Frontic Tools - Frontic Docs

The Frontic Tools MCP server connects AI coding assistants directly to your Frontic projects, letting you build storefronts and configure your Frontic API from your IDE. Pair it with the Frontic Docs MCP server so the agent can also answer “how does X work?” questions without leaving the editor.

Overview

Connect your favorite AI coding assistants directly to your Frontic Commerce projects through secure OAuth authentication. Build a frontend AND the corresponding Frontic API - all from your coding agent or IDE of choice.

Claude Code

Full project management and content creation workflows

Claude Desktop

Interactive Frontic operations from your desktop

Cursor

Integrated development experience with AI assistance

The MCP server provides tools across these categories:

Project & Resource Discovery

Browse projects, list each resource kind, inspect storage records, and preview Fetch API responses

Blocks, Listings, Pages & Trees

Create and manage blocks, listings, pages, and menu trees that define your API

Storage Management

Create Data Storages, then add, rename, and remove top-level fields

Project Configuration

Explore domains, scopes, regions, and locales

Browse Data

Search storage records and raw feed records

Integrations

Inspect connected integrations and their feed schemas before mapping

Data Syncs & Value Composer

Create data syncs, configure field mappings, and preview composer operations

Context Documents

Manage skills, rules, guides, and other context documents

Shopware Insights

Read the promotions and sales channels of a connected Shopware store

Setup

The fastest path is the CLI - pick your package manager:

frontic mcp init --client cursor

pnpm dlx @frontic/cli@latest mcp init --client cursor

npx @frontic/cli@latest mcp init --client cursor

yarn dlx @frontic/cli@latest mcp init --client cursor

bunx @frontic/cli@latest mcp init --client cursor

Swap cursor for claude, windsurf, vscode, or codex. The CLI prompts whether to install Frontic Tools, Frontic Docs, or both - the recommended default is both. See the Frontic CLI reference for full options. If you’d rather edit the config yourself, drop this entry into your client’s MCP config file:

{
  "mcpServers": {
    "frontic-tools": {
      "type": "http",
      "url": "https://mcp.frontic.com/mcp"
    }
  }
}

The first time you connect, you’ll be prompted to authenticate via OAuth. The token carries your Frontic account’s permissions - whatever role you have on a project, the MCP can do on that project, and nothing more.

Client-specific configuration

Claude Code

Add to .mcp.json at your project root - Claude Code reads project MCP config from this file:

{
  "mcpServers": {
    "frontic-tools": {
      "type": "http",
      "url": "https://mcp.frontic.com/mcp"
    }
  }
}

Claude Desktop

Edit the Claude Desktop config file:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "frontic-tools": {
      "type": "http",
      "url": "https://mcp.frontic.com/mcp"
    }
  }
}

Cursor

Add to .cursor/mcp.json in your project root, or Cursor’s global settings:

{
  "mcpServers": {
    "frontic-tools": {
      "type": "http",
      "url": "https://mcp.frontic.com/mcp"
    }
  }
}

Common Workflows

Most tools are a step in a chain, and the chain is what gets work done. These four cover almost everything people build through the MCP server.

Build a backend from zero

list_projects to find the project, then create_storage for the Data Storage and update_storage to add its fields. Wrap it in a Detail Block with create_block, expose a collection with create_listing, and route it with create_page. Finish with preview_fetch_response to see exactly what the frontend will receive.

Map a feed into a storage

list_integrations to find the connected system, then get_integration_feed for the flat field list in dot-path notation. Shape the storage to match, then create_data_sync - which creates the sync shell only. Call get_data_sync for the field tree, map the fields with update_data_sync_value_configs, and dry-run the result with preview_composer_record before anything is persisted. Verify with search_storage_records once the sync has run.

Build a navigation menu

list_storages for the source data and list_blocks to find an item block worth reusing - multiple Menu Trees can share one. create_tree builds the hierarchy, update_tree narrows it with query conditions.

Debug what the frontend receives

preview_fetch_response for the served payload, search_storage_records for the stored values behind it - filtered by scope, region, and locale to see contextualized output. When you need a storage’s real keys and slugs rather than a search, list_storage_record_examples returns up to five records with their system fields.

Example Prompts

Once connected, try these prompts to explore the MCP server’s capabilities:

Have your agent list your projects and inspect the main one

Ask the agent to map out your Data Storages and their fields

Let the agent build a Search Listing with filters and sorting

Tell the agent to scaffold a hero block with the standard fields

Have the agent assemble a homepage from existing primitives

Get the agent to plan and set up a basic storefront end to end

Have the agent extend a Data Storage schema with a new translatable field

Have the agent create a Products storage and connect a feed

Have the agent set up a content storage from scratch

Have the agent add a relation field linking two storages

Walk the integration → feed → data sync workflow end to end

Ask the agent to inspect your data syncs and their field mappings

Have the agent preview a Value Composer transformation on a sample record

Ask the agent what the connected Shopware store is running

Ask the agent to walk your scope, region, and domain hierarchy

Available Tools

What each tool may do

Every tool declares whether it only reads, as an MCP annotation. Clients that act on those annotations can run the reading tools without stopping to ask and keep prompting for the rest; for clients that want it spelled out, the allow-list below does the same thing explicitly.

May do	Tools
Reads only	every `get_`, `list_`, `search_`, and `preview_` tool
Writes	every `create_` tool, and the `update_` tools that change configuration
Destroys data	`delete_resource`, `update_storage`

The preview_* tools count as reads because all three are dry-runs - they show you the transformed value or the API response without persisting anything. update_storage sits with delete_resource because its field operations include deleting a field, and deleting a field takes the stored values with it. Adding and renaming fields is safe; removing one is not reversible.

Auto-approve the read-only tools in Claude Code

Add to .claude/settings.json in your project. Everything not listed still prompts:

{
  "permissions": {
    "allow": [
        "mcp__frontic-tools__get_context_document",
        "mcp__frontic-tools__get_data_sync",
        "mcp__frontic-tools__get_integration_feed",
        "mcp__frontic-tools__get_project",
        "mcp__frontic-tools__list_blocks",
        "mcp__frontic-tools__list_composer_operations",
        "mcp__frontic-tools__list_context_documents",
        "mcp__frontic-tools__list_data_syncs",
        "mcp__frontic-tools__list_domains",
        "mcp__frontic-tools__list_integrations",
        "mcp__frontic-tools__list_listings",
        "mcp__frontic-tools__list_locales",
        "mcp__frontic-tools__list_pages",
        "mcp__frontic-tools__list_projects",
        "mcp__frontic-tools__list_regions",
        "mcp__frontic-tools__list_scopes",
        "mcp__frontic-tools__list_shopware_promotions",
        "mcp__frontic-tools__list_shopware_sales_channels",
        "mcp__frontic-tools__list_storage_record_examples",
        "mcp__frontic-tools__list_storages",
        "mcp__frontic-tools__list_trees",
        "mcp__frontic-tools__preview_composer_record",
        "mcp__frontic-tools__preview_composer_value",
        "mcp__frontic-tools__preview_fetch_response",
        "mcp__frontic-tools__search_feed_records",
        "mcp__frontic-tools__search_storage_records"
    ]
  }
}

Project & Resource Discovery

`list_projects`

Browse all your Frontic projects with their names, IDs, data types, and configured locales. Use it when you don’t already know which project you’re working against - once a project is selected, the resource tools take its ID directly.

Show Parameters

enum

default:"content"

Output format: ‘content’ (default) returns project data directly in the response, ‘resource’ returns MCP resource URIs that can be fetched separately. Use ‘content’ in most cases. One of: resource, content

`preview_fetch_response`

Get a sample Fetch API response for a block, listing, tree, or page - exactly what the frontend receives when calling the Frontic API. Useful for understanding the response shape before building your frontend.

Show Parameters

enum

required

Resource type to get the example for. Can be “block”, “listing”, “tree” or “page”. If you pass the wrong type for a given name, the tool will auto-correct to match the resource’s actual mode. One of: block, listing, tree, page

string

The type name of the block, listing, or tree (e.g., ‘ProductCard’, ‘CategoryListing’, ‘CategoryNavigation’). Use list_blocks, list_listings or list_trees to find available names if unsure. Not needed for pages - use the slug parameter instead.

string

JSON object with entry point parameters. For blocks: {“key”: “<record-uuid>”} (omit to return any result). For listings: depends on configuration, e.g., {“categoryId”: “<id>”} - most listings require at least one parameter. For trees: optional {“key”: “<record-uuid>”} to start from a subtree, {“depth”: <n>} to limit depth (1-20). Use list_storage_record_examples to find real record IDs.

string

Only required for pages. The full URL of the page - e.g. “myshop.com/shoes/sneakers/nike-air-max-90”, “myshop.com/fr-FR/Women/Clothing”

string

required

Project ID in which the resource exists.

`delete_resource`

Permanently delete a block, listing, page, storage, or tree from the project. This action cannot be undone.

Show Parameters

string

required

The unique identifier of the project containing the resource to delete

enum

required

Entity type: “block” for content components, “listing” for listings, “page” for website pages, “storage” for data storages “tree” for trees One of: block, listing, page, storage, tree

string

required

The unique identifier of the specific resource to delete

Block Management

`list_blocks`

Single-record endpoints with their fields, parameters, and storageId

Show Parameters

string

required

Project ID in which the resource exists.

`create_block`

Create a new block - a reusable content component that defines which fields are exposed in Fetch API responses. Blocks serve single-record endpoints.

Show Parameters

string

required

The unique identifier of the project in which the block should be created

string

required

The name of the block. IMPORTANT: Always use normal capitalized words with spaces (e.g., ‘Product Card’, ‘Hero Banner’, ‘Cart Summary’). Never use PascalCase or concatenated words - the platform derives the URL slug and API type name from this value, and incorrect casing breaks namespacing.

string

required

A brief description of the block’s purpose (e.g., ‘Serves product detail data including name, price, images, and description’).

string

required

Must reference a valid storage ID from the available storages.

boolean

default:"false"

If the storage used in storageId is of type “commerce”, you can choose to retrieve product variants from this block. This will determine the fetchMode for this block. It will make the block return variant records. The default is to return product records.

`update_block`

Update an existing block by adding or removing fields, and control its active status.

Show Parameters

string

required

The unique identifier of the project containing the block

string

required

The unique identifier of the block to modify

object[]

Optional array of field operations. Each field can have “add” or “delete” operation, allowing mixed operations in a single call. Omit this parameter if you only want to change the active status.

Show Fields

enum

required

The operation for this field: “add” to add a new field, “delete” to remove an existing field One of: add, delete

string

required

Name of the block field. For “add” operation, this will be the new field name. For “delete” operation, this must match an existing field name.

enum

The type of field to add. Required for “add” operations. Options: “field” (storage-based), “block” (nested block), “resultList”, “variantList”, “productPrice” (only possible for blocks with fetchMode=product), “pageRoute” One of: field, block, resultList, variantList, productPrice, pageRoute

string

Required for fieldType=“field”. The ID of the field from the block’s storage to link to.

string

Required for fieldType=“block”, “resultList”, or “variantList”. The ID of the nested block to embed.

object[]

Optional parameter values for nested blocks. Maps parent block fields to nested block parameters.

Show Fields

string

required

The ID of the block field to pass as value

enum

The source type for the block parameter value One of: blockField, static, storageSchemaField

string

required

The static value or block field ID or block field name or storage field ID for the block parameter value source

string

The path in the block field or storage field for the block parameter value source, if it is a composite field

enum

Required for fieldType=“productPrice”. Strategy for selecting price from variants. One of: first, lowest, highest

string

Required for fieldType=“pageRoute”. The ID of the target page to link to.

string

Required for fieldType=“pageRoute”. The ID of the block field that provides the key value that is used to get the correct page value.

string

Required for fieldType=“pageRoute”. The ID of the storage schema field used as the page key.

string

Optional for fieldType=“pageRoute”. Path in the key field if it is a composite field (e.g., “name.en”).

boolean

Optional. Set whether the block is active (enabled) or inactive (disabled). When inactive, the block cannot be used in pages or other blocks.

Listing Management

`list_listings`

Collections with their search, filter, and sort fields, parameters, and query conditions

Show Parameters

string

required

Project ID in which the resource exists.

`create_listing`

Create a new listing - a searchable, filterable, sortable collection that performs search queries on a block index.

Show Parameters

string

required

The unique identifier of the project in which the listing should be created

string

required

The name of the listing. IMPORTANT: Always use normal capitalized words with spaces (e.g., ‘Product List’, ‘Category Grid’, ‘Search Results’). Never use PascalCase or concatenated words - the platform derives the URL slug and API type name from this value.

string

required

A brief description of the listing’s purpose (e.g., ‘Searchable product collection with category filtering and price sorting’).

string

required

Must reference a valid storage ID from the available storages.

enum

required

Choose if the listing should return only the first result or all results. Use “first” for single item searches, “all” for listing multiple items (e.g. product listings in categories). One of: all, first

string

required

Must reference a valid block ID from the available blocks. This block will be used to render each item in the listing.

`update_listing`

Configure all aspects of an existing listing: search fields, filter fields, sort fields, parameters, query conditions, search settings, and pagination.

Show Parameters

string

required

The unique identifier of the project containing the listing

string

required

The unique identifier of the listing to modify

object[]

Optional array of field operations for search/filter/sort fields. Each field can have “add”, “update”, or “delete” operation.

Show Fields

enum

required

The operation for this field: “add” to add a new field, “update” to modify an existing field, “delete” to remove an existing field One of: add, update, delete

enum

required

The type of field: “search” for text search, “filter” for filtering/faceting, “sort” for sorting One of: search, filter, sort

string

required

Name of the field. For “add” operation, this will be the new field name. For “update” and “delete” operations, this must match an existing field name.

enum

Required for “add” operations. Source of the field: “blockField” (from listing’s nested block) or “storageField” (from storage referenced in storageId). Always prefer block fields over storage fields. One of: blockField, storageField

string

Optional path of the field, if it is nested/composite (e.g. if the field you want to use is vendor.label, then the path is “label”)

boolean

Required for type=“filter” add operations. Whether the filter field is facetted (shows available options)

boolean

Optional for type=“filter” add operations. Whether to aggregate values for this filter field. Defaults to true if not specified.

object[]

Optional for type=“filter”. Numeric ranges for the filter field

Show Fields

number

required

Start of the range

number

required

End of the range

enum

Required for type=“sort” add operations. The sorting direction: “asc” (ascending) or “desc” (descending) One of: asc, desc

boolean

Required for type=“sort” add operations. Indicates if this is the default sort field. Only one sort field can be default.

object[]

Optional array of parameter operations. Parameters can be referenced in query conditions.

Show Fields

enum

required

The operation for this parameter: “add”, “update”, or “delete” One of: add, update, delete

string

required

Name of the parameter. For “update” and “delete” operations, this must match an existing parameter name.

string

Required for “add” operations. The data type of the parameter (e.g., “string”, “int”, “bool”)

boolean

Required for “add” operations. Whether the parameter is required

boolean

Required for “add” operations. Whether the parameter is an array

number

Optional minimum number of items (for array parameters)

number

Optional maximum number of items (for array parameters)

string

Optional default value for the parameter

object[]

Optional array of query condition operations. Conditions filter which records appear in the listing.

Show Fields

enum

required

The operation for this condition: “add” to add a new condition, “delete” to remove by index One of: add, delete

number

Required for “delete” operation. The index of the condition to delete (0-based)

object

Required for simple comparison conditions. The left-hand side of the condition (field to check). Omit for pure group conditions.

Show Fields

enum

required

Source type: “blockField” or “storageField” (will be normalized to storageSchemaField) One of: blockField, storageField

string

required

UUID of the field to compare

string

Optional path if the field is composite (e.g., “name.en”)

enum

Required for simple comparison conditions. The comparison operator. Omit for pure group conditions. One of: like, notLike, equals, notEquals, gt, gte, lt, lte

object

Required for simple comparison conditions. The right-hand side of the condition (value to compare against). Omit for pure group conditions.

Show Fields

enum

required

Value type: “blockParameter” (reference to parameter) or “value” (static value) One of: blockParameter, value

string

required

For blockParameter: parameter name. For value: the static value

any[]

Array of nested conditions for complex AND/OR logic. Required for pure group conditions. Each rule follows the same structure and can itself be either a simple condition or a group.

enum

Required when rules are present. Logical operator (“and” or “or”) for combining nested rules. One of: and, or

boolean

Optional. Enable fuzzy search for better matching of similar terms. Defaults to true if not specified.

number

Optional. Default number of items per page for pagination. Must be at least 1. Defaults to 24 if not specified.

number

Optional. Maximum number of items per page that can be requested. Must be at least 1. Defaults to 48 if not specified.

boolean

Optional. Set whether the listing is active (enabled) or inactive (disabled). When inactive, the listing cannot be used in the storefront.

Page Management

`list_pages`

URL routing patterns with their block and that block’s storageId

Show Parameters

string

required

Project ID in which the resource exists.

`create_page`

Create a page - a URL routing pattern that generates SEO-friendly routes from block data records.

Show Parameters

string

required

The unique identifier of the project in which the page should be created

string

required

The name of the page (max 100 characters). IMPORTANT: Always use normal capitalized words with spaces (e.g., ‘Product Detail’, ‘Category Page’). Never use PascalCase or concatenated words.

string

required

Must reference a valid block ID from the available blocks. This must be a block, not a listing.

string

required

The ID of the block field to use as the URL slug. This field will be used to generate the URL path for each page.

string

The path in the slug block field referenced in slugBlockFieldId, if it is a composite field (e.g., “label” if the slugBlockFieldId references the composite field “cover” and the full path would be “cover.label”). Leave empty for simple fields.

enum

Strategy for handling URL conflicts: “suffix” adds the conflict suffix to make URLs unique, “discard” skips records with conflicting URLs. Defaults to “suffix”. One of: suffix, discard

string

Only required for urlConflictStrategy=“suffix”. The ID of the block field to use as suffix when URL conflicts occur (e.g., when multiple items would generate the same URL slug). It often makes sense to use a unique identifier for the item, like a unique id or product-number etc.

string

Only relevant for urlConflictStrategy=“suffix”. The path in the URL conflict suffix block field referenced in urlConflictSuffixBlockFieldId, if it is a composite field (e.g., “label” if the slugBlockFieldId references the composite field “cover” and the full path would be “cover.label”). Leave empty for simple fields.

enum

Behavior when a source record is deleted: “temporaryRedirect” (302), “permanentRedirect” (301), “notFound” (404), or “delete” (removes the page URL completely). Defaults to “notFound”. One of: temporaryRedirect, permanentRedirect, notFound, delete

boolean

Indicates if the URLs of this page should be included in generated sitemaps. Defaults to false.

`update_page`

Update an existing page’s URL generation settings like slug source, conflict strategy, and delete behavior.

Show Parameters

string

required

The unique identifier of the project containing the page

string

required

The unique identifier of the page to update

string

Optional: The ID of the block field to use as the URL slug. Only provide if you want to change it.

string

Optional: The path in the slug block field, if it is a composite field (e.g., “label” for “cover.label”). Use empty string "" for simple fields. Only provide if you want to change it.

string

Optional: The ID of the block field to use as suffix when URL conflicts occur. Required if you set urlConflictStrategy to “suffix”. It often makes sense to use a unique identifier like a product number.

string

Optional: The path in the URL conflict suffix block field, if it is a composite field. Use empty string "" for simple fields. Only provide if you want to change it.

enum

Optional: Strategy for handling URL conflicts: “suffix” adds the conflict suffix to make URLs unique, “discard” skips records with conflicting URLs. Set to null to clear the value. If set to “suffix”, you must also provide urlConflictSuffixBlockFieldId.

enum

Optional: Behavior when a source record is deleted: “temporaryRedirect” (302), “permanentRedirect” (301), “notFound” (404), or “delete” (removes the page URL completely). Set to null to clear the value.

boolean

Indicates if the URLs of this page should be included in generated sitemaps..

Tree Management

`list_trees`

Hierarchical collections with their storageId, query conditions, and nested item block

Show Parameters

string

required

Project ID in which the resource exists.

`create_tree`

Create a Menu Tree - a hierarchical collection of records (navigation menus, category trees, support article hierarchies). Each tree references a connected Data Storage and a Detail Block that defines the node shape. Multiple trees can share the same block; the agent prefers reusing an existing block over creating a new one.

Show Parameters

string

required

The unique identifier of the project in which the tree should be created

string

required

The name of the tree. IMPORTANT: Always use normal capitalized words with spaces (e.g., ‘Category Menu’, ‘Footer Navigation’, ‘Support Links’). Never use PascalCase or concatenated words - the platform derives the URL slug and API type name from this value.

string

required

A brief description of the tree’s purpose (e.g., ‘Main category navigation for headers’).

string

required

Must reference a valid storage ID from the available storages.

string

required

Must reference a valid block ID from the available blocks. This block will be used to render each item in the tree.

`update_tree`

Configure an existing tree’s query conditions - the business-logic filters that narrow which records the tree includes. Use it after create_tree to make a tree more specific.

Show Parameters

string

required

The unique identifier of the project containing the tree

string

required

The unique identifier of the tree to modify

object[]

Optional array of query condition operations. Conditions filter which records appear in the tree.

Show Fields

enum

required

The operation for this condition: “add” to add a new condition, “delete” to remove by index One of: add, delete

number

Required for “delete” operation. The index of the condition to delete (0-based)

object

Required for simple comparison conditions. The left-hand side of the condition (field to check). Omit for pure group conditions.

Show Fields

enum

required

Source type: “storageSchemaField” One of: storageSchemaField

string

required

UUID of the field to compare

string

Optional path if the field is composite (e.g., “name.en”)

enum

Required for simple comparison conditions. The comparison operator. Omit for pure group conditions. One of: like, notLike, equals, notEquals, gt, gte, lt, lte

object

Required for simple comparison conditions. The right-hand side of the condition (value to compare against). Omit for pure group conditions.

Show Fields

enum

required

Value type: “value” (static value) One of: value

string

required

the static value

any[]

Array of nested conditions for complex AND/OR logic. Required for pure group conditions. Each rule follows the same structure and can itself be either a simple condition or a group.

enum

Required when rules are present. Logical operator (“and” or “or”) for combining nested rules. One of: and, or

boolean

Optional. Set whether the tree is active (enabled) or inactive (disabled). When inactive, the tree cannot be used in the storefront.

Storage Management

`list_storages`

Data schemas with their fields

Show Parameters

string

required

Project ID in which the resource exists.

`create_storage`

Create a new Data Storage with a name and schema type.

default creates a regular Content storage for content or reference data.
commerce creates a Products storage for product catalog data and adds commerce-specific system fields for variants, prices, parent keys, and options.

After creating the storage, use update_storage to add custom fields.

Show Parameters

string

required

The unique identifier of the project in which the storage should be created

string

required

The name of the storage. IMPORTANT: Always use normal capitalized words with spaces (e.g., ‘Products’, ‘Blog Articles’, ‘Brands’). Never use PascalCase or concatenated words.

enum

required

Storage type. Use “default” for regular content/reference data and “commerce” for product catalog data with commerce-specific default fields. One of: default, commerce

`update_storage`

Update an existing Data Storage by adding, renaming, or removing top-level fields. Removing a field deletes the values stored in it, which is why this tool is marked as destroying data. Field type is fixed at creation - to change a field’s type you delete and recreate it. Multiple operations can be batched in a single call and run sequentially, so a later operation can target a field added or renamed earlier in the same call. The tool works on top-level fields only - nested sub-fields of composite types aren’t editable through this tool. To create a new storage from scratch through MCP, use create_storage.

Show Parameters

string

required

Project ID in which the resource exists.

string

required

Storage ID

object[]

required

Ordered list of field operations. Operations run sequentially, so later operations can target fields created or renamed earlier in the same call.

Show Fields

enum

required

Field operation: “add” creates a new top-level storage field, “update” modifies an existing field, “delete” removes one. One of: add, update, delete

string

Existing storage field ID to update or delete. Prefer this when available from list_storages.

string

Existing storage field name to update or delete when fieldId is unknown. Only top-level field names are supported.

string

New field name for “add”, or the replacement name when renaming an existing field during “update”.

string

Data type key for “add” operations, e.g. “string”, “integer”, “boolean”, or a custom composite type key. If relation is provided and type is omitted, “any” is used automatically.

string

Optional field description for add or update operations.

boolean

Whether the field stores multiple values. For add operations, defaults to false.

boolean

Whether the field is shared across variants. For add operations, defaults to false.

boolean

Whether the field is translatable. For add operations, defaults to false.

boolean

Whether the field is scope-aware. For add operations, defaults to false.

object

Optional relation configuration. The source field must already exist in the current storage.

Show Fields

string

required

Source field name in the current storage used as the relation key

string

Optional path in the source field when it is composite, e.g. “code”.

string

required

Target storage ID

string

required

Field name in the target storage to match against

string

Optional path in the target field when it is composite, e.g. “slug”.

string

Optional mock generator module, e.g. “faker.datatype.string”.

string

Optional mock generator parameters string.

integer

Optional minimum array size for generated mock data.

integer

Optional maximum array size for generated mock data.

Project Configuration

Explore the context hierarchy of your project. get_project returns the whole configuration in one call: the project, its scopes with their regions and locales, and its domains with each domain’s scope, region, and locale resolved into names rather than left as IDs. The list_* tools enumerate one resource kind at a time.

Domains map a scope, region, and locale combination to URL routing.
Scopes are data segmentation layers (e.g., B2B vs B2C). Every project has at least the default public scope. A scoped field can return different values per scope - for example, true in scope A and false in scope B.
Regions are subdivisions of a scope (e.g., EU, US, APAC). Each region holds a currency and one or more locales, determining which prices to show.
Locales determine which translation to return for translatable fields. If a field is set as translatable, the response value matches the translation for that locale. list_locales is the complete list: a locale that no region exposes appears there and not in get_project.

`get_project`

How one project is configured, in a single call: the project itself, its scopes with their regions and locales, and its domains with each domain’s scope, region, and locale resolved into names and keys rather than left as IDs. Start here when you already know which project you are working on, and reach for list_projects only when you do not. fetchApi carries the hostname and the version token that addresses this environment. Send the token as the fs-version header on Fetch API calls: without it the Fetch API serves the public environment. environments lists the project’s other stages, each with its own version token, and a stage only carries a token when one is set - the public environment has none, because sending no fs-version header is how a caller reaches it. list_locales remains the complete locale list. A locale that no region exposes appears there and not here.

Show Parameters

string

required

Project ID in which the resource exists.

`list_domains`

List all domains configured in a project

Show Parameters

string

required

Project ID in which the resource exists.

`list_scopes`

List all scopes in a project

Show Parameters

string

required

Project ID in which the resource exists.

`list_regions`

List all regions within a scope

Show Parameters

string

required

Project ID in which the resource exists.

string

required

Scope ID

`list_locales`

List all configured locales in a project

Show Parameters

string

required

Project ID in which the resource exists.

Browse Data

Search and inspect the actual data in your project - both transformed storage records and raw feed records from integrations.

`list_storage_record_examples`

List up to five example records from a storage, each carrying the system fields the record has (key, active, parentKey, position, plus price and options on Products storages). Use it to learn a storage’s real shape - slugs, keys, field values - before writing query conditions against it.

Show Parameters

string

required

Project ID in which the resource exists.

string

required

ID of the storage to read - usually the one connected to the block, listing or page. list_blocks, list_listings, list_trees and list_pages show each resource’s storageId.

string[]

Storage fields to return next to the system fields, e.g. [“slug”] for a page or [“categoryId”] for a listing parameter. Only records with a value in every one of them are returned. System fields come with every example.

boolean

Commerce storages only: true returns variant records instead of parent products. Default false.

`search_storage_records`

Search records in a storage to see the transformed values blocks and listings serve to the frontend. Useful for verifying data after a sync, debugging unexpected output, or pulling a realistic sample for a composer dry-run with preview_composer_value or preview_composer_record. Filter by scope, region, and locale to see contextualized values like scoped pricing or translated content. By default records[] contains bare payload objects - exactly what the user stored, no wrapper. Pass includeMetadata: true to get each record wrapped with its id, sourceId, parentSourceId, createdAt, and updatedAt. Opt in when you need to know when a record was updated/created or walk parent–child relationships in a follow-up call. metadata.parentSourceId is the storage-tree parent - distinct from any parentKey a user happens to have inside their own payload. That namespacing is why metadata is opt-in rather than always-on. availableFilters and metaSearchResult (pagination + total count) are always returned alongside records[], regardless of includeMetadata. Omit conditions entirely to return all records.

Show Parameters

string

required

Project ID in which the resource exists.

string

required

Storage ID

object

required

Show Fields

boolean

Search product variants instead of parent products. Only applicable to commerce-type storages.

object

Filter conditions. Pass an and or or array of {field, operator, value} items - NOT a field-keyed object. Example: {“and”: [{“field”: “slug”, “operator”: “like”, “value”: “polizei”}]}. The field-keyed shape {“slug”: {“like”: ”…”}} is silently ignored by the backend and returns all records.

Show Fields

object[]

All items must match (logical AND).

Show Fields

string

required

Field name to filter on, as it appears in availableFilters (e.g. “slug”, “name”, “parentKey”).

enum

required

Comparison operator. “like”/“notLike” accept raw substrings (no SQL wildcards - the backend wraps them). “empty” matches a missing, null or empty-string value, “notEmpty” the opposite. One of: equals, notEquals, gt, lt, gte, lte, like, notLike, empty, notEmpty

string | number | boolean | string[]

Value to compare against. Strings, numbers, booleans, or string arrays (for IN-style matches). Not needed for “empty” and “notEmpty”.

object[]

Any item must match (logical OR).

Show Fields

string

required

Field name to filter on, as it appears in availableFilters (e.g. “slug”, “name”, “parentKey”).

enum

required

string | number | boolean | string[]

Value to compare against. Strings, numbers, booleans, or string arrays (for IN-style matches). Not needed for “empty” and “notEmpty”.

object[]

Sort order for results as array of {field, direction: ‘ASC’|‘DESC’}.

Show Fields

string

required

enum

required

One of: ASC, DESC

integer

Page number for pagination (1-based). Defaults to 1.

integer

Maximum records to return per page. Defaults to server limit.

string

Scope ID for scope-contextualized field values (e.g., B2B pricing). Falls back to default scope if omitted.

string

Region ID within the scope for region-specific values. Falls back to default region if omitted.

string

Locale ID for translated field values (e.g., German product descriptions). Falls back to default locale if omitted.

boolean

default:"false"

Include storage metadata (id, sourceId, parentSourceId, createdAt, updatedAt) on each record. Default false returns the bare payload - what the user actually stored - under records[]. Opt in when you need to reference a record by its sourceId, parentSourceId, or id, e.g. for follow-up tool calls or debugging tree relationships.

`search_feed_records`

Search feed records from an integration to see the connector-normalized data before the composer transforms it. Every commerce platform names fields differently (Shopify: flat keys; Shopware: dot-notation; commercetools: deeply nested objects), so inspecting an actual feed record is the only reliable way to know what’s available and how it’s shaped. Same metadata contract as search_storage_records: bare payloads by default, full envelope (id, sourceId, parentSourceId, createdAt, updatedAt) when you pass includeMetadata: true. availableFilters and metaSearchResult are always returned.

Show Parameters

string

required

Feed ID

string

required

Integration ID

string

Integration channel ID. Omit to use the integration’s default channel. Read the ids from list_integrations.

string

Integration translation ID, which belongs to a channel. Omit to use the chosen channel’s first translation.

string

Filter by a specific record source ID to find a single record. Finds a parent record unless variants is set. Use this for targeted lookups when you know the record identifier.

string

Filter by parent record source ID to find its variants (e.g., all variants of a main product). Only applicable to feeds with parent-child relationships. Implies a variant search.

boolean

Commerce feeds only: true makes sourceId look up a variant record instead of a parent. Not needed when filtering by parentSourceId, which always searches variants.

boolean

default:"false"

Integrations

Discover and inspect connected source systems and their feed schemas. The typical workflow is list_integrations → pick a feed → get_integration_feed → use the field paths in update_data_sync_value_configs or composer previews. Integrations belong to the Team rather than to a single project, so both tools here take no project and return the same results wherever you are working. The same holds for search_feed_records in Browse Data, which reads a feed rather than a storage.

`list_integrations`

Connected source systems with their channels and a compact feed inventory (id and name). Integrations belong to the Team, so this takes no project and returns the same list whichever project you are working in. Takes no parameters.

`get_integration_feed`

Get the full schema for a single integration feed by feed ID. Returns the integration metadata, feed metadata (including type, token, urls, methods, configurations, and refreshSupported), and a flat field list in dot-path notation with types - for example, title (string), manufacturer (object), manufacturer.name (string). The flat representation is what you copy into composer configs. If the feed ID isn’t found, the tool returns the list of available feed IDs across the Team’s integrations so the agent can recover without a separate discovery call.

Show Parameters

string

required

Integration feed ID

Data Syncs

`list_data_syncs`

List all data sync configurations, showing which integrations feed into which storages

Show Parameters

string

required

Project ID in which the resource exists.

`create_data_sync`

Create a data sync shell that connects one integration feed to one storage with a name, feed, storage, and optional scope/channel/locale mappings. New data syncs are enabled on creation; call get_data_sync, then update_data_sync_value_configs, to configure field value mappings

Show Parameters

string

required

Project ID in which the resource exists.

string

required

The name of the data sync. Always use normal capitalized words with spaces (e.g., ‘Sync Products’, ‘Sync Categories’).

string

required

Integration feed ID

string

required

Storage ID

object[]

Optional scope/channel/locale mappings. This is not the storage field value-composer configuration.

Show Fields

string

required

Project scope ID

string

required

Integration channel ID

object[]

Locale to integration translation mappings for this scope and channel. Leave empty when the scope has no locale mappings.

Show Fields

string

required

Project locale ID

string

required

Integration translation ID

`get_data_sync`

Get a single data sync with full field-level detail including value-composer configurations

Show Parameters

string

required

Project ID in which the resource exists.

string

required

Data sync ID

`update_data_sync_value_configs`

Update data sync value composer field configurations by field ID

Show Parameters

string

required

Project ID in which the resource exists.

string

required

Data sync ID

object[]

required

List of field-level config replacements. Each entry replaces one target field config by fieldId, without requiring the caller to pass all data sync fields.

Show Fields

string

required

Storage field UUID from get_data_sync.storage.fields[].id (supports nested composite fields).

object

required

Full replacement config for this field only. Valid examples: {“source”:“data-field”,“field”:“name”}, {“source”:“static”,“value”:“Default”}, {“source”:“composed”,“composition”:{“type”:“object”,“properties”:{“label”:{“source”:“data-field”,“field”:“manufacturer.name”}}}}.

Value Composer

`list_composer_operations`

List all available value-composer operations with type, input/output shapes, and documentation

Show Parameters

string

required

Project ID in which the resource exists.

`preview_composer_value`

Preview a computed value by running a single value-composer config against a sample record (dry-run)

Show Parameters

string

required

Project ID in which the resource exists.

object

required

Value-composer config object. Common valid examples: {“source”:“data-field”,“field”:“name”}, {“source”:“static”,“value”:“Default”}, {“source”:“composed”,“composition”:{“type”:“object”,“properties”:{“label”:{“source”:“data-field”,“field”:“manufacturer.name”}}}}. Use list_composer_operations to discover operation-specific composition schemas.

object

required

Sample input record payload used during the dry run. You can source realistic examples from search_feed_records or search_storage_records.

boolean

default:"false"

Include intermediate transformation steps in the response for troubleshooting composition chains. Useful when a multi-step composition produces unexpected results and you need to see which operation went wrong.

`preview_composer_record`

Preview a full transformed record by running all storage-field mapping configs against a sample source record

Show Parameters

string

required

Project ID in which the resource exists.

string

required

Storage ID

object

required

Object mapping storage field names/keys to value-composer configs, e.g. {“title”:{“source”:“data-field”,“field”:“name”},“seoTitle”:{“source”:“static”,“value”:“Fallback”},“manufacturer”:{“source”:“composed”,“composition”:{“type”:“object”,“properties”:{“label”:{“source”:“data-field”,“field”:“manufacturer.name”}}}}}.

object

required

Sample source record payload for the dry run. Prefer realistic records from search_feed_records or search_storage_records.

boolean

default:"false"

Context Documents ^Plus

Context document tools are part of Frontic Studio, available on the Plus plan.

Manage skills, rules, commands, guides, and other context documents that provide instructions and knowledge to AI agents working with your project.

`list_context_documents`

List all active documents in a collection (skills, rules, commands, guides)

Show Parameters

string

required

Project ID in which the resource exists.

enum

required

Document collection One of: skills, rules, commands, guides

`get_context_document`

Fetch a single document by key, returning full markdown with YAML frontmatter

Show Parameters

string

required

Document key in format “collection/slug”

string

required

Project ID in which the resource exists.

`create_context_document`

Creates the document as a draft unless activate is true, which publishes it in the same call.

Show Parameters

string

required

Project ID in which the resource exists.

enum

required

Document collection One of: skills, rules, commands, guides

string

required

Document title

string

required

Document description

string

Initial document content in markdown format

boolean

Set to true to make this document active immediately. When false or omitted, the document starts as a draft.

`update_context_document`

Update a document’s metadata and/or content

Show Parameters

string

required

Document key in format “collection/slug”

string

required

Project ID in which the resource exists.

string

New document title

string

New document description

string

New document content in markdown format

enum

Set the document status. “active” makes it available to agents, “draft” hides it, “inactive” disables it. Omit to keep the current status unchanged. One of: active, draft, inactive

Shopware Insights

Read the store behind a project that has a Shopware integration configured. Both tools are read-only.

`list_shopware_promotions`

List active promotions with their discounts and validity window (validFrom/validUntil, ISO-8601, null = unbounded)

Show Parameters

string

required

Project ID in which the resource exists.

`list_shopware_sales_channels`

List sales channels with their configuration, theme configuration, languages, and currencies

Show Parameters

string

required

Project ID in which the resource exists.

​Overview

Claude Code

Claude Desktop

Cursor

Project & Resource Discovery

Blocks, Listings, Pages & Trees

Storage Management

Project Configuration

Browse Data

Integrations

Data Syncs & Value Composer

Context Documents

Shopware Insights

​Setup

​Client-specific configuration

​Common Workflows

​Example Prompts

​Available Tools

​What each tool may do

​Project & Resource Discovery

​list_projects

​preview_fetch_response

​delete_resource

​Block Management

​list_blocks

​create_block

​update_block

​Listing Management

​list_listings

​create_listing

​update_listing

​Page Management

​list_pages

​create_page

​update_page

​Tree Management

​list_trees

​create_tree

​update_tree

​Storage Management

​list_storages

​create_storage

​update_storage

​Project Configuration

​get_project

​list_domains

​list_scopes

​list_regions

​list_locales

​Browse Data

​list_storage_record_examples

​search_storage_records

​search_feed_records

​Integrations

​list_integrations

​get_integration_feed

​Data Syncs

​list_data_syncs

​create_data_sync

​get_data_sync

​update_data_sync_value_configs

​Value Composer

​list_composer_operations

​preview_composer_value

​preview_composer_record

​Context Documents Plus

​list_context_documents

​get_context_document

​create_context_document

​update_context_document

​Shopware Insights

​list_shopware_promotions

​list_shopware_sales_channels

Overview

Setup

Client-specific configuration

Common Workflows

Example Prompts

Available Tools

What each tool may do

Project & Resource Discovery

`list_projects`

`preview_fetch_response`

`delete_resource`

Block Management

`list_blocks`

`create_block`

`update_block`

Listing Management

`list_listings`

`create_listing`

`update_listing`

Page Management

`list_pages`

`create_page`

`update_page`

Tree Management

`list_trees`

`create_tree`

`update_tree`

Storage Management

`list_storages`

`create_storage`

`update_storage`

Project Configuration

`get_project`

`list_domains`

`list_scopes`

`list_regions`

`list_locales`

Browse Data

`list_storage_record_examples`

`search_storage_records`

`search_feed_records`

Integrations

`list_integrations`

`get_integration_feed`

Data Syncs

`list_data_syncs`

`create_data_sync`

`get_data_sync`

`update_data_sync_value_configs`

Value Composer

`list_composer_operations`

`preview_composer_value`

`preview_composer_record`

Context Documents ^Plus

`list_context_documents`

`get_context_document`

`create_context_document`

`update_context_document`

Shopware Insights

`list_shopware_promotions`

`list_shopware_sales_channels`