A skill your agent uses when creating or updating a website, web app, or browser game, or when a visual layout or interactive tool would help with what the user is doing.

CC0-1.0Auto-check passed

Install Sites

skills CLI
$ npx skills add asgeirtj/system_prompts_leaks --skill sites -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install asgeirtj/system_prompts_leaks sites --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/asgeirtj/system_prompts_leaks.git skills-src && mkdir -p .claude/skills && cp -r skills-src/OpenAI/dots/skills/sites .claude/skills/sites && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
sites
GitHub stars
69k
Token cost
~1.6k tokens
SKILL.md length
895 words
Files
1
Skills in repo
128
Repo updated
First seen
Licence
CC0-1.0

At a glance

A skill your agent uses when creating or updating a website, web app, or browser game, or when a visual layout or interactive tool would help with what the user is doing.

  • Updating a website
  • SKILL.md covers Build and deliver, Creation attribution, When the user asks for a Site and Proactive creation
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md
  • A visual layout

What it does

Sites is an agent skill from asgeirtj/system_prompts_leaks. Use when creating or updating a website, web app, or browser game, or when a visual layout or interactive tool would help with what the user is doing. Read even without a website request; use the skill to decide whether to create one.

Its SKILL.md is about 1.6k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

The repository describes itself as: Documented system prompts from Anthropic - Claude Fable 5.1, Opus 5.5, Claude Design, Claude Code. OpenAI - ChatGPT GPT-6-Astra, Codex. Google - Gemini 3.8 Flash, 3.1 Pro… The licence is CC0-1.0.

When your agent uses it

  • Updating a website
  • A visual layout
  • Interactive tool would help with what the user is doing

Example prompts

  • “/sites”

What it can do on your machine

Read from SKILL.md and the folder at commit 60d44cc. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Sites loads about 1.6k tokens when it runs. Until then it costs about 60 tokens; SKILL.md has 895 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~60
When it runs · the whole SKILL.md, loaded when a task matches
~1.6k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from asgeirtj/system_prompts_leaks at commit 60d44cc, republished under its CC0-1.0 licence (© asgeirtj). 895 words, ~1,557 tokens.

Download SKILL.mdSave it as .claude/skills/sites/SKILL.md (or your agent's skills folder).
name
sites
description
Use when creating or updating a website, web app, or browser game, or when a visual layout or interactive tool would help with what the user is doing. Read even without a website request; use the skill to decide whether to create one.
metadata.version
2026-09-29.creation-intent.1

Sites with your dot

Build and deliver

  • Use the Sites building and hosting skills. Pass these requirements when delegating.
  • Prefer static sites unless the features need a backend, authentication, or a database.
  • Design for desktop and mobile, with particular care for phone layouts.
  • Send the link with a short description of what you created. Explain access simply, using your own wording. For private sites: "Your private site is ready: [link]." or "Here's your site, private to you: [link]."
  • State the Site's current access level. If it is private, immediately explain in the same delivery message that the user can ask you to change its sharing settings. Offer to add collaborators only when invitation eligibility is confirmed; do not infer that eligibility from the available access modes. Use the Site's available sharing options as the source of truth, tailored to the user's account and workspace policies: public access when supported, or workspace access for Business and Enterprise accounts when available. Distinguish workspace access from public access; do not change sharing without the user's request.
  • If creation fails, explain briefly and provide the useful content in chat. Never claim an unfinished Site is ready.
  • When you're sending the link of the site, give a short summary of the content of this site

Creation attribution

  • On the initial create_site or create_and_deploy_basic_site call, set creation_intent from the original human request: user_requested when the user asked for a Site, web app, or browser game; proactive when you chose to create one without that request; unknown when the available context does not establish which. A request for camping advice is not itself a Site request.
  • Decide before delegating and pass the intent through every child that may create the Site. Preserve the original human intent; a parent's instruction to build does not make a proactive Site user-requested.
  • Send creation_intent only when the tool schema supports it. Otherwise omit it and leave attribution unknown; do not invent another field or block creation.
  • Attribution describes the initial creation. Do not reclassify a Site when reusing, editing, or republishing it, including when the user later requests changes.

When the user asks for a Site

  • Send only two build messages. For requested Sites only, this overrides the communication guidance in Sites building and hosting and the delivery-summary instructions above.
  • When reactions are supported, react to the user's request immediately, before sending a message, researching, or delegating. Choose an emoji related to the Site, such as 🐶 for dogs or 🥔 for potatoes.
  • In the first message, acknowledge warmly and briefly say what you will build. Give the only time estimate here: "It will take a few minutes," followed by a promise to send the result when ready. This replaces the numeric estimate from Sites building. Example: "I'll put those options into a trip comparison. It will take a few minutes and I'll send the link when it's ready!"
  • Build quietly without routine progress or publishing updates. When delegating, pass these requirements and have delegates return results to the parent without sending user-facing messages.
  • Once publication succeeds, or explicitly requested local-only work is complete, send one short final message with the verified Site URL or requested local artifact and simple, accurate access wording. For published Sites, include the sharing guidance above in this final message. For requested scheduling, also confirm the outcome and, on success, its timing and enabled or paused state. Mention meaningful differences from the opening plan, without repeating the plan or adding a content summary, feature list, or other suggested next steps.
  • Keep the final text and URL together in one message. On phone channels, use the raw URL, not a Markdown link.
  • Answer user interjections and ask essential blocking questions as needed; these are exceptions to the two-message limit. If creation fails, use the final message to explain briefly and provide useful content in chat. Never claim an unfinished Site is ready.
  • Replace the opening reaction with ✅ only when the Site is ready.
Show full SKILL.md (241 more words)Show less

Proactive creation

Proactiveness means creating a useful Site without waiting for the user to ask. For example, bring scattered trip options, prices, and routes into one comparison.

  • For new proactive briefs, plans, comparisons, and guides, use the template for the user's dot in assets/template/README.md. Pass its instructions and absolute template folder when delegating. Use its HTML/CSS directly for static Sites or as the design reference when backend features are needed.
  • Reconsider as new details arrive. Start when there is enough concrete information for a Site to help beyond the chat answer. Keep quick answers, simple lists, and early brainstorming in chat.
  • Respect the user's requested format, disinterest, and saved preferences. If they decline proactive Sites, remember that preference using the available memory mechanism.
  • Reuse a Site that exists or is being built for the same purpose, preserving its template unless the user asks to change it. Added details or images do not justify another.
  • Those sites must be private. A request to share information does not authorize wider access or sending the Site to others.
  • Give your answer in chat without waiting. Build quietly in the background, without an announcement or delivery estimate, and continue answering the user's messages.
  • Send the finished link once, following the delivery instructions above.
  • Stay within the current conversation. Do not schedule updates or send messages hours later.
  • Examples of proactive sites:
    • events
    • trip planning
    • product comparison/buying guide
    • project plans
    • meal planning
    • workout trackers
    • learning guide

© asgeirtj, CC0-1.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in OpenAI/dots/skills/sites of asgeirtj/system_prompts_leaks.

Open the folder on GitHubat commit 60d44cc

Compare with similar skills

Sites next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Sites compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Sites this skillasgeirtj/system_prompts_leaks69k—~1.6kAutomated safety check: PassCC0-1.0
Docs Website Updatenrwl/nx29k—~1.2kAutomated safety check: PassMIT
Openclaw Updateopenclaw/openclaw392k—~1.5kAutomated safety check: PassMIT
Update Docsvercel/next.js143k—~1.9kAutomated safety check: PassMIT
Website Searchthedaviddias/Front-End-Checklist74k—~556Automated safety check: PassMIT
UI Updatesickn33/agentic-awesome-skills47k1 repos~1.4kAutomated safety check: PassMIT

Similar skills

  • Sync docs commits from master out to the live docs branches in the nx repo: cherry-picks docs( / feat(nx-dev) commits onto website-<major AND the latest <major.<minor.x release branch.

    29k GitHub stars~1.2k tokensUpdated today
    Auto-check passed
  • Openclaw Update

    openclaw/openclaw

    Route routine updates of an existing local or remote OpenClaw Gateway to its installation owner, with automatic session recovery.

    392k GitHub stars~1.5k tokensUpdated today
    DevOps & CloudAuto-check passed
  • Update Docs

    vercel/next.js

    Official

    This skill should be used when the user asks to "update documentation for my changes", "check docs for this PR", "what docs need updating", "sync docs with code", "scaffold docs for this feature"…

    143k GitHub stars~1.9k tokensUpdated today
    Documents & OfficeAuto-check passed
  • Website Search

    thedaviddias/Front-End-Checklist

    A skill your agent uses when applies to sites with internal site search functionality.

    74k GitHub stars~556 tokensUpdated 4 days ago
    Marketing & SEOAuto-check passed
  • UI Update

    sickn33/agentic-awesome-skills

    Update StyleSeed engine in your project — analyzes what's outdated and updates safely

    47k GitHub starsUsed in 1 repo~1.4k tokens
    Auto-check passed
  • Guides edits to the Repomix documentation site's VitePress configuration, per-language content, and translation workflow across 14 languages.

    29k GitHub starsUsed in 1 repo~342 tokens
    Frontend & DesignAuto-check passed

More from asgeirtj/system_prompts_leaks

All 128 skills in this repo
  • Fleet Manager for Agent Sessions

    asgeirtj/system_prompts_leaks

    Shows one digest of coding-agent sessions across your connected machines and lets you open, read, steer, approve, stop and close them, over Herdr, tmux or MSP.

    69k GitHub stars~2.5k tokensUpdated today
    Auto-check passed
  • Muse Code Product Doctor

    asgeirtj/system_prompts_leaks

    Diagnoses a Muse Code installation's own failures from binary and session evidence, instead of treating the report as an ordinary repository bug.

    69k GitHub stars~3.5k tokensUpdated today
    Auto-check passed
  • DOCX

    asgeirtj/system_prompts_leaks

    A skill your agent uses whenever the user wants to create, read, edit, or manipulate Word documents (.docx) or Word templates (.dotx).

    69k GitHub stars~1.9k tokensUpdated today
    Auto-check passed
  • Agents Project Coordinator

    asgeirtj/system_prompts_leaks

    Runs a goal as a project in which the agent coordinates separate agent threads, judging when to split the work, and interviews you first when nothing can be verified.

    69k GitHub stars~2.9k tokensUpdated today
    Auto-check passed
  • Muse Plugin Creator

    asgeirtj/system_prompts_leaks

    Creates and validates a new native Muse plugin package in the current workspace, limited to five capability families, and leaves installation to you.

    69k GitHub stars~1.8k tokensUpdated today
    Auto-check passed
  • Deep Research

    asgeirtj/system_prompts_leaks

    A skill your agent uses when the user's prompt requires (1) researching a topic across multiple sources, comparing options or alternatives, analyzing trends or history, understanding markets or…

    69k GitHub stars~3.3k tokensUpdated today
    Auto-check passed

Questions about Sites

What does Sites do?

A skill your agent uses when creating or updating a website, web app, or browser game, or when a visual layout or interactive tool would help with what the user is doing. Sites is an agent skill from asgeirtj/system_prompts_leaks. Use when creating or updating a website, web app, or browser game, or when a visual layout or interactive tool would help with what the user is doing.

When should I use Sites?

Sites fits situations like: updating a website; A visual layout; interactive tool would help with what the user is doing.

How do I install Sites in Claude Code?

Run `npx skills add asgeirtj/system_prompts_leaks --skill sites -a claude-code`. Or copy the skill folder (OpenAI/dots/skills/sites in asgeirtj/system_prompts_leaks) into .claude/skills/sites in your project. Claude Code loads it when a task matches its description.

How do I install Sites in Codex?

Run `npx skills add asgeirtj/system_prompts_leaks --skill sites -a codex`. Or copy the skill folder (OpenAI/dots/skills/sites in asgeirtj/system_prompts_leaks) into .agents/skills/sites in your project. Codex loads it when a task matches its description.

Can I use Sites in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add asgeirtj/system_prompts_leaks --skill sites -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/sites, .gemini/skills/sites, .github/skills/sites and .opencode/skills/sites in your project.

What does Sites need to run?

SKILL.md names no scripts, command-line tools or credentials: Sites is instructions for the agent only.

Does Sites access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Sites safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Sites use?

Sites is published under the CC0-1.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Sites use?

About 1.6k tokens (SKILL.md is roughly 6.2k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Sites?

Skills that share tags, products or a category with Sites: Docs Website Update (nrwl/nx, 29k stars), Openclaw Update (openclaw/openclaw, 392k stars), Update Docs (vercel/next.js, 143k stars) and Website Search (thedaviddias/Front-End-Checklist, 74k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Sites?

asgeirtj (a GitHub user) maintains it in asgeirtj/system_prompts_leaks, which has 69,280 GitHub stars. The repository holds 128 skills in this directory. The repository was last updated on October 10, 2026.

Source: asgeirtj/system_prompts_leaks on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.