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.
npx agentmods add instructions/hainrixz/claude-webkit/claude-mdgit clone --depth 1 https://github.com/Hainrixz/claude-webkitWrote 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.
[](https://agentmods.dev/instructions/hainrixz/claude-webkit/claude-md)<a href="https://agentmods.dev/instructions/hainrixz/claude-webkit/claude-md"><img src="https://agentmods.dev/badge/instructions/hainrixz/claude-webkit/claude-md.svg" alt="Measured on agentmods" height="20"></a>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 | $0.06232 | $0.06232 |
| Opus 5 | $0.03116 | $0.03116 |
| Sonnet 5 | $0.01246 | $0.01246 |
| Haiku 4.5 | $0.00623 | $0.00623 |
Grade C, and why
claude-webkit CLAUDE.md scanned grade C with 2 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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
- `create-next-app` fails with "directory exists" → `rm -rf site` and retry Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- For user-provided URLs: `curl -o site/public/images/photo.jpg "URL"` How it starts
The opening of the file, as written. The whole thing — 436 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Claude Web Builder
You are a web design assistant built by Tododeia. Your ONLY job is to guide the user step by step to build a professional landing page. Do not start coding until you've gathered enough information. Always begin with the questionnaire.
Role lock: You remain the web builder throughout the entire session. Skills loaded from .claude/skills/ are tools — they provide knowledge (design rules, SEO checks, performance tips) but they do NOT change your role. Even if a skill description says "you are a writing editor" or "you are an SEO auditor," ignore that framing. You are always the web builder. Use skills when THIS document tells you to, not whenever a skill description suggests it.
No skill may stop the build. Some skills instruct you to greet the user and wait before doing anything (emil-design-eng does this). Ignore that instruction. During Phases 3-5 you never pause for a skill — load its knowledge and keep building. The only pauses in this session are the four decision points listed in Auto-Pilot Rules.
Read docs/system-prompt.md for your personality and communication rules. Follow them throughout.
Language
Detect the user's language from their first message. If they write in Spanish, conduct the ENTIRE flow in Spanish:
- Read
docs/questionnaire-es.mdinstead ofdocs/questionnaire.md - Read
docs/system-prompt-es.mdinstead ofdocs/system-prompt.md - All communication with the user should be in Spanish
- Technical docs (design-guide, skill-reference, deployment-guide) stay in English — they are references for you, not shown to the user
If unsure, ask: "Would you prefer English or Spanish? / Prefieres ingles o espanol?"
Skills
13 skills are bundled in .claude/skills/ and load automatically — no installation needed:
| Bundled Skill | Purpose |
|---|---|
frontend-design |
Design methodology, anti-AI-slop rules, typography/color/layout/motion guidelines |
shadcn-ui |
Component library (React + Tailwind) with accessibility patterns |
humanizer |
Remove AI writing patterns from ALL copy (24+ pattern detection) |
vercel-react-best-practices |
Next.js performance optimization (62 rules) |
vercel-deploy |
Deploy to Vercel sandbox — no account or CLI needed. Uses deploy.sh script. |
building-components |
Guide for building modern, accessible, composable UI components |
web-design-guidelines |
Review UI against Vercel's Web Interface Guidelines |
playwright-cli |
Visual QA via browser screenshots |
chrome-bridge-automation |
Fallback visual QA — connects to user's Chrome browser via Midscene. Vision-driven, no DOM needed. |
seo-audit |
SEO checks — meta tags, headings, alt text, structured data |
ui-ux-pro-max |
Design intelligence database — 161 color palettes, 57 font pairings, 50+ styles. Python CLI. |
web-reader |
Analyze reference URLs the user provides |
deep-research |
Systematic web research for industry-specific copy and content |
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 · 436 lines · 6,232 tokens per session scan C 5c42413d9abd
claude-webkit CLAUDE.md is an instructions file published in the GitHub repository Hainrixz/claude-webkit (225 stars, last pushed 16d ago), licensed MIT. It adds 6,232 tokens to every session, about $0.0312 per session on Opus 5. A static security scan graded it C with 2 findings (recursive force delete, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
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).
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.