Skip to main content
Glama

Pagetool

Create a site from a brief

create_site_from_brief

BUILD FIRST, ASK LATER. Creates a complete, live draft website in one call - free, instant, no account and no sign-in. Call this as soon as someone asks for a website, using whatever you already know and sensible defaults for the rest, then show them the returned preview_url and refine it with edit_section and set_theme. Do not interview the user first. Composition is deterministic and no language model is involved, so anything you leave blank renders as a visible placeholder rather than invented copy. You can edit or claim this draft for 7 days from creation.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "business_name": "<business_name>"
    },
    "name": "create_site_from_brief"
  }
}

Input Schema

NameRequiredDescriptionDefault
taglineNoShort tagline/subheading, max 200 chars.
sectionsNoComma-separated section kinds to include, in order, e.g. "hero,about,services,contact". Allowed values: hero, about, services, features, gallery, testimonials, pricing, faq, hours, contact, cta, footer, menu, map, social, header, products, reviews. Defaults to hero,about,services,contact when omitted.
templateNoOptional starter template - the same ones people pick in the editor. storefront: products with buy buttons, why-us, shipping FAQ. event: a recurring night - what to expect, rules, sign-up, where. shop: a place to visit - story, hours, photos, map, socials. services: services with prices, reviews, service area, hours. nonprofit: mission, get involved, give. Your fields win: sections replace the template sections; content_json replaces it per kind, other kinds keep the sample text. business_name is still required.
font_bodyNoBody font family name. Defaults to Inter.
site_nameNoOptional display name for the site row (defaults to business_name).
draft_tokenNoThe draft token returned by draft_create, if you are working on an anonymous draft. Only needed when your client cannot send the x-pagetool-draft-token header.
content_jsonNoPer-section content, keyed by section kind (the kind must also be in sections), as a JSON string or an object, max 64KB. Keys per kind - hero: heading, body, ctaLabel, ctaHref, image, alt. about: heading, body, image, alt. cta: heading, body, ctaLabel, ctaHref. services / features / testimonials: heading, items[] of {title, body, href, linkLabel, image, alt}. pricing: heading, items[] of {title, body, price, href, linkLabel, image, alt}. gallery: heading, items[] of {image, alt, title}. faq: heading, items[] of {title, body} (question/q and answer/a are accepted for title/body). hours: heading, body. contact: heading, body. footer: none. menu: heading, items[] of {title, body, price, group} (grouped by group). map: heading, address (defaults to contact_address; embeds a Google map). social: heading, items[] of {title, href}. products: heading, items[] of {title, body, price, href, linkLabel, image, alt, soldOut} (a shop grid: photo, name, price and a Buy now button; soldOut true shows a Sold out badge instead of the button). reviews: heading, placeId (the business Google place id; live reviews show on Creator plans and up, at most 5, fetched from Google at view time; never write review text). header: none (a new site and a new page always start with one: the business name plus links to its sections). An item href renders a link (text = linkLabel, default "Learn more"; in pricing a button, default "Buy now"). Hrefs may be http(s), mailto:, tel:, /path or an in-page anchor like #pricing (every section gets its kind as its id). An image renders a photo with alt - use import_image first to get a hosted copy. Max 12 items per kind. Anything ignored, cut or dropped is listed in the response warnings array - read it. Example: {"hero":{"heading":"Fresh bread daily","body":"Baked before sunrise.","ctaLabel":"Order now","ctaHref":"/order"},"services":{"items":[{"title":"Sourdough","body":"24-hour ferment."},{"title":"Croissants","body":"All butter."}]}}
font_headingNoHeading font family name. Defaults to Inter.
business_nameYesThe business or site name. Required, max 120 chars.
contact_emailNoContact email shown in the contact section, if included.
contact_phoneNoContact phone shown in the contact section, if included.
contact_addressNoContact address shown in the contact section, if included.
palette_neutralNoNeutral text/background colour as a #rrggbb hex string. Defaults to #0f172a.
palette_primaryNoPrimary brand colour as a #rrggbb hex string. Defaults to #2563eb.

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observed

TDQS

A4.9/5.0
Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Goes well beyond the annotations, which only say it is a non-idempotent, non-destructive, non-open-world write. The description adds that it is free and needs no account or sign-in, that composition is deterministic with no LLM so blanks become visible placeholders rather than invented copy, and that the draft is editable/claimable for 7 days — all material behavior an agent cannot infer from structured fields.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The imperative 'BUILD FIRST, ASK LATER' is front-loaded as the operative rule, followed by what it does, the trigger, the anti-pattern, the determinism caveat, and the retention window. Every sentence carries a distinct instruction; none is filler.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a 14-parameter creation tool with no output schema, the description covers what is returned (preview_url, warnings array), what happens to unspecified fields (placeholders), and what the agent should do next. Nothing needed to call it correctly is missing, and the heavy per-field documentation is correctly delegated to the schema.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so the baseline is 3, but the description adds real meaning: it tells the agent to use 'whatever you already know and sensible defaults for the rest' and warns that omitted content renders as a visible placeholder, which informs the omit-vs-fill decision. It also notes the response carries preview_url and a warnings array, though it does not add per-parameter semantics beyond the schema.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource — 'Creates a complete, live draft website in one call' — and immediately scopes it as draft, not published. The follow-up sentence naming edit_section and set_theme for refinement makes the boundary between creation and iteration explicit.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

'Call this as soon as someone asks for a website' defines the trigger precisely, and 'Do not interview the user first' is an explicit when-not directive that prevents a common failure mode. It also names the follow-on tools (edit_section, set_theme) so the agent knows where this tool's responsibility ends.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources