What is get-shit-pretty CLAUDE.md?

Repository instructions for developing GSP, a framework of reusable instructions and coding agents that works with several coding-agent tools. They describe required file locations, versioning, installation, and Git workflow.

Use them when changing GSP source, adding dependencies, preparing pull requests, publishing the package, or deciding which files should and should not be edited.

They give contributors shared rules for making changes safely and consistently, including how to work with protected branches and generated files.

get-shit-pretty CLAUDE.md — Instructions file for Claude Code and Codex by jubscodes: Repository instructions for developing GSP, a framewor

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods

npx agentmods add instructions/jubscodes/get-shit-pretty/claude-md

Clone the repo

git clone --depth 1 https://github.com/jubscodes/get-shit-pretty

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/instructions/jubscodes/get-shit-pretty/claude-md.svg)](https://agentmods.dev/instructions/jubscodes/get-shit-pretty/claude-md)

Your own site

<a href="https://agentmods.dev/instructions/jubscodes/get-shit-pretty/claude-md"><img src="https://agentmods.dev/badge/instructions/jubscodes/get-shit-pretty/claude-md.svg" alt="Measured on agentmods" height="20"></a>

Per session 2,792 This file is loaded in full into every session.

When invoked 2,792 The same file — it is already loaded in full.

Security scan A 0 findings. Scan, not verified.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.02792	$0.02792
Opus 5	$0.01396	$0.01396
Sonnet 5	$0.00558	$0.00558
Haiku 4.5	$0.00279	$0.00279

Measured 6d ago against content hash 0f282dd99ba8, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

get-shit-pretty CLAUDE.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

CLAUDE.md · 193 lines

How it starts

The opening of the file, as written. The whole thing — 193 lines — stays where its author put it; the contents beside it link to each section on GitHub.

GSP — Get Shit Pretty

Must-always / Must-never

Always edit source under gsp/ — never inside .claude/, .opencode/, .gemini/, .agents/, .codex/ (symlinks or installer-managed).
Always put new package dependencies in devDependencies — the npm package ships zero prod deps.
Always ship changes via PR — main is protected, no direct push. Squash-merge via gh pr merge --squash.
Never edit inside .design/ from a skill — it's user/agent output only.
Never add model: or effort: to skill frontmatter — model selection is the user's choice.
Never hardcode output paths in agents — use "path provided by the skill that spawned you".

Design engineering system for Claude Code, OpenCode, Gemini, and Codex.

Git workflow

main is protected — all changes require a PR (no direct push). Use feature/release branches and squash merge via gh pr merge --squash.

This repository

This repo is both:

Source and npm package — where the GSP agentic framework is built, versioned, and published via npm publish. The package.json files field controls what ships: .mcp.json, bin, scripts, gsp.
GSP consumer — GSP is installed here too (e.g. .claude/ symlinks). You can run GSP workflows in this workspace while developing the framework.

Edit source under gsp/; the installer keeps runtimes in sync. Never edit inside .claude/ (or other runtime dirs) directly — they point at or are populated from source.

Architecture

Dual-diamond: Branding (discover → strategy → identity → patterns) + Project (brief → research → design → critique → build → review). Verbal identity is merged into brand-strategy (4 phases, not 5).

Two skill layers

Expertise skills (knowledge owners): gsp-color, gsp-typography, gsp-visuals, gsp-accessibility, gsp-style

Own domain knowledge as sibling files (domains/, references/). Serve the full pipeline, not just one phase. Two consumption patterns:

Read (passive) — pipeline skill or agent reads expertise skill's sibling files for domain context
Invoke (active) — pipeline skill calls the expertise skill to run its logic (e.g. --enrich, --validate)

Read the full file on GitHub · 193 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

6d ago First seen · 193 lines · 2,792 tokens per session scan A 0f282dd99ba8

Subscribe to this mod's changes

get-shit-pretty CLAUDE.md is an instructions file published in the GitHub repository jubscodes/get-shit-pretty (54 stars, last pushed 3mo ago), licensed MIT. It adds 2,792 tokens to every session, about $0.0140 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

ai ai-tools claude-code cli design-engineering design-systems

design-sandbox AGENTS.md

Instructions for memi-design/design-sandbox, covering this is not the next.js you know, design2code.ground, quick orientation and pointers for non-claude tools.

memi-design/design-sandbox · 306 tokens

design-sandbox CLAUDE.md

Instructions for memi-design/design-sandbox, a project described as: memi-ready Next.js, Tailwind, shadcn, MCP, and Agent Skills sandbox for interface understanding and design-to-code exploration.

memi-design/design-sandbox · 5 tokens

vscode buildNext.instructions.md

Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).

microsoft/vscode · 6,785 tokens

get-shit-pretty CLAUDE.md