Skip to main content
Glama

create_metaobject

Create a new metaobject instance by specifying type and field values. Use to store custom data objects linked to other entities via metafield references.

Instructions

Create a new metaobject (instance) of an existing type. The type must match a registered metaobject definition — call list_metaobject_definitions first if you're unsure. fields is an array of {key, value} pairs; values are always strings (JSON/reference fields take a JSON-encoded string, primitives take literal text). handle is optional; Shopify generates one from the displayName field if present. status only applies to types that have the publishable capability — passing it for non-publishable types is silently ignored. Returns the new metaobject's GID for use in subsequent set_metafield calls (e.g. linking the metaobject to a product via a metaobject_reference metafield).

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
typeYesType handle from a registered metaobject definition. The definition must already exist; this tool does not create new types/schemas.
handleNoOptional URL-safe handle. If the type has a 'displayName' field, Shopify generates a handle from it; otherwise pass one here.
fieldsYesField values. Provide at least the required fields from the type's definition. Required fields without values cause a validation error.
statusNoPublish status. Only applies to types that declared the `publishable` capability — passing this for non-publishable types is silently ignored. ACTIVE = visible on storefront, DRAFT = hidden.

Implementation Reference

  • The handler function for 'create_metaobject' tool. It builds the metaobject payload from args (type, fields, optional handle, optional publishable status), calls the Shopify GraphQL mutation METAOBJECT_CREATE_MUTATION, checks for user errors, and returns the created metaobject's ID, name, handle, and type.
      server.tool(
      "create_metaobject",
      "Create a new metaobject (instance) of an existing type. The `type` must match a registered metaobject definition — call list_metaobject_definitions first if you're unsure. `fields` is an array of {key, value} pairs; values are always strings (JSON/reference fields take a JSON-encoded string, primitives take literal text). `handle` is optional; Shopify generates one from the displayName field if present. `status` only applies to types that have the `publishable` capability — passing it for non-publishable types is silently ignored. Returns the new metaobject's GID for use in subsequent set_metafield calls (e.g. linking the metaobject to a product via a metaobject_reference metafield).",
      createMetaobjectSchema,
      async (args) => {
        const metaobject: Record<string, unknown> = {
          type: args.type,
          fields: args.fields,
        };
        if (args.handle) metaobject.handle = args.handle;
        if (args.status) {
          metaobject.capabilities = {
            publishable: { status: args.status },
          };
        }
    
        const data = await client.graphql<{
          metaobjectCreate: {
            metaobject: MetaobjectNode | null;
            userErrors: ShopifyUserError[];
          };
        }>(METAOBJECT_CREATE_MUTATION, { metaobject });
        throwIfUserErrors(data.metaobjectCreate.userErrors, "metaobjectCreate");
        const m = data.metaobjectCreate.metaobject;
        if (!m) {
          return {
            content: [
              { type: "text" as const, text: "metaobjectCreate returned no metaobject." },
            ],
          };
        }
        return {
          content: [
            {
              type: "text" as const,
              text: `Created metaobject ${m.displayName ?? m.handle} (${m.type}) — ${m.id}`,
            },
          ],
        };
      },
    );
  • The Zod-based input schema for 'create_metaobject', defining `type` (string), optional `handle` (string), `fields` (array of {key, value}), and optional `status` (ACTIVE/DRAFT enum).
    const createMetaobjectSchema = {
      type: z
        .string()
        .describe(
          "Type handle from a registered metaobject definition. The definition must already exist; this tool does not create new types/schemas.",
        ),
      handle: z
        .string()
        .optional()
        .describe(
          "Optional URL-safe handle. If the type has a 'displayName' field, Shopify generates a handle from it; otherwise pass one here.",
        ),
      fields: z
        .array(fieldInputSchema)
        .min(1)
        .describe(
          "Field values. Provide at least the required fields from the type's definition. Required fields without values cause a validation error.",
        ),
      status: z
        .enum(["ACTIVE", "DRAFT"])
        .optional()
        .describe(
          "Publish status. Only applies to types that declared the `publishable` capability — passing this for non-publishable types is silently ignored. ACTIVE = visible on storefront, DRAFT = hidden.",
        ),
    };
  • The `registerMetaobjectTools` function that registers all metaobject tools on the MCP server. The 'create_metaobject' tool is registered on line 380 within this function.
    export function registerMetaobjectTools(
      server: McpServer,
      client: ShopifyClient,
    ): void {
      server.tool(
        "list_metaobject_definitions",
        "List the metaobject definitions (custom types/schemas) registered on this Shopify store, with their field definitions. Each definition declares a `type` handle, a set of typed fields, and which fields are required. Use this tool to discover what custom data shapes the store supports before calling list_metaobjects (which queries instances of one type) or create_metaobject (which creates a new instance). Cursor-paginated.",
        listDefinitionsSchema,
        async (args) => {
          const data = await client.graphql<{
            metaobjectDefinitions: Connection<MetaobjectDefinitionNode>;
          }>(LIST_METAOBJECT_DEFINITIONS_QUERY, {
            first: args.first,
            after: args.after,
          });
          const edges = data.metaobjectDefinitions.edges;
          if (edges.length === 0) {
            return {
              content: [
                { type: "text" as const, text: "No metaobject definitions on this store." },
              ],
            };
          }
          const rows: string[] = [`Found ${edges.length} definition(s):`];
          for (const { node } of edges) {
            rows.push(`  ${node.name} (${node.type}) — ${node.metaobjectsCount ?? "?"} objects — ${node.id}`);
            for (const f of node.fieldDefinitions) {
              const req = f.required ? "*" : "";
              rows.push(`    - ${f.key}${req}: ${f.type.name}`);
            }
          }
          return {
            content: [{ type: "text" as const, text: rows.join("\n") }],
          };
        },
      );
    
      server.tool(
        "list_metaobjects",
        "List instances of a single metaobject type — e.g. all 'lookbook' or 'product_feature' entries. Returns each metaobject's display name, handle, GID, and (when the type is publishable) ACTIVE/DRAFT status. The type handle comes from list_metaobject_definitions. Cursor-paginated; pass `after` to advance pages. To inspect an individual metaobject's full field values, follow up with get_metaobject.",
        listMetaobjectsSchema,
        async (args) => {
          const data = await client.graphql<{
            metaobjects: Connection<MetaobjectNode>;
          }>(LIST_METAOBJECTS_QUERY, {
            type: args.type,
            first: args.first,
            after: args.after,
          });
          const edges = data.metaobjects.edges;
          if (edges.length === 0) {
            return {
              content: [
                {
                  type: "text" as const,
                  text: `No metaobjects of type "${args.type}".`,
                },
              ],
            };
          }
          const rows: string[] = [
            `Found ${edges.length} metaobject(s) of type "${args.type}":`,
          ];
          for (const { node } of edges) {
            const status = node.capabilities?.publishable?.status;
            const label = node.displayName ?? node.handle;
            rows.push(`  ${label}${status ? ` [${status}]` : ""} — ${node.handle} — ${node.id}`);
          }
          return {
            content: [{ type: "text" as const, text: rows.join("\n") }],
          };
        },
      );
    
      server.tool(
        "get_metaobject",
        "Fetch a single metaobject by GID and return its display name, handle, type, publishable status, and all of its field values. Field values longer than 120 characters are truncated in the rendered output (full values are still on the underlying record). Use list_metaobjects to discover GIDs first.",
        getMetaobjectSchema,
        async (args) => {
          const data = await client.graphql<{
            metaobject: MetaobjectNode | null;
          }>(GET_METAOBJECT_QUERY, { id: args.id });
          if (!data.metaobject) {
            return {
              content: [
                { type: "text" as const, text: `Metaobject not found: ${args.id}` },
              ],
            };
          }
          const m = data.metaobject;
          const status = m.capabilities?.publishable?.status;
          return {
            content: [
              {
                type: "text" as const,
                text: [
                  `${m.displayName ?? m.handle} (${m.type})${status ? ` [${status}]` : ""}`,
                  `  ID: ${m.id}`,
                  `  Handle: ${m.handle}`,
                  `  Updated: ${m.updatedAt}`,
                  "  Fields:",
                  ...formatMetaobjectFields(m.fields),
                ].join("\n"),
              },
            ],
          };
        },
      );
    
        server.tool(
        "create_metaobject",
        "Create a new metaobject (instance) of an existing type. The `type` must match a registered metaobject definition — call list_metaobject_definitions first if you're unsure. `fields` is an array of {key, value} pairs; values are always strings (JSON/reference fields take a JSON-encoded string, primitives take literal text). `handle` is optional; Shopify generates one from the displayName field if present. `status` only applies to types that have the `publishable` capability — passing it for non-publishable types is silently ignored. Returns the new metaobject's GID for use in subsequent set_metafield calls (e.g. linking the metaobject to a product via a metaobject_reference metafield).",
        createMetaobjectSchema,
        async (args) => {
          const metaobject: Record<string, unknown> = {
            type: args.type,
            fields: args.fields,
          };
          if (args.handle) metaobject.handle = args.handle;
          if (args.status) {
            metaobject.capabilities = {
              publishable: { status: args.status },
            };
          }
    
          const data = await client.graphql<{
            metaobjectCreate: {
              metaobject: MetaobjectNode | null;
              userErrors: ShopifyUserError[];
            };
          }>(METAOBJECT_CREATE_MUTATION, { metaobject });
          throwIfUserErrors(data.metaobjectCreate.userErrors, "metaobjectCreate");
          const m = data.metaobjectCreate.metaobject;
          if (!m) {
            return {
              content: [
                { type: "text" as const, text: "metaobjectCreate returned no metaobject." },
              ],
            };
          }
          return {
            content: [
              {
                type: "text" as const,
                text: `Created metaobject ${m.displayName ?? m.handle} (${m.type}) — ${m.id}`,
              },
            ],
          };
        },
      );
    
      server.tool(
        "update_metaobject",
        "Update an existing metaobject's handle, field values, or publishable status. Fields are upserted by key — pass only the fields you want to change; omitted fields keep their current values. To clear a field, pass an empty string or null-ish value matching the field type. If you change the handle, set redirectNewHandle=true to have Shopify redirect from the old handle on the storefront. The `type` cannot be changed by this tool — delete and recreate to change type.",
        updateMetaobjectSchema,
        async (args) => {
          const metaobject: Record<string, unknown> = {};
          if (args.handle !== undefined) metaobject.handle = args.handle;
          if (args.fields) metaobject.fields = args.fields;
          if (args.redirectNewHandle !== undefined) {
            metaobject.redirectNewHandle = args.redirectNewHandle;
          }
          if (args.status) {
            metaobject.capabilities = {
              publishable: { status: args.status },
            };
          }
          const data = await client.graphql<{
            metaobjectUpdate: {
              metaobject: MetaobjectNode | null;
              userErrors: ShopifyUserError[];
            };
          }>(METAOBJECT_UPDATE_MUTATION, { id: args.id, metaobject });
          throwIfUserErrors(data.metaobjectUpdate.userErrors, "metaobjectUpdate");
          const m = data.metaobjectUpdate.metaobject;
          if (!m) {
            return {
              content: [
                { type: "text" as const, text: "metaobjectUpdate returned no metaobject." },
              ],
            };
          }
          return {
            content: [
              {
                type: "text" as const,
                text: `Updated metaobject ${m.displayName ?? m.handle} (${m.type}) — ${m.id}`,
              },
            ],
          };
        },
      );
    
      server.tool(
        "delete_metaobject",
        "Permanently delete a metaobject by GID. Irreversible. Any metafield references pointing at this metaobject will become broken — Shopify does NOT auto-clean references, you have to find and fix them. Use get_metaobject to confirm the right record before deleting. Returns the deleted GID, or a no-op message if nothing matched.",
        deleteMetaobjectSchema,
        async (args) => {
          const data = await client.graphql<{
            metaobjectDelete: {
              deletedId: string | null;
              userErrors: ShopifyUserError[];
            };
          }>(METAOBJECT_DELETE_MUTATION, { id: args.id });
          throwIfUserErrors(data.metaobjectDelete.userErrors, "metaobjectDelete");
          return {
            content: [
              {
                type: "text" as const,
                text: data.metaobjectDelete.deletedId
                  ? `Deleted metaobject ${data.metaobjectDelete.deletedId}.`
                  : "No metaobject matched; nothing deleted.",
              },
            ],
          };
        },
      );
    }
  • src/server.ts:67-67 (registration)
    Where `registerMetaobjectTools` is called with the MCP server and Shopify client, registering the tool on the overall server.
    registerMetaobjectTools(s, shopify);
  • The GraphQL mutation `METAOBJECT_CREATE_MUTATION` used by the handler to create metaobjects via Shopify's Admin API.
    const METAOBJECT_CREATE_MUTATION = /* GraphQL */ `
      mutation MetaobjectCreate($metaobject: MetaobjectCreateInput!) {
        metaobjectCreate(metaobject: $metaobject) {
          metaobject {
            id
            type
            handle
            displayName
            fields { key type value }
            capabilities { publishable { status } }
          }
          userErrors { field message code }
        }
      }
    `;
Behavior4/5

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

With no annotations, the description covers key behaviors: silently ignoring status for non-publishable types, optional handle generation from displayName, and serialization of field values as strings. It also mentions the return value (GID).

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 description is concise yet comprehensive, front-loaded with the main action, and each sentence adds unique value. No redundancy.

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?

Given 4 parameters, no output schema, and no annotations, the description covers prerequisites, parameter details, edge cases, and return value. It is self-contained for an agent to use correctly.

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

Parameters5/5

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

Schema coverage is 100%, but the description adds significant value: explaining that type must match a registered definition, handle generation logic, field value serialization (JSON-encoded for complex types), and status restrictions. This goes well beyond the schema descriptions.

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?

The description clearly states it creates a new metaobject instance of an existing type. It distinguishes from sibling tools like delete_metaobject, update_metaobject, and get_metaobject by specifying creation and referencing list_metaobject_definitions for type verification.

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

Usage Guidelines4/5

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

The description advises calling list_metaobject_definitions first if unsure of the type, and explains optional handle generation and status applicability. It does not explicitly state when not to use, but provides clear context for usage.

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

Install Server

Other Tools

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/miller-joe/shopify-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server