Rules

safety-guard helps prevent destructive operations when agents work autonomously or on production systems. It adds careful mode, write freeze mode, and guard mode to block risky commands, confine edits to one directory, and reduce mistakes during deploys, migrations, and sensitive repo work.

rules-distill is a maintenance skill for Skill Authors and prompt library curators. It scans installed skills, distills repeated patterns into reusable rules, and helps you append, revise, or create rule files with less guesswork than a generic review prompt.

hookify-rules helps you create Hookify rule files with YAML frontmatter for .claude/hookify.{rule-name}.local.md. Use it to define the right event, pattern, and action for warnings or blocks, and to turn a rough policy into a reliable hookify-rules guide for Skill Authoring.

The configure-ecc skill is an interactive installer for Everything Claude Code (ECC). It guides selective installation of skills and rules, helps choose user-level or project-level destinations, verifies target paths, and can optimize installed files after setup. Ideal for a reliable configure-ecc guide.

The coding-standards skill gives a baseline for naming, readability, immutability, consistency, and code review across projects, before applying framework-specific rules.

verification-before-completion is a final-check skill that blocks unsupported completion claims. Learn when to use it, how to install it from obra/superpowers, and how to match each status claim to fresh verification evidence.

git-guardrails-claude-code adds a PreToolUse hook to block dangerous git commands before Claude Code runs them. Install it to prevent destructive pushes, hard resets, forced cleans, and branch deletes, with scoped control for this project or all projects. Useful when you need git-guardrails-claude-code for Access Control boundaries in Claude Code.

The context-engineering skill helps you structure project context so agents follow conventions, reduce hallucinations, and stay focused. Use it when starting a session, switching tasks, or building a context-engineering guide for a codebase.

figma-create-design-system-rules helps you generate project-specific Design System rules for Figma-to-code work. Use it to capture conventions for components, naming, tokens, file locations, and what not to hardcode so AI coding agents stay consistent across your repo. Requires a Figma MCP server connection.

git-guardrails-claude-code adds a Claude Code PreToolUse hook that blocks dangerous Git commands like push, reset --hard, clean -f, branch -D, checkout ., and restore . before execution. It supports project or global install, manual hook setup, executable script permissions, and behavior testing with the bundled shell script.

The zhangxuefeng-perspective skill turns education, career, and life-choice questions into blunt, practical guidance in Zhang Xuefeng’s public thinking style. Use it when you need a decision lens for majors, school choice, job prospects, family constraints, or考研. It’s best for questions that need a direct recommendation, not a neutral explanation.

semgrep-rule-variant-creator helps port existing Semgrep rules into target languages with applicability analysis, test-first validation, and separate rule/test outputs. Use the semgrep-rule-variant-creator skill when you need a reliable guide for Semgrep rule expansion across polyglot codebases, not a brand-new rule from scratch.

semgrep-rule-creator creates production-quality Semgrep rules for security vulnerabilities, bug patterns, taint-flow detections, and coding standards. Use the semgrep-rule-creator skill for Security Audit work when you need precise rules, test cases, and validation instead of a generic draft.

agent-md-refactor helps refactor bloated AGENTS.md, CLAUDE.md, or COPILOT.md files using progressive disclosure. It surfaces contradictions, keeps universal rules at the root, and splits the rest into linked docs for cleaner Context Engineering and lower context load.

tinybird-cli-guidelines is a practical guide for Tinybird CLI commands, workflows, and operations. It helps backend development teams and agents choose the right tb command, manage local development, deploy safely, and handle data, tokens, and secrets with less guesswork.

moyu-strict is a strict anti-over-engineering skill for Code Editing. It helps agents keep changes narrow, confirm scope before edits, avoid new abstractions, and skip unrequested tests, docs, dependencies, or rewrites. Use it when you want the smallest safe diff and predictable review.

moyu-lite is a lightweight guardrail skill for code editing that keeps changes tightly scoped to the user’s exact request. It avoids extra refactors, new abstractions, broad rewrites, and side changes, making moyu-lite useful when you need minimal diffs, careful scope control, and a clear moyu-lite guide for restrained editing.

moyu-ja is a Japanese code editing skill for keeping changes tightly scoped. It favors the smallest safe diff, avoids unrelated edits, and asks for clarification when a request is ambiguous. Use moyu-ja when you need disciplined, minimal updates instead of broad refactors. The moyu-ja guide focuses on restraint, exact scope, and practical change control.

moyu-fr is a code-editing guardrail skill for small, surgical changes. It helps agents avoid overengineering, keep diffs minimal, and focus on the exact edit requested. Use moyu-fr when you want a simple patch, not a rewrite, and when tight scope control matters more than expansion.

moyu is a Code Editing skill that keeps changes tightly scoped, avoids over-engineering, and prefers the smallest safe diff. It helps agents stay within the user’s request, making moyu useful for precise fixes, narrow edits, and restrained workflows.

moyu-en is a restraint-first skill for code editing. It helps keep changes surgical, avoid over-engineering, and limit diffs to exactly what the user asked for. Use moyu-en when you need precise, small patches, narrow scope control, and a better guide for minimal code updates.

reflect is a Skill Validation tool for reviewing a prior response or output. It uses complexity triage and verification to catch missed flaws, weak reasoning, and overconfident approval before work ships.

full-output-enforcement is a writing and execution guardrail skill for complete, unabridged output. It bans placeholders, partial deliveries, and skipped middle sections, making it useful for code generation, documentation, migrations, and Skill Authoring when completeness matters.

freeze is a guardrail skill that restricts file edits to one directory for the session. It blocks Edit and Write outside the allowed path, making it useful for workflow automation, debugging, and scoped refactors when you want the agent to stay inside a module or folder.