Guardrails

Rendered guardrail framework: keep a repo's hard rules in one-rule-per-file GRL sources behind a compact router, and render only the rules a task needs. Use when gathering guardrail context before implementing or reviewing code, when creating, editing, retiring, or reviewing GRL rule files or .agents/guardrails/index.md, when running the guardrails CLI (validate, render, review-metadata, next-id), or when setting up, adopting, or bootstrapping a guardrail framework in a repo that does not have one yet. Do NOT use for ordinary lint rule authoring unless the lint rule backs a guardrail, for runt

Install

skills CLI (Claude Code, Cursor, Codex and more)
npx skills add bolasblack/claude-skills --skill guardrails
Manual install
git clone --depth 1 https://github.com/bolasblack/claude-skills.git /tmp/claude-skills
mkdir -p ~/.claude/skills
cp -r /tmp/claude-skills/skills/guardrails ~/.claude/skills/guardrails

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# Guardrails

A repository accumulates hard rules that an agent must know before writing code. When those rules live in grouped topic documents, routing an agent to a group loads every rule's text, examples, and rationale at once, and burns a large share of the context window before any task code is read.

This framework keeps the default hot-path cost to one compact router file plus the one-line summaries of the IDs a task actually selected. Rule bodies, examples, and rationale stay on disk and are rendered on demand.

`<SKILL_PATH>` means the loaded base directory of this skill. Use it in skill documentation; use concrete installed paths only in repo-level executable files such as package scripts, Makefile targets, git hooks, or CI config.

## What the Framework Owns

- `.agents/guardrails/index.md` — compact routing document that lists GRL IDs per task or scenario. Read first, read whole.
- `.agents/guardrails/rules/GRL-<number>.md` — active one-rule-per-file sources. Never opened directly for ordinary implementation context.
- `.agents/guardrails/retired-rules/GRL-<number>.md` — retired sources, kept for ID history.
- `<SKILL_PATH>/scripts/guardrails.ts` — Bun CLI for validation, rendering, review metadata, and next-id lookup.

These paths and the `GRL-` ID prefix are fixed by the CLI and are not configurable.

## Requirements

- Bun on PATH (https://bun.sh).
- git, for `review-metadata`.
Read the full skill on GitHub

Details

Skill name
guardrails
Path
skills/guardrails/SKILL.md
Repo updated
18 November 2026

FAQ

What does the Guardrails skill do?

Rendered guardrail framework: keep a repo's hard rules in one-rule-per-file GRL sources behind a compact router, and render only the rules a task needs. Use when gathering guardrail context before implementing or reviewing code, when creating, editing, retiring, or reviewing GRL rule files or .agents/guardrails/index.md, when running the guardrails CLI (validate, render, review-metadata, next-id), or when setting up, adopting, or bootstrapping a guardrail framework in a repo that does not have one yet. Do NOT use for ordinary lint rule authoring unless the lint rule backs a guardrail, for runt

How do I install the Guardrails skill?

skills CLI (Claude Code, Cursor, Codex and more): npx skills add bolasblack/claude-skills --skill guardrails. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is Guardrails popular?

It lives in bolasblack/claude-skills, which has 5 GitHub stars and 0 forks.

Related skills

Caveman

5

bolasblack/claude-skills

Terse caveman voice: answer first, fluff gone, every technical fact kept. Use for /caveman, "caveman mode", "talk like caveman", "be brief", "less tokens". Stays on until "stop caveman" or "normal mode".

Images, Video & Audio

Agent Centric

5

bolasblack/claude-skills

Track decisions (AGD) with validation and indexing. Use when making design decisions, recording important choices, discussing trade-offs, or when user mentions AGD or decision record.

AI, LLMs & Agents

Codex

5

bolasblack/claude-skills

OpenAI Codex second opinion via the codex CLI — three modes. Review: independent diff review with a pass/fail gate. Challenge: adversarial pass that tries to break the code. Consult: ask Codex anything, with session continuity for follow-ups. Use when the user says "codex review", "codex challenge", "ask codex", or wants a second opinion from Codex.

Software Development

Hexagonal Clean Architecture

5

vnovakovits/claude-skills

Apply Alistair Cockburn's Hexagonal Architecture (Ports and Adapters), Jeffrey Palermo's Onion Architecture, and Robert C. Martin's Clean Architecture — the family of architectural styles that put the domain at the center, push frameworks and I/O to the edges, and enforce inward-pointing dependencies. Use when structuring a non-trivial application, separating business logic from delivery and persistence, deciding where ports live and how adapters plug in, or testing domain logic without spinning up infrastructure.

Software Development

Grok

5

bolasblack/claude-skills

xAI Grok second opinion via the grok CLI — three modes. Review: independent diff review with a pass/fail gate. Challenge: adversarial pass that tries to break the code. Consult: ask Grok anything, with session continuity for follow-ups. Use when the user says "grok review", "grok challenge", "ask grok", or wants a second opinion from Grok.

Software Development

Heyreach Master

8

Freddy-CHEN-ux/00_claude_skills

Internal resource library for HeyReach integration. Contains shared API client, operation scripts, and reference documentation.

Software Development

Browse the most popular Claude skills, all software development skills, or write your own with the skills generator.