Claude Code Subagents By 0xfurai
Production-ready collection of 100+ specialized subagents covering programming languages, frameworks, database
Cb Truth
CoalBoard's formal lens (sub2) — reasons from first principles, invariants, and proof; checks internal consist