Create Site logo

Create Site

Organization
adobe
create-site

Use this when a user wants to create a brand-new AEM Edge Delivery site from scratch and no GitHub repository or DA content exists yet — triggers include 'set up a new site', 'create a new EDS project', or 'onboard a new site'. Covers creating the GitHub repo from the boilerplate, installing aem-code-sync, seeding initial DA content (nav, footer, homepage), and returning a live preview URL. For importing or migrating existing pages use page-import; for building blocks on an existing site use content-driven-development.

Overview

Publisheradobe
Repositoryskills
Skill namecreate-site
Stars
298
Forks
92
Bundled files
2
LicenseApache-2.0
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 2 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by adobe on GitHub. Read the source before you install it.

Installation

Install the Create Site AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/adobe/skills/tree/main/plugins/aem/edge-delivery-services/skills/create-site
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/adobe/skills.git /tmp/skills
mkdir -p .claude/skills
cp -r /tmp/skills/plugins/aem/edge-delivery-services/skills/create-site .claude/skills/create-site
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Create Site in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Create Site on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Create Site is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Create a New AEM Edge Delivery Site

This skill walks through the full onboarding flow for a new AEM Edge Delivery site. It handles everything that can be automated and clearly signals the steps that require human action.

When to Use This Skill

Use this skill when:

  • A user wants to create a brand-new AEM Edge Delivery site from scratch
  • A user asks to "set up a new site", "create a new EDS project", or "onboard a new site"
  • No GitHub repository or DA content exists yet for the project

Do NOT use this skill for:

  • Importing or migrating existing pages (use page-import skill)
  • Building or modifying blocks on an existing site (use content-driven-development skill)

Prerequisites

  • A GitHub account with permission to create repositories in the target org
  • An Adobe IMS account with access to DA (da.live)
  • gh CLI authenticated (gh auth status) or a GitHub personal access token with repo scope
  • Node.js (for DA token management via da-auth-helper)

Related Skills

  • page-import — Import existing pages into the newly created site
  • content-driven-development — Build and modify blocks once the site exists
  • building-blocks — Implement new block code

Step 0: Create TodoList

Create a checklist to track progress (use your agent's task-tracking tool if available):

  1. Gather inputs — org, repo name, site name collected
  2. Create GitHub repository — repo created from boilerplate template
  3. Install aem-code-sync (human action) — GitHub App installed on repo
  4. Authenticate with DA — valid IMS token obtained
  5. Create initial content in DA — nav, footer, index created
  6. Trigger preview — all three paths return 200/201
  7. Hand off — preview URL and DA links delivered to user

Step 1: Gather Inputs

Ask the user for the following. Do not proceed until all required inputs are provided.

  1. GitHub org — the GitHub organization or username where the repo will be created (e.g. my-org)
  2. Project name — the repository name, lowercase, hyphens only (e.g. my-site)
  3. Site name — the human-readable name used in content (e.g. My Site). If not provided, derive it from the project name.

Store as: {{ORG}}, {{REPO}}, {{SITE_NAME}}


Step 2: Create GitHub Repository

Create a new repository using the adobe/aem-boilerplate template.

Option A — GitHub CLI (preferred, handles auth automatically):

bash
gh repo create {{ORG}}/{{REPO}} \
  --template adobe/aem-boilerplate \
  --description "{{SITE_NAME}} — AEM Edge Delivery site" \
  --public

Check if gh is available with gh auth status. If not authenticated, run gh auth login first.

Option B — GitHub API (if gh CLI is not available):

POST https://api.github.com/repos/adobe/aem-boilerplate/generate
Authorization: Bearer {{GITHUB_TOKEN}}
Content-Type: application/json

{
  "owner": "{{ORG}}",
  "name": "{{REPO}}",
  "description": "{{SITE_NAME}} — AEM Edge Delivery site",
  "private": false,
  "include_all_branches": false
}

To obtain a token: https://github.com/settings/tokens/new — scope: repo.

Success: HTTP 201 (API) or exit code 0 (CLI). The repo is now live at https://github.com/{{ORG}}/{{REPO}}.


Step 3: Install aem-code-sync (human action required)

The aem-code-sync GitHub App connects the repository to AEM's content delivery pipeline. This step cannot be automated — the user must complete it in the browser.

Tell the user:

Action required: Install the AEM Code Sync app on your new repository.

  1. Open this URL: https://github.com/apps/aem-code-sync/installations/new
  2. Under "Repository access", select Only select repositories
  3. Choose {{ORG}}/{{REPO}} from the list
  4. Click Save

Reply "done" when complete.

Wait for confirmation before proceeding.

Verify: After confirmation, check that https://admin.hlx.page/status/{{ORG}}/{{REPO}}/main/ returns a valid JSON response (not 404). If it does, the app is correctly installed.


Step 4: Authenticate with DA

DA requires Adobe IMS authentication. Choose the appropriate path:

Option A — da-auth-helper (preferred)

da-auth-helper (https://github.com/adobe-rnd/da-auth-helper) caches IMS tokens at ~/.aem/da-token.json. Always check the cache first before triggering a new OAuth flow.

  1. Check for a valid cached token:
bash
node -e "
  const fs = require('fs');
  const p = process.env.HOME + '/.aem/da-token.json';
  if (!fs.existsSync(p)) { console.log('No cache'); process.exit(1); }
  const t = JSON.parse(fs.readFileSync(p));
  console.log('Valid:', t.expires_at > Date.now());
  console.log('Expires:', new Date(t.expires_at).toISOString());
"
  1. If valid, capture the token and skip to Step 5:
bash
DA_TOKEN=$(node -e "const t = require(process.env.HOME + '/.aem/da-token.json'); process.stdout.write(t.access_token);")
  1. If missing or expired, install da-auth-helper from GitHub (it is not published to npm) and refresh:
bash
npm install -g github:adobe-rnd/da-auth-helper
da-auth-helper token

This opens a browser for Adobe IMS login and writes the new token to ~/.aem/da-token.json. Then capture it as in step 2.

Option B — DA MCP is configured

If the DA MCP server is available, trigger the authentication tool to start the OAuth flow and share the authorization URL with the user.

Option C — Manual token

Ask the user to obtain an IMS token from their browser (e.g. from the DA network tab or an existing session) and paste it. Store as {{DA_TOKEN}}.


Step 5: Create Initial Content in DA

Create the three mandatory pages every EDS site requires. Use the templates below exactly — they are pre-validated for EDS compliance.

Option A — DA MCP: Call the DA create source tool three times with the content below.

Option B — DA API:

Write each file to a temp file first, then POST using @ syntax. Inline multiline content with -F 'data=...' causes curl to fail (exit 26). Use /usr/bin/curl explicitly to avoid PATH resolution issues in subshells.

bash
cat > /tmp/nav.html << 'EOF'
<nav content>
EOF
/usr/bin/curl -s -o /dev/null -w "%{http_code}" -X POST "https://admin.da.live/source/{{ORG}}/{{REPO}}/nav.html" \
  -H "Authorization: Bearer {{DA_TOKEN}}" \
  -F "data=@/tmp/nav.html;type=text/html"

Repeat for footer.html and index.html.

Verify: After each POST, expect HTTP 201. If you get 401, the token has expired — return to Step 4.


nav.html

html
<main>
  <div>
    <p><a href="/">{{SITE_NAME}}</a></p>
  </div>
  <div>
    <ul>
      <li><a href="/">Home</a></li>
    </ul>
  </div>
  <div></div>
</main>

footer.html

html
<main>
  <div>
    <p>© 2024 {{SITE_NAME}}. All rights reserved.</p>
  </div>
</main>

index.html

html
<main>
  <div>
    <h1>Welcome to {{SITE_NAME}}</h1>
    <p>Your new site is ready. Start editing this page in DA.</p>
  </div>
</main>

Step 6: Trigger Preview

Preview pulls the DA content into the AEM delivery pipeline and makes it accessible on the .aem.page domain.

DA-sourced content requires the Bearer token on preview requests — even for public repos. Use /usr/bin/curl explicitly.

bash
/usr/bin/curl -s -o /dev/null -w "%{http_code}" -X POST "https://admin.hlx.page/preview/{{ORG}}/{{REPO}}/main/nav" \
  -H "Authorization: Bearer {{DA_TOKEN}}"
/usr/bin/curl -s -o /dev/null -w "%{http_code}" -X POST "https://admin.hlx.page/preview/{{ORG}}/{{REPO}}/main/footer" \
  -H "Authorization: Bearer {{DA_TOKEN}}"
/usr/bin/curl -s -o /dev/null -w "%{http_code}" -X POST "https://admin.hlx.page/preview/{{ORG}}/{{REPO}}/main/" \
  -H "Authorization: Bearer {{DA_TOKEN}}"

Success: HTTP 200 or 201 for each. The homepage is now live at:

https://main--{{REPO}}--{{ORG}}.aem.page/

Step 7: Confirm and Hand Off

Tell the user:

Your site is ready!

  • Preview: https://main--{{REPO}}--{{ORG}}.aem.page/
  • Browse content in DA: https://da.live/#/{{ORG}}/{{REPO}}/
  • Edit homepage: https://da.live/edit#/{{ORG}}/{{REPO}}/index
  • Edit nav: https://da.live/edit#/{{ORG}}/{{REPO}}/nav
  • Edit footer: https://da.live/edit#/{{ORG}}/{{REPO}}/footer
  • GitHub repo: https://github.com/{{ORG}}/{{REPO}}

To start developing locally:

bash
git clone https://github.com/{{ORG}}/{{REPO}}.git
cd {{REPO}}
npm install
aem up

What would you like to do next — add more pages, customize a block, or set up a custom domain?


Troubleshooting

SymptomLikely causeFix
Step 2 returns 422Repo name already existsAsk user for a different name
Step 3 verify returns 404aem-code-sync not installedRe-send the installation URL
Step 4 cached token missing/expiredNo prior DA session on this machineInstall da-auth-helper from GitHub (npm install -g github:adobe-rnd/da-auth-helper) and run da-auth-helper token
Step 5 curl exits with code 26Inline multiline content in -F flagWrite content to a temp file and use @/tmp/file.html syntax
Step 5 returns 401Expired or missing IMS tokenRe-check ~/.aem/da-token.json expiry; ask user for a fresh token
Step 5 returns 403Token lacks permission for this org/repoConfirm the user has write access to {{ORG}}/{{REPO}} in DA
Step 6 returns 401DA-sourced content requires auth on previewAdd -H "Authorization: Bearer {{DA_TOKEN}}" to preview requests
Step 6 returns 404aem-code-sync not installed correctlyVerify Step 3, then retry
curl: command not found in scriptsPATH not resolved in subshellUse /usr/bin/curl explicitly
Preview URL shows blank pagenav or index not previewedRe-run Step 6 for the failing path

Reference

DA URL patterns

  • Browse folder: https://da.live/#/{{org}}/{{repo}}{{folder-path}}
  • Edit HTML document: https://da.live/edit#/{{org}}/{{repo}}{{path-without-extension}}
  • Edit JSON/sheet: https://da.live/sheet#/{{org}}/{{repo}}{{path-without-extension}}

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Create Site AI skill do?

Use this when a user wants to create a brand-new AEM Edge Delivery site from scratch and no GitHub repository or DA content exists yet — triggers include 'set up a new site', 'create a new EDS project', or 'onboard a new site'. Covers creating the GitHub repo from the boilerplate, installing aem-code-sync, seeding initial DA content (nav, footer, homepage), and returning a live preview URL. For importing or migrating existing pages use page-import; for building blocks on an existing site use content-driven-development.

Why use Create Site on TypingMind?

Because you install it once and use it with any model. Create Site is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Create Site in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/adobe/skills/tree/main/plugins/aem/edge-delivery-services/skills/create-site. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Create Site?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Create Site?

As many as you like. As long as a model supports skills, you can use Create Site with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Create Site AI skill free?

Yes. It is published on GitHub by adobe under the Apache-2.0 license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇