MCP Tools — Content & Storefront

These tools let you manage every layer of your site's content from an MCP client: websites and pages, media, navigation, taxonomies, reusable block templates, custom post types, branding profiles, and your storefront (products, orders, customers, discounts, and reviews). Most write tools run through an approval workflow — when your API key is configured to require approval, mutations return a pending: true response with an approval.url you or a reviewer must click before the change goes live.

For authentication setup and scope reference, see MCP Overview.

Approval responses#

When a write tool is staged for approval instead of applied immediately, the response shape is:

{
  "pending": true,
  "pendingId": 42,
  "summary": "Create post \"About Us\" on website 7",
  "status": "pending",
  "approval": {
    "url": "https://app.simplerdevelopment.com/approve/abc123",
    "expiresAt": "2026-06-05T08:00:00.000Z"
  }
}

Tools that always mint an approval URL (even on direct-apply keys) also return approval on success — you can share that link with a reviewer to preview the result.

Websites#

sites_list#

List all websites owned by your account.

• Auth: sites:read
• Inputs: none

Response

[
  {
    "id": 7,
    "clientId": 3,
    "name": "Acme Corp",
    "domain": "acme.com",
    "active": true,
    "publicAccess": true,
    "brandingProfileId": 2,
    "createdAt": "2025-01-10T12:00:00.000Z"
  }
]

Tool call example

{ "name": "sites_list", "arguments": {} }

sites_update#

Update metadata on a website (name, domain, description, active flag, public-access gating, branding profile). DNS/Vercel provisioning is not triggered.

• Auth: sites:write

Response — the updated website row, plus approval if staged.

Tool call example

{
  "name": "sites_update",
  "arguments": { "id": 7, "name": "Acme Corp v2", "active": true }
}

sites_get_custom_code#

Get the site-wide custom CSS and JS. Cascade order is: site → CPT → per-post.

• Auth: sites:read

Response

{ "customCss": "body { font-size: 16px; }", "customJs": "" }

sites_update_custom_code#

Stage changes to site-wide custom CSS/JS. Changes are written to draft fields and do not go live until you call sites_publish_custom_code.

• Auth: sites:write

Response

{
  "draftCustomCss": "body { color: red; }",
  "draftCustomJs": "",
  "liveCustomCss": "body { color: black; }",
  "liveCustomJs": "",
  "draftUpdatedAt": "2026-06-04T10:00:00.000Z",
  "note": "Wrote to draft fields. Call sites_publish_custom_code to make changes live."
}

sites_publish_custom_code#

Promote the draft site-wide CSS/JS to live. Copies draft_custom_css → custom_css and draft_custom_js → custom_js, then clears the draft fields.

• Auth: sites:write

Response

{ "customCss": "body { color: red; }", "customJs": "" }

Posts & Pages#

posts_list#

List content posts for a website. Returns a slim projection (no content blob) by default.

• Auth: sites:read

Response

[
  {
    "id": 101,
    "websiteId": 7,
    "title": "About Us",
    "slug": "about-us",
    "postType": "page",
    "published": true,
    "publishedAt": "2026-01-15T09:00:00.000Z",
    "excerpt": null,
    "createdAt": "2026-01-10T08:00:00.000Z"
  }
]

posts_get#

Fetch a single post by id. Prefer this over posts_list when you need one post in full.

• Auth: sites:read

Response — the post row (plus content / customCss / customJs / SEO fields if includeContent: true).

Tool call example

{ "name": "posts_get", "arguments": { "id": 101, "includeContent": true } }

posts_create#

Create a blog post or page. Mints an approval URL you can share with a reviewer.

• Auth: sites:write

Response

{
  "id": 202,
  "websiteId": 7,
  "title": "New Landing Page",
  "slug": "new-landing",
  "postType": "page",
  "published": false,
  "approval": {
    "url": "https://app.simplerdevelopment.com/approve/xyz789",
    "expiresAt": "2026-06-05T10:00:00.000Z"
  }
}

posts_update#

Update a post. Supports full SEO fields. Mints an approval URL.

• Auth: sites:write

Response — updated slim post row + approval.

posts_fork#

Duplicate a published post into a new draft tied to the original via parentPostId. Edit the fork, share its approval URL for review, and publish it when approved without taking the live page down.

• Auth: sites:write

Response

{
  "id": 203,
  "title": "About Us (fork)",
  "slug": "about-us-fork-1abc2",
  "parentPostId": 101,
  "published": false,
  "approval": { "url": "https://app.simplerdevelopment.com/approve/forklink" }
}

posts_delete#

Permanently delete a post. Revisions cascade.

• Auth: sites:write

Response — { "success": true, "id": 101 } or a pending approval envelope.

posts_upload_html#

Upload a single HTML/XHTML file (base64-encoded) as a draft page post wrapping an html-embed block. Nav/header tags are stripped, referenced assets are imported to media, and the cleaned file is stored in S3. Max 1 MB decoded. Restricted to staff-role API keys.

• Auth: sites:write

Response

{
  "id": 204,
  "title": "my-page",
  "slug": "my-page",
  "postType": "page",
  "published": false,
  "importedAssets": 3,
  "skippedAssets": 1,
  "url": "https://media.example.com/media/uuid.html"
}

posts_upload_html_zip#

Upload a zip archive (base64-encoded) containing index.html + supporting assets as a draft page. Every file is uploaded to S3; relative refs resolve through a shared media-proxy prefix. Limits: 50 MB uncompressed, 200 files, 10 MB per file. Restricted to staff-role API keys.

• Auth: sites:write

Response — same shape as posts_upload_html plus bundleFileCount (number of files extracted from the zip) and bundlePrefix (the shared S3 key prefix for all extracted files).

posts_list_revisions#

Revision history for a post (autosaves, manual saves, publishes).

• Auth: sites:read

Response — array of revision rows ordered newest-first.

Taxonomies (Categories & Tags)#

taxonomies_list#

List categories and tags for a website.

• Auth: sites:read

Response

{
  "categories": [{ "id": 5, "name": "News", "slug": "news", "color": "#2563eb" }],
  "tags": [{ "id": 12, "name": "announcement", "slug": "announcement" }]
}

taxonomies_create_category#

Create a category on a website. May be staged for approval.

• Auth: sites:write

taxonomies_create_tag#

Create a tag on a website.

• Auth: sites:write

posts_set_taxonomies#

Replace the categories and/or tags assigned to a post. Call taxonomies_list first to look up ids. Omitted arrays are left unchanged.

• Auth: sites:write

Response

{ "postId": 101, "categoryIds": [5], "tagIds": [12, 14] }

Media#

media_list#

List uploaded media assets for your account.

• Auth: media:read

Response — array of media rows (id, filename, mimeType, fileSize, url, alt, caption, createdAt).

media_upload_from_url#

Fetch a public URL and store the file in your media library. Max 25 MB. SSRF-guarded — internal/private URLs are rejected.

• Auth: media:write

Response — the new media row including url (the internal URL to use in posts, decks, emails).

Tool call example

{
  "name": "media_upload_from_url",
  "arguments": {
    "url": "https://example.com/hero.jpg",
    "alt": "Hero image",
    "websiteId": 7
  }
}

media_upload_presign#

Mint a short-lived S3 PUT URL for a direct local-file upload. After your curl --upload-file succeeds, call media_register to create the media row.

Allowed MIME types: image/png, image/jpeg, image/gif, image/webp, image/avif, image/svg+xml, application/pdf, video/mp4, video/webm, video/quicktime, audio/mpeg, audio/ogg, audio/wav. Max 25 MB.

• Auth: media:write

Response

{
  "mediaKey": "media/uuid.jpg",
  "storedFilename": "uuid.jpg",
  "uploadUrl": "https://s3.amazonaws.com/bucket/media/uuid.jpg?X-Amz-...",
  "requiredHeaders": { "Content-Type": "image/jpeg", "Content-Length": "204800" },
  "expiresAt": "2026-06-04T09:05:00.000Z"
}

Typical two-step flow

# Step 1 — presign
RESULT=$(mcp call media_upload_presign '{"filename":"hero.jpg","mimeType":"image/jpeg","fileSize":204800}')
UPLOAD_URL=$(echo $RESULT | jq -r '.uploadUrl')
MEDIA_KEY=$(echo $RESULT | jq -r '.mediaKey')

# Step 2 — upload directly to S3
curl -X PUT "$UPLOAD_URL" \
  -H "Content-Type: image/jpeg" \
  --upload-file ./hero.jpg

# Step 3 — register
mcp call media_register "{\"mediaKey\":\"$MEDIA_KEY\",\"originalFilename\":\"hero.jpg\",\"mimeType\":\"image/jpeg\"}"

media_register#

Finalize a presigned-upload flow: HEAD the S3 object, verify the size cap, and insert a media row. Pairs with media_upload_presign.

• Auth: media:write

Response — the new media row.

media_delete#

Permanently delete a media asset from the library (does not remove the S3 object).

• Auth: media:write

Response — { "success": true, "id": 55 }

Navigation#

Nav changes use a draft overlay: nav_create, nav_update, and nav_delete all write to a draft JSON field and leave the live nav untouched until you call nav_publish or nav_publish_all.

nav_list#

List nav items for a website, sorted by sortOrder. Hierarchical via parentId.

• Auth: sites:read

Response — array of nav rows including draft overlay if pending.

nav_create#

Stage a new nav item (draft only — hidden from the live nav until nav_publish).

• Auth: sites:write

nav_update#

Stage changes to a nav item into its draft overlay. Live columns are untouched until nav_publish.

• Auth: sites:write

nav_delete#

Stage a tombstone on a nav item (draft.pendingDelete). The row and live nav are unchanged until nav_publish runs.

• Auth: sites:write

nav_publish#

Promote a single nav item's draft to live. If pendingDelete: row is removed. If pendingCreate: item becomes visible. Otherwise: draft fields are applied to live columns.

• Auth: sites:write

nav_publish_all#

Promote every nav row with a non-null draft on a website in one call. Same per-row semantics as nav_publish.

• Auth: sites:write

Response

{ "websiteId": 7, "count": 3, "items": [{ "id": 10, "published": true }, { "id": 11, "deleted": true }] }

Block Templates#

Block templates are reusable block trees you can insert into any post. They follow the same draft overlay pattern as navigation: create/update/delete stage into a draft jsonb field; block_templates_publish makes them live.

scope values:

• block — copy-on-insert; post gets its own detached copy.
• section — copy-on-insert at section level.
• global — posts embed a live reference; when you update and publish the template, all embedders sync automatically (version is bumped).

block_templates_list#

List reusable block templates. Returns global (platform-curated) + your client's own templates.

• Auth: sites:read

Response — slim projection (no blocks JSON blob). Call block_templates_get for the full tree.

block_templates_get#

Fetch a full template including its blocks JSON.

• Auth: sites:read

block_templates_create#

Create a new reusable block template (starts as a draft — hidden from the picker until published).

• Auth: sites:write

Response — new template row + approval URL.

block_templates_update#

Stage changes to a block template into its draft overlay. Live columns and version are untouched until block_templates_publish.

• Auth: sites:write

block_templates_delete#

Stage a tombstone on a block template. Blocked if any posts currently embed it as a global template — remove or convert those usages first.

• Auth: sites:write

block_templates_publish#

Promote a block template's draft to live. If pendingDelete: row is removed. If pendingCreate: template appears in picker. Otherwise: draft fields applied to live columns; version is bumped when blocks changed (triggers sync on global embedders).

• Auth: sites:write

block_templates_fork#

Duplicate a published block template into a new draft tied to the original via parent_template_id. Use when you want to build a variant without modifying the source. Returns the new template id + an approval URL.

• Auth: sites:write

Custom Domains#

website_domains_list#

List custom domains attached to a website.

• Auth: sites:read

Response

[{ "id": 3, "websiteId": 7, "domain": "acme.com", "isPrimary": true, "status": "pending" }]

website_domains_add#

Attach a custom domain to a website. Starts in pending status until DNS verification. Does not provision DNS records — you must configure them externally.

• Auth: sites:write

website_domains_remove#

Detach a custom domain. Does not affect external DNS.

• Auth: sites:write

Environment Variables#

website_env_vars_list#

List env vars for a website environment. Values are included — treat output as secrets.

• Auth: sites:read

Response — array of { id, key, value, syncedToVercel } rows, sorted by key.

website_env_vars_set#

Upsert an env var (creates or overwrites). Sets syncedToVercel: false — actual Vercel sync happens via the portal UI.

• Auth: sites:write

website_env_vars_delete#

Remove an env var by id.

• Auth: sites:write

Custom Post Types#

Custom post types (CPTs) extend the CMS beyond blog and page. You can define custom field schemas, a block-tree template that wraps every post, and per-type CSS/JS. Reads use sites:read; writes use sites:write. Only site-owned CPTs are editable; built-in/global types (managed by admins) are read-only.

post_types_list#

List CPTs available on a website (site-specific + global built-ins).

• Auth: sites:read

post_types_get#

Fetch a single CPT including its template and custom code.

• Auth: sites:read

post_types_create#

Create a new CPT scoped to a website. Slug must be unique within the site and must not collide with global types.

• Auth: sites:write

Response — the new post type row.

post_types_update#

Update the name, slug, description, icon, or active flag of a site-owned CPT.

• Auth: sites:write

post_types_delete#

Permanently delete a site-owned CPT. Posts of that type cascade to deletion.

• Auth: sites:write

post_types_get_template#

Get the block-tree template that wraps every post of this CPT. The template always contains exactly one { type: "post-content" } placeholder — at render time the post's own blocks are substituted in. Returns { template, defaulted: true } if no template has been saved yet.

• Auth: sites:read

post_types_update_template#

Replace the block-tree template for a CPT. The server enforces "exactly one post-content placeholder" — extras are dropped (first wins), and a placeholder is prepended if absent. Pass template: null to reset to the default starter.

• Auth: sites:write

post_types_get_code#

Get the type-wide custom CSS and JS that cascades to every post of this CPT.

• Auth: sites:read

Response — { "customCss": "...", "customJs": "..." }

post_types_update_code#

Update the type-wide custom CSS/JS. Pass an empty string to clear a field; omit to leave unchanged.

• Auth: sites:write

post_types_fields_list#

List custom field definitions for a CPT, ordered by order. Children of repeater/group fields have parentId set.

• Auth: sites:read

post_types_fields_create#

Add a custom field to a CPT.

• Auth: sites:write

post_types_fields_update#

Update a custom field. Reparenting (parentId) requires the new parent to be a repeater or group on the same CPT.

• Auth: sites:write

post_types_fields_delete#

Delete a custom field. Stored values cascade. For repeater/group parents, child fields cascade too.

• Auth: sites:write

Branding#

Branding profiles hold your color palette, fonts, logos, and brand voice copy. Reads use branding:read; writes use branding:write.

branding_list_profiles#

List all branding profiles for your account.

• Auth: branding:read
• Inputs: none

Response — array of profile rows (id, name, isDefault, colors, fonts, logo URLs).

branding_get_profile#

Fetch a full branding profile (colors, fonts, logos, button style). Omit profileId to get the default profile.

• Auth: branding:read

branding_get_messaging#

Fetch brand voice and copy context: tagline, value proposition, elevator pitch, tone, voice samples, differentiators.

• Auth: branding:read

Response

{
  "tagline": "Build better, ship faster.",
  "valueProposition": "...",
  "elevatorPitch": "...",
  "toneOfVoice": "confident, approachable",
  "keyDifferentiators": ["no-code", "AI-first"],
  "targetAudience": "SMB agencies"
}

branding_audit#

Run the rule-based consistency audit on a branding profile. Returns WCAG contrast issues, missing-field warnings, and structural problems.

• Auth: branding:read

branding_check_contrast#

Compute the WCAG contrast ratio between two CSS colors. Returns the ratio plus AA/AAA pass/fail.

• Auth: branding:read

Response

{ "ratio": 4.56, "aa": true, "aaa": false }

branding_create_profile#

Create a new branding profile.

• Auth: branding:write

branding_update_profile#

Update any combination of colors, fonts, logos, or the isDefault flag on an existing profile.

• Auth: branding:write

branding_delete_profile#

Permanently delete a branding profile. Sites that referenced it fall back to the client default.

• Auth: branding:write

branding_update_messaging#

Update brand voice / copy context (tagline, elevator pitch, value prop, tone, audience, differentiators). Creates the row if it does not exist yet. Pass profileId to scope messaging to a specific profile; omit for the client-level default.

• Auth: branding:write

Storefront#

Storefront tools are website-scoped — all reads require store:read, all writes require store:write.

Products#

store_products_list

List products for a website. Filter by status, category, featured flag, or search term.

• Auth: store:read

store_products_get

Fetch a product including all its images and variants.

• Auth: store:read

Response

{
  "product": { "id": 1, "name": "T-Shirt", "price": 2999, "status": "active" },
  "images": [{ "id": 10, "url": "...", "order": 0 }],
  "variants": [{ "id": 20, "name": "Large / Red", "price": 2999 }]
}

store_products_create

Create a new product. Price is in cents. Starts in draft status — use store_products_update to activate.

• Auth: store:write

store_products_update

Update any mutable field on a product. For images and variants, use their dedicated tools.

• Auth: store:write

store_products_delete

Permanently delete a product. Images and variants cascade. Order items retain historical data via productName/variantName.

• Auth: store:write

store_products_adjust_inventory

Adjust product quantity by a positive or negative delta. Use store_products_update to SET an absolute quantity.

• Auth: store:write

store_product_options_create

Add a product option axis (e.g. "Size", "Color"). Use store_product_option_values_create to add its values.

• Auth: store:write

store_product_option_values_create

Add a value (e.g. "Red", "Large") to an existing product option.

• Auth: store:write

store_product_variants_create

Create a product variant. optionValues ties the variant to specific option values (e.g. Size=Large, Color=Red). Price in cents.

• Auth: store:write

store_product_variants_update

Update any mutable field on a variant.

• Auth: store:write

Product Categories#

store_categories_list

List product categories for a website.

• Auth: store:read

store_categories_create

Create a product category. Supports parent/child hierarchy via parentId.

• Auth: store:write

Orders#

store_orders_list

List orders for a website. Filter by status, payment status, customer email, or date.

• Auth: store:read

store_orders_get

Fetch order detail: the order row, line items, and full status history.

• Auth: store:read

Response

{
  "order": { "id": 55, "orderNumber": "ORD-0055", "status": "processing", "total": 5998 },
  "items": [{ "id": 1, "productName": "T-Shirt", "quantity": 2, "unitPrice": 2999 }],
  "history": [{ "status": "pending", "createdAt": "..." }, { "status": "processing", "createdAt": "..." }]
}

store_orders_update_status

Transition an order through fulfillment states. Stamps shippedAt/deliveredAt automatically. Does not send customer notifications.

• Auth: store:write

store_analytics_get

Return order and revenue aggregates for a storefront over a time window: total revenue, order count, average order value, and order counts by status. The siteId must belong to the authenticated client.

• Auth: store:read

Response

{
  "totalRevenue": 124900,
  "totalOrders": 42,
  "averageOrderValue": 2974,
  "ordersByStatus": { "processing": 5, "shipped": 30, "delivered": 7 },
  "windowDays": 30
}

store_orders_add_note

Append or overwrite the internal staff-only note on an order. Does not affect customer-facing fields.

• Auth: store:write

Customers#

store_customers_list

List storefront customers. Filter by status or search by email/name.

• Auth: store:read

Response — slim projection: id, email, first/last name, phone, status, orderCount, totalSpent.

store_customers_get

Fetch a store customer (PII-included but no password hash). Returns up to 10 recent orders.

• Auth: store:read

Discounts#

store_discounts_list

List discount codes for a website.

• Auth: store:read

store_discounts_create

Create a discount code. percent amounts are in basis points (1000 = 10%); fixed_amount amounts are in cents.

• Auth: store:write

store_discounts_toggle

Flip the active flag on a discount code.

• Auth: store:write

store_discounts_delete

Permanently delete a discount code.

• Auth: store:write

Reviews#

store_reviews_list

List product reviews across a website. Filter by approval status and/or product.

• Auth: store:read

store_reviews_moderate

Approve or reject a product review.

• Auth: store:write

Customer Support Messages#

store_customer_messages_list

List customer support messages for a website.

• Auth: store:read

store_customer_messages_reply

Post a staff reply on a customer support thread. Does not email the customer.

• Auth: store:write

Store Settings#

store_settings_get

Get storefront configuration for a website (currency, tax, shipping, payout schedule, enabled flag).

• Auth: store:read

Common Errors#