> **Building with AI coding agents?** Install the authstack plugin with one command. This equips your agent with accurate Scalekit implementation patterns.
>
> **Recommended** (the `-y` flags skip prompts, so the command runs without an interactive terminal):
> ```bash
> npx -y @scalekit-inc/cli setup -y
> ```
>
> Global:
> ```bash
> npm install -g @scalekit-inc/cli
> scalekit setup -y
> ```
>
> Supports Claude Code, Cursor, GitHub Copilot, Codex + skills for other Agent Skills-compatible agents in `~/.agents/skills`.
> Skills: integrate-agentkit, implement-saaskit, add-mcp-oauth, implement-sso, implement-scim.
> [Full setup guide](https://docs.scalekit.com/dev-kit/build-with-ai/)

---

# Connect AI agents to the Holded MCP server

The Holded MCP connector routes your AI agent's tool calls to Holded's own MCP server through Scalekit. Each user signs in to Holded once, and Scalekit stores and refreshes their tokens, so your agent never handles credentials. It comes with 142 tools.

**Authentication:** OAuth 2.1/DCR
**Categories:** Accounting & Finance, CRM & Sales
**Tools:** 142: 142 read, 0 write, 0 destructive
**Users sign in with:** OAuth
**OAuth app:** Your own Holded MCP server app
**Built by:** Vendor MCP
**Try it:** [Playground in the Scalekit dashboard](https://app.scalekit.com/ws/signup?sk_intent=playground&provider=HOLDEDMCP)

## Setup

1. ### Install the SDK

   **Node.js**

   ```bash
   npm install @scalekit-sdk/node dotenv
   ```

   **Python**

   ```bash
   pip install scalekit-sdk-python python-dotenv
   ```

2. ### Set your credentials

   Add your Scalekit credentials to your `.env` file. Find values in **[app.scalekit.com](https://app.scalekit.com)** > **Developers** > **API Credentials**.

   ```sh title=".env"
   SCALEKIT_ENVIRONMENT_URL=<your-environment-url>
   SCALEKIT_CLIENT_ID=<your-client-id>
   SCALEKIT_CLIENT_SECRET=<your-client-secret>
   ```

3. ### Create the Holded MCP connection

   In **AgentKit > Connections**, create a Holded MCP connection and copy its redirect URI. The name you give it is the `connection_name` your code passes. See [Configure connections](/agentkit/connections/).

4. ### Register an OAuth app

   Holded MCP server connections use your own OAuth app. Register one with Holded MCP server and add the redirect URI you copied.

   Then enter the app's Client ID and Client Secret on the Holded MCP connection.

5. ### Authorize a user and make your first call

   **Node.js** (`quickstart.mts`)

   ```typescript
   import { ScalekitClient } from '@scalekit-sdk/node'
   import 'dotenv/config'
   import { createInterface } from 'node:readline/promises'

   const scalekit = new ScalekitClient(
     process.env.SCALEKIT_ENVIRONMENT_URL,
     process.env.SCALEKIT_CLIENT_ID,
     process.env.SCALEKIT_CLIENT_SECRET,
   )
   const actions = scalekit.actions

   const connector = 'holdedmcp'
   const identifier = 'user_123'

   // Generate an authorization link for the user
   const { link } = await actions.getAuthorizationLink({ connectionName: connector, identifier })
   console.log('Authorize Holded MCP:', link)
   const rl = createInterface({ input: process.stdin, output: process.stdout })
   await rl.question('Press Enter after authorizing...')
   rl.close()

   // Make your first call
   const result = await actions.executeTool({
     connector,
     identifier,
     toolName: 'holdedmcp_get_usage',
     toolInput: {},
   })
   console.log(result)
   ```

   ```bash
   npx tsx quickstart.mts
   ```

   **Python** (`quickstart.py`)

   ```python
   import os
   from scalekit import ScalekitClient
   from dotenv import load_dotenv
   load_dotenv()

   scalekit_client = ScalekitClient(
       env_url=os.getenv("SCALEKIT_ENVIRONMENT_URL"),
       client_id=os.getenv("SCALEKIT_CLIENT_ID"),
       client_secret=os.getenv("SCALEKIT_CLIENT_SECRET"),
   )
   actions = scalekit_client.actions

   connection_name = "holdedmcp"
   identifier = "user_123"

   # Generate an authorization link for the user
   link_response = actions.get_authorization_link(
       connection_name=connection_name,
       identifier=identifier,
   )
   print("Authorize Holded MCP:", link_response.link)
   input("Press Enter after authorizing...")

   # Make your first call
   result = actions.execute_tool(
       tool_input={},
       tool_name="holdedmcp_get_usage",
       connection_name=connection_name,
       identifier=identifier,
   )
   print(result)
   ```

   ```bash
   python quickstart.py
   ```

   Each user signs in once. See [Authorize a user](/agentkit/tools/authorize/) for the full flow and statuses.

## Tools

Pass the exact name to `execute_tool`, with an input like each tool's example. To ask for a tool that's missing, use the [request form](https://scalekitsupport.portal.usepylon.com/forms/request-a-connector-tool).

**Node.js**

```typescript
const result = await actions.executeTool({
  toolName: 'holdedmcp_download_inbox_file',
  toolInput: {
    filename: '<filename>',
    incomingDocumentId: '<incomingDocumentId>',
  },
  connector: 'holdedmcp',
  identifier: 'user_123',
})
```

**Python**

```python
result = actions.execute_tool(
    tool_name="holdedmcp_download_inbox_file",
    tool_input={
        "filename": "<filename>",
        "incomingDocumentId": "<incomingDocumentId>",
    },
    connection_name="holdedmcp",
    identifier="user_123",
)
```

`result.data` is the app's response as JSON, and `result.execution_id` (`executionId` in Node.js) is the ID of the call.

### `holdedmcp_download_inbox_file`

Download a file from an inbox document · Read-only

**Required permission:** `inbox:inbox.read`

Returns the raw content of a file attached to an incoming document. Use `?variant=inline` to display the file in-browser with 24-hour caching instead of triggering a download.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `filename` (`string`, required): Name of the file to retrieve
- `incomingDocumentId` (`string`, required): Incoming document identifier
- `variant` (`string`, optional): Delivery mode: download (attachment) or inline (display with cache). One of: `download`, `inline`. Default: `download`.

Example input: `{"filename":"<filename>","incomingDocumentId":"<incomingDocumentId>"}`

### `holdedmcp_find_invoices_by_document_number`

Find invoices by document number · Read-only

**Required permission:** `sales:invoices.read`

Returns invoices in your account whose document number matches the given value.

The match is **fuzzy autocomplete**: partial prefixes (e.g. `F260`) return all invoices that start with that token, and minor typos are tolerated. Results are scored by relevance and limited to the authenticated account. Draft, deleted and cancelled invoices are excluded.

Results are cursor-paginated: the response includes a `cursor` value and a `has_more` boolean. Pass the `cursor` to the next request to fetch the following page.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `document_number` (`string`, required): Document number to look up. Supports prefixes (e.g. `F260`) and tolerates minor typos.
- `cursor` (`string`, optional): Pagination cursor returned by the previous response. Omit to start from the first page.
- `limit` (`integer`, optional): Maximum number of invoices to return per page. Default: `50`.

Example input: `{"document_number":"F260"}`

### `holdedmcp_get_banking_account`

Get a banking account by ID · Read-only

**Required permission:** `accounting:banks.read`

Retrieves the full details of a single banking account by its unique identifier.

The response includes the current balance, bank details (IBAN, BIC), synchronization status, and archived state.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `id` (`string`, required): Banking account identifier

Example input: `{"id":"<id>"}`

### `holdedmcp_get_booking`

Get a booking by ID · Read-only

**Required permission:** `crm:booking-services.read`

Returns a single booking identified by its unique ID.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `bookingId` (`string`, required): Booking ID (24-char hex ObjectId)

Example input: `{"bookingId":"<bookingId>"}`

### `holdedmcp_get_booking_location_slots`

Get available slots for a location · Read-only

**Required permission:** `crm:booking-services.read`

Returns available time slots for a specific service at a location on a given day. The response also includes the booking form fields (under `form`) that must be collected from the user before calling Create a booking.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `day` (`string`, required): Date in YYYY-MM-DD format
- `locationId` (`string`, required): Location ID (24-char hex ObjectId)
- `service_id` (`string`, required): Service identifier

Example input: `{"day":"<day>","locationId":"<locationId>","service_id":"<service_id>"}`

### `holdedmcp_get_contact`

Get a contact by ID · Read-only

**Required permission:** `contacts:contacts.read`

Retrieves the full details of a single contact by its unique identifier.

The response includes all contact data: billing address, shipping addresses, contact persons, custom fields, default invoicing settings, social networks, and accounting records. Use this endpoint when you need the complete contact profile — for list views, prefer the List contacts endpoint which returns the same fields but paginated.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `contactId` (`string`, required): Contact ID (24-char hex ObjectId)

Example input: `{"contactId":"<contactId>"}`

### `holdedmcp_get_contact_attachment`

Get a contact attachment · Read-only

**Required permission:** `contacts:contacts.read`

Downloads a single attachment from a contact.

The response body contains the raw file bytes. Use the `Content-Disposition` header to obtain the original file name.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `contactId` (`string`, required): Contact ID (24-char hex ObjectId)
- `filename` (`string`, required): Attachment file name (case-insensitive match)

Example input: `{"contactId":"<contactId>","filename":"<filename>"}`

### `holdedmcp_get_contact_group`

Get a contact group · Read-only

**Required permission:** `contacts:contacts.read`

Returns a single contact group by its ID.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `contactGroupId` (`string`, required): Contact Group ID (24-char hex ObjectId)

Example input: `{"contactGroupId":"<contactGroupId>"}`

### `holdedmcp_get_contact_portal_link`

Get client portal link for a contact · Read-only

**Required permission:** `contacts:contacts.read`

Generates a portal access link for a specific contact. The link allows the contact to access their client portal where they can view documents. The link expires after 2 days.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `contactId` (`string`, required): Contact ID (24-char hex ObjectId)

Example input: `{"contactId":"<contactId>"}`

### `holdedmcp_get_credit_note`

Get a credit note · Read-only

**Required permission:** `sales:invoices.read`

Retrieves the full details of a single credit note by its unique identifier.

The response includes line items, payment status and breakdown, custom fields, and the source document if the credit note was created from an invoice.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `creditNoteId` (`string`, required): Credit Note ID (24-char hex ObjectId)

Example input: `{"creditNoteId":"<creditNoteId>"}`

### `holdedmcp_get_credit_note_attachment`

Get credit note attachment · Read-only

**Required permission:** `sales:invoices.read`

Downloads a specific attachment from a credit note. The response contains the raw binary content of the file.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `attachmentId` (`string`, required): Attachment identifier (filename)
- `creditNoteId` (`string`, required): Credit Note ID (24-char hex ObjectId)

Example input: `{"attachmentId":"<attachmentId>","creditNoteId":"<creditNoteId>"}`

### `holdedmcp_get_credit_note_pdf`

Get credit note PDF · Read-only

**Required permission:** `sales:invoices.read`

Downloads the credit note as a PDF file. The response Content-Type is `application/pdf`.

The PDF is generated using the template configured in your account settings.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `creditNoteId` (`string`, required): Credit Note ID (24-char hex ObjectId)

Example input: `{"creditNoteId":"<creditNoteId>"}`

### `holdedmcp_get_employee`

Get an employee · Read-only

**Required permission:** `team:employees.read`

Returns a single employee by its ID.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `employeeId` (`string`, required): Employee ID

Example input: `{"employeeId":"<employeeId>"}`

### `holdedmcp_get_employee_contract`

Get active contract · Read-only

**Required permission:** `team:employees.read`

Returns all fields of the employee's active contract. Use this as the first step before updating a contract field: fetch the current data here, then send ALL fields back to the PUT endpoint modifying only what you want to change. Field names in this response match exactly the field names expected by the PUT endpoint.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `employeeId` (`string`, required): Employee ID

Example input: `{"employeeId":"<employeeId>"}`

### `holdedmcp_get_employee_time`

Get an employee time entry · Read-only

**Required permission:** `team:time-tracking.read`

Returns a single employee time tracking entry by its ID.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `timeId` (`string`, required): Employee time entry ID

Example input: `{"timeId":"<timeId>"}`

### `holdedmcp_get_estimate`

Get an estimate · Read-only

**Required permission:** `sales:estimates.read`

Retrieves the full details of a single estimate by its unique identifier.

The response includes line items, custom fields, pipeline stage, and the source document if the estimate was converted from another document.

A signed-in user with limited access to estimates only reaches the estimates assigned to them or created by them; any other is reported as not found. API keys still need the scope of this endpoint.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `estimateId` (`string`, required): Estimate ID (24-char hex ObjectId)

Example input: `{"estimateId":"<estimateId>"}`

### `holdedmcp_get_estimate_attachment`

Get estimate attachment · Read-only

**Required permission:** `sales:estimates.read`

Downloads a specific attachment from an estimate. The response contains the raw binary content of the file. A signed-in user with limited access to estimates only reaches the estimates assigned to them or created by them; any other is reported as not found. API keys still need the scope of this endpoint.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `attachmentId` (`string`, required): Attachment identifier (filename)
- `estimateId` (`string`, required): Estimate ID (24-char hex ObjectId)

Example input: `{"attachmentId":"<attachmentId>","estimateId":"<estimateId>"}`

### `holdedmcp_get_estimate_pdf`

Get estimate PDF · Read-only

**Required permission:** `sales:estimates.read`

Downloads the estimate as a PDF file. The response Content-Type is `application/pdf`.

The PDF is generated using the template configured in your account settings and reflects the current state of the estimate.

A signed-in user with limited access to estimates only reaches the estimates assigned to them or created by them; any other is reported as not found. API keys still need the scope of this endpoint.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `estimateId` (`string`, required): Estimate ID (24-char hex ObjectId)

Example input: `{"estimateId":"<estimateId>"}`

### `holdedmcp_get_event`

Get an event by ID · Read-only

**Required permission:** `crm:events.read`

Retrieves the full details of a single CRM event by its unique identifier.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `eventId` (`string`, required): Event ID (24-char hex ObjectId)

Example input: `{"eventId":"<eventId>"}`

### `holdedmcp_get_expenses_account`

Get an expenses account by ID · Read-only

**Required permission:** `accounting:expenses-accounts.read`

Retrieves the details of a single expenses account by its unique identifier.

The response includes the account name, color, accounting number, and archived status.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `expensesAccountId` (`string`, required): Expenses Account ID (24-char hex ObjectId)

Example input: `{"expensesAccountId":"<expensesAccountId>"}`

### `holdedmcp_get_funnel`

Get a funnel · Read-only

**Required permission:** `crm:funnels.read`

Returns a single funnel by its ID.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `funnelId` (`string`, required): Funnel ID

Example input: `{"funnelId":"<funnelId>"}`

### `holdedmcp_get_inbox`

Get an incoming document by ID · Read-only

**Required permission:** `inbox:inbox.read`

Retrieves the full details of a single incoming document by its unique identifier.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `incomingDocumentId` (`string`, required): Incoming document identifier (24-char hex ObjectId)

Example input: `{"incomingDocumentId":"<incomingDocumentId>"}`

### `holdedmcp_get_invoice`

Get an invoice · Read-only

**Required permission:** `sales:invoices.read`

Retrieves the full details of a single invoice by its unique identifier.

The response includes line items, payment status and breakdown, custom fields, pipeline stage, and the source document if the invoice was converted from an estimate or sales order.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `invoiceId` (`string`, required): Invoice ID (24-char hex ObjectId)

Example input: `{"invoiceId":"<invoiceId>"}`

### `holdedmcp_get_invoice_attachment`

Get invoice attachment · Read-only

**Required permission:** `sales:invoices.read`

Downloads a specific attachment from an invoice. The response contains the raw binary content of the file.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `attachmentId` (`string`, required): Attachment identifier (filename)
- `invoiceId` (`string`, required): Invoice ID (24-char hex ObjectId)

Example input: `{"attachmentId":"<attachmentId>","invoiceId":"<invoiceId>"}`

### `holdedmcp_get_invoice_pdf`

Get invoice PDF · Read-only

**Required permission:** `sales:invoices.read`

Downloads the invoice as a PDF file. The response Content-Type is `application/pdf`.

The PDF is generated using the template configured in your account settings. It reflects the current state of the invoice, including any updates made after creation.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `invoiceId` (`string`, required): Invoice ID (24-char hex ObjectId)

Example input: `{"invoiceId":"<invoiceId>"}`

### `holdedmcp_get_invoice_recurring`

Get a recurring invoice · Read-only

**Required permission:** `sales:invoices.read`

Retrieves the full details of a single recurring invoice by its unique identifier.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `invoiceRecurringId` (`string`, required): Recurring invoice ID (24-char hex ObjectId)

Example input: `{"invoiceRecurringId":"<invoiceRecurringId>"}`

### `holdedmcp_get_invoicing_forecast`

Get an invoicing forecast · Read-only

**Required permission:** `accounting:cashflow.read`

Returns the full details of a single invoicing forecast by its unique identifier. An invoicing forecast represents a scheduled payment milestone on an estimate or purchase order.

Use when the user asks for the details of a specific cashflow forecast by id.

Distinct from the List endpoint: this returns one complete forecast record rather than a paginated collection.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `forecastId` (`string`, required): Invoicing forecast identifier (24-character hex ObjectId)

Example input: `{"forecastId":"<forecastId>"}`

### `holdedmcp_get_lead`

Get a lead · Read-only

**Required permission:** `crm:deals.read`

Returns a single lead by its ID.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `leadId` (`string`, required): Lead ID

Example input: `{"leadId":"<leadId>"}`

### `holdedmcp_get_ledger_entry`

Get a ledger entry · Read-only

**Required permission:** `accounting:daily-ledger.read`

Retrieves a single ledger entry (asiento contable) by its ID, including every debit/credit line that makes up the entry.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `ledgerEntryId` (`string`, required): Ledger entry ID (24-char hex ObjectId)

Example input: `{"ledgerEntryId":"<ledgerEntryId>"}`

### `holdedmcp_get_payment`

Get a payment by ID · Read-only

**Required permission:** `accounting:payments.read`

Retrieves the full details of a single payment record by its unique identifier.

The response includes amount, date, status, reconciliation status, and the linked contact and source document.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `paymentId` (`string`, required): Payment ID (24-char hex ObjectId)

Example input: `{"paymentId":"<paymentId>"}`

### `holdedmcp_get_payment_method`

Get a payment method by ID · Read-only

**Required permission:** `sales:invoicing-settings.read`

Returns a single configured payment method by its ID. The ID comes from the 'id' field in the list endpoint response. Note: this returns a configured payment method (manual or connected gateway such as Stripe, PayPal, GoCardless, Square, Monei, or Wallet) — not an individual transaction. Use /api/v2/payments/{id} to fetch a specific payment transaction.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `paymentMethodId` (`string`, required): Payment method identifier

Example input: `{"paymentMethodId":"<paymentMethodId>"}`

### `holdedmcp_get_price_list`

Get a price list by ID · Read-only

**Required permission:** `inventory:rates.read`

Retrieves the details of a single price list (rate) by its unique identifier.

The response includes the price list name and optional description.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `priceListId` (`string`, required): Price List ID (24-char hex ObjectId)

Example input: `{"priceListId":"<priceListId>"}`

### `holdedmcp_get_product`

Get a product by ID · Read-only

**Required permission:** `inventory:products.read`

Retrieves the full details of a single product by its unique identifier.

The response includes variants, stock levels across warehouses, custom fields, tax configuration, and pricing information.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `productId` (`string`, required): Product ID (24-char hex ObjectId)

Example input: `{"productId":"<productId>"}`

### `holdedmcp_get_product_image`

Get a product image by ID · Read-only

**Required permission:** `inventory:products.read`

Retrieves a single image from a product by its image identifier.

The response includes the image URL, available sizes with dimensions, position, and description.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `imageId` (`string`, required): Image ID (24-char hex ObjectId)
- `productId` (`string`, required): Product ID (24-char hex ObjectId)

Example input: `{"imageId":"<imageId>","productId":"<productId>"}`

### `holdedmcp_get_product_main_image`

Get the main product image · Read-only

**Required permission:** `inventory:products.read`

Retrieves the main (first) image of a product. Returns 404 if the product has no images.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `productId` (`string`, required): Product ID (24-char hex ObjectId)

Example input: `{"productId":"<productId>"}`

### `holdedmcp_get_product_stock`

Get stock availability for a product · Read-only

**Required permission:** `inventory:products.read`

Returns the stock availability of a product, broken down into total, available, reserved and in-transit quantities across all warehouses, plus the physical stock stored in each warehouse.

Reserved stock represents units committed to pending sales orders. In-transit stock represents units incoming from purchase orders. Available stock equals total minus reserved.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `productId` (`string`, required): Product ID (24-char hex ObjectId)

Example input: `{"productId":"<productId>"}`

### `holdedmcp_get_product_stock_transit`

Get stock in transit for a product · Read-only

**Required permission:** `inventory:products.read`

Returns the stock in transit for a product, including per-document movement details.

Each movement represents a document (purchase order, sales order or production order) that affects the product's transit stock. Incoming movements come from purchase orders. Compromised movements come from sales or production orders.

For pack products, the response aggregates transit data from all component products.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `productId` (`string`, required): Product ID (24-char hex ObjectId)

Example input: `{"productId":"<productId>"}`

### `holdedmcp_get_production_order`

Get a production order · Read-only

**Required permission:** `inventory:production-orders.read`

Retrieves the full details of a single production order by its unique identifier.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `productionOrderId` (`string`, required): Production Order ID (24-char hex ObjectId)

Example input: `{"productionOrderId":"<productionOrderId>"}`

### `holdedmcp_get_proforma`

Get a proforma · Read-only

**Required permission:** `sales:proforms.read`

Retrieves the full details of a single proforma invoice by its unique identifier.

The response includes line items, custom fields, pipeline stage, and the source document if the proforma was converted from another document.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `proformaId` (`string`, required): Proforma ID (24-char hex ObjectId)

Example input: `{"proformaId":"<proformaId>"}`

### `holdedmcp_get_proforma_attachment`

Get proforma attachment · Read-only

**Required permission:** `sales:proforms.read`

Downloads a specific attachment from a proforma. The response contains the raw binary content of the file.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `attachmentId` (`string`, required): Attachment identifier (filename)
- `proformaId` (`string`, required): Proforma ID (24-char hex ObjectId)

Example input: `{"attachmentId":"<attachmentId>","proformaId":"<proformaId>"}`

### `holdedmcp_get_proforma_pdf`

Get proforma PDF · Read-only

**Required permission:** `sales:proforms.read`

Downloads the proforma invoice as a PDF file. The response Content-Type is `application/pdf`.

The PDF reflects the current state of the proforma.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `proformaId` (`string`, required): Proforma ID (24-char hex ObjectId)

Example input: `{"proformaId":"<proformaId>"}`

### `holdedmcp_get_project`

Get a project · Read-only

**Required permission:** `projects:projects.read`

Returns a single project by its ID.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `projectId` (`string`, required): Project ID

Example input: `{"projectId":"<projectId>"}`

### `holdedmcp_get_project_summary`

Get a project summary · Read-only

**Required permission:** `projects:projects.read`

Returns the summary of a project by its ID.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `projectId` (`string`, required): Project ID

Example input: `{"projectId":"<projectId>"}`

### `holdedmcp_get_project_time`

Get a project time entry · Read-only

**Required permission:** `projects:projects.read`

Returns a single project time tracking entry by its ID.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `projectId` (`string`, required): Project ID
- `timeId` (`string`, required): Project time entry ID

Example input: `{"projectId":"<projectId>","timeId":"<timeId>"}`

### `holdedmcp_get_purchase`

Get a purchase · Read-only

**Required permission:** `accounting:purchases.read`

Retrieves the full details of a single purchase invoice by its unique identifier.

The response includes line items, payment status and breakdown, custom fields, and the source document if the purchase was created from a purchase order.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `purchaseId` (`string`, required): Purchase ID (24-char hex ObjectId)

Example input: `{"purchaseId":"<purchaseId>"}`

### `holdedmcp_get_purchase_attachment`

Get purchase attachment · Read-only

**Required permission:** `accounting:purchases.read`

Downloads a specific attachment from a purchase. The response contains the raw binary content of the file.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `attachmentId` (`string`, required): Attachment identifier (filename)
- `purchaseId` (`string`, required): Purchase ID (24-char hex ObjectId)

Example input: `{"attachmentId":"<attachmentId>","purchaseId":"<purchaseId>"}`

### `holdedmcp_get_purchase_order`

Get a purchase order · Read-only

**Required permission:** `inventory:purchase-orders.read`

Retrieves the full details of a single purchase order by its unique identifier.

The response includes line items, delivery tracking information, custom fields, and pipeline stage.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `purchaseOrderId` (`string`, required): Purchase Order ID (24-char hex ObjectId)

Example input: `{"purchaseOrderId":"<purchaseOrderId>"}`

### `holdedmcp_get_purchase_order_attachment`

Get purchase order attachment · Read-only

**Required permission:** `inventory:purchase-orders.read`

Downloads a specific attachment from a purchase order. The response contains the raw binary content of the file.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `attachmentId` (`string`, required): Attachment identifier (filename)
- `purchaseOrderId` (`string`, required): Purchase Order ID (24-char hex ObjectId)

Example input: `{"attachmentId":"<attachmentId>","purchaseOrderId":"<purchaseOrderId>"}`

### `holdedmcp_get_purchase_order_pdf`

Get purchase order PDF · Read-only

**Required permission:** `inventory:purchase-orders.read`

Downloads the purchase order as a PDF file. The response Content-Type is `application/pdf`.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `purchaseOrderId` (`string`, required): Purchase Order ID (24-char hex ObjectId)

Example input: `{"purchaseOrderId":"<purchaseOrderId>"}`

### `holdedmcp_get_purchase_order_received_items`

Get received items for a purchase order · Read-only

**Required permission:** `inventory:purchase-orders.read`

Returns the items that have already been received for this purchase order.

The response includes received quantities and pending quantities for each line item, so you can track receipt progress.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `purchaseOrderId` (`string`, required): Purchase Order ID (24-char hex ObjectId)

Example input: `{"purchaseOrderId":"<purchaseOrderId>"}`

### `holdedmcp_get_purchase_refund`

Get a purchase refund · Read-only

**Required permission:** `accounting:purchases.read`

Retrieves the full details of a single purchase refund (compra rectificativa) by its unique identifier.

The response includes line items, payment status and breakdown, custom fields, and the source purchase document if the refund was created from an existing purchase.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `purchaseRefundId` (`string`, required): Purchase Refund ID (24-char hex ObjectId)

Example input: `{"purchaseRefundId":"<purchaseRefundId>"}`

### `holdedmcp_get_purchase_shipment`

Get a purchase shipment · Read-only

**Required permission:** `inventory:purchase-shipments.read`

Retrieves the full details of a single purchase shipment by its unique identifier.

The response includes line items, custom fields, and warehouse information.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `purchaseShipmentId` (`string`, required): Purchase Shipment ID (24-char hex ObjectId)

Example input: `{"purchaseShipmentId":"<purchaseShipmentId>"}`

### `holdedmcp_get_receipt_note`

Get a receipt note · Read-only

**Required permission:** `sales:invoices.read`

Retrieves the full details of a single receipt note by its unique identifier.

The response includes line items, payment status and breakdown, custom fields, and the source document if the receipt note was created from a sales receipt.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `receiptNoteId` (`string`, required): Receipt Note ID (24-char hex ObjectId)

Example input: `{"receiptNoteId":"<receiptNoteId>"}`

### `holdedmcp_get_receipt_note_attachment`

Get receipt note attachment · Read-only

**Required permission:** `sales:invoices.read`

Downloads a specific attachment from a receipt note. The response contains the raw binary content of the file.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `attachmentId` (`string`, required): Attachment identifier (filename)
- `receiptNoteId` (`string`, required): Receipt Note ID (24-char hex ObjectId)

Example input: `{"attachmentId":"<attachmentId>","receiptNoteId":"<receiptNoteId>"}`

### `holdedmcp_get_receipt_note_pdf`

Get receipt note PDF · Read-only

**Required permission:** `sales:invoices.read`

Downloads the receipt note as a PDF file. The response Content-Type is `application/pdf`.

The PDF is generated using the template configured in your account settings.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `receiptNoteId` (`string`, required): Receipt Note ID (24-char hex ObjectId)

Example input: `{"receiptNoteId":"<receiptNoteId>"}`

### `holdedmcp_get_recurring_schedule`

Get recurring invoice schedule · Read-only

**Required permission:** `sales:invoices.read`

Returns the occurrence schedule for a recurring invoice, including past converted, skipped, and future pending dates.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `invoiceRecurringId` (`string`, required): Recurring invoice ID (24-char hex ObjectId)
- `until` (`string`, optional): Upper date boundary (YYYY-MM-DD). Defaults to one year from today.

Example input: `{"invoiceRecurringId":"<invoiceRecurringId>"}`

### `holdedmcp_get_remittance`

Get a remittance · Read-only

**Required permission:** `accounting:remittances.read`

Retrieves the full details of a remittance by its unique identifier. A remittance groups multiple invoices or purchases into a single SEPA direct debit or credit transfer for batch processing.

The response includes the list of documents bundled in the remittance, each with its contact, amount, and payment status.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `remittanceId` (`string`, required): Remittance ID

Example input: `{"remittanceId":"<remittanceId>"}`

### `holdedmcp_get_sales_channel`

Get a sales channel by ID · Read-only

**Required permission:** `sales:sales-channels.read`

Retrieves the details of a single sales channel by its unique identifier.

The response includes the channel name, color, accounting number, and archived status.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `salesChannelId` (`string`, required): Sales Channel ID (24-char hex ObjectId)

Example input: `{"salesChannelId":"<salesChannelId>"}`

### `holdedmcp_get_sales_order`

Get a sales order · Read-only

**Required permission:** `inventory:sales-orders.read`

Retrieves the full details of a single sales order by its unique identifier.

The response includes line items, shipping tracking information, custom fields, pipeline stage, and shipped item quantities.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `salesOrderId` (`string`, required): Sales Order ID (24-char hex ObjectId)

Example input: `{"salesOrderId":"<salesOrderId>"}`

### `holdedmcp_get_sales_order_attachment`

Get sales order attachment · Read-only

**Required permission:** `inventory:sales-orders.read`

Downloads a specific attachment from a sales order. The response contains the raw binary content of the file.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `attachmentId` (`string`, required): Attachment identifier (filename)
- `salesOrderId` (`string`, required): Sales Order ID (24-char hex ObjectId)

Example input: `{"attachmentId":"<attachmentId>","salesOrderId":"<salesOrderId>"}`

### `holdedmcp_get_sales_order_pdf`

Get sales order PDF · Read-only

**Required permission:** `inventory:sales-orders.read`

Downloads the sales order as a PDF file. The response Content-Type is `application/pdf`.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `salesOrderId` (`string`, required): Sales Order ID (24-char hex ObjectId)

Example input: `{"salesOrderId":"<salesOrderId>"}`

### `holdedmcp_get_sales_order_shipped_items`

Get shipped items for a sales order · Read-only

**Required permission:** `inventory:sales-orders.read`

Returns the items that have already been shipped for this sales order.

The response includes shipped quantities and dates for each line item, so you can track fulfillment progress.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `salesOrderId` (`string`, required): Sales Order ID (24-char hex ObjectId)

Example input: `{"salesOrderId":"<salesOrderId>"}`

### `holdedmcp_get_sales_receipt`

Get a sales receipt · Read-only

**Required permission:** `sales:receipts.read`

Retrieves the full details of a single sales receipt by its unique identifier.

The response includes line items, payment status and breakdown, custom fields, and pipeline stage.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `salesReceiptId` (`string`, required): Sales Receipt ID (24-char hex ObjectId)

Example input: `{"salesReceiptId":"<salesReceiptId>"}`

### `holdedmcp_get_sales_receipt_attachment`

Get sales receipt attachment · Read-only

**Required permission:** `sales:receipts.read`

Downloads a specific attachment from a sales receipt. The response contains the raw binary content of the file.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `attachmentId` (`string`, required): Attachment identifier (filename)
- `salesReceiptId` (`string`, required): Sales Receipt ID (24-char hex ObjectId)

Example input: `{"attachmentId":"<attachmentId>","salesReceiptId":"<salesReceiptId>"}`

### `holdedmcp_get_sales_receipt_pdf`

Get sales receipt PDF · Read-only

**Required permission:** `sales:receipts.read`

Downloads the sales receipt as a PDF file. The response Content-Type is `application/pdf`.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `salesReceiptId` (`string`, required): Sales Receipt ID (24-char hex ObjectId)

Example input: `{"salesReceiptId":"<salesReceiptId>"}`

### `holdedmcp_get_service`

Get a service · Read-only

**Required permission:** `sales:services.read`

Retrieves the full details of a single service by its unique identifier.

The response includes pricing, tax configuration, sales channel assignment, and duration.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `serviceId` (`string`, required): Service ID (24-char hex ObjectId)

Example input: `{"serviceId":"<serviceId>"}`

### `holdedmcp_get_task`

Get a task · Read-only

**Required permission:** `projects:projects.read`

Returns a single task by its ID.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `taskId` (`string`, required): Task ID

Example input: `{"taskId":"<taskId>"}`

### `holdedmcp_get_tax_per_country_sorted_keys`

Get sorted tax keys by country and section · Read-only

**Required permission:** `accounting:taxes.read`

Returns the ordered list of tax keys applicable for a given country and ecommerce section. Combines default country rules with section-specific overrides.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `country` (`string`, required): ISO 3166-1 alpha-2 country code (e.g. ES, FR, DE)
- `section` (`string`, optional): Ecommerce section (all|shopify|prestashop|woocommerce|amazon|wix|stripe). Defaults to "all" when omitted. Default: `all`.

Example input: `{"country":"<country>"}`

### `holdedmcp_get_usage`

Get API usage (automation token) · Read-only

**Required permission:** `developers:usage.read`

Returns the current month API usage for the API.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `holdedmcp_get_usage_by_type`

Get API usage by type · Read-only

**Required permission:** `developers:usage.read`

Returns the current month API usage for the given type.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `type` (`string`, required): Usage type identifier. One of: `automation_token`, `digital_signature`, `ai_banking_rule_suggestion`, `llm_tokens`.

Example input: `{"type":"automation_token"}`

### `holdedmcp_get_warehouse`

Get a warehouse by ID · Read-only

**Required permission:** `inventory:warehouses.read`

Retrieves the details of a single warehouse by its unique identifier.

The response includes the warehouse name, contact information, address, and whether it is the default warehouse.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `warehouseId` (`string`, required): Warehouse ID (24-char hex ObjectId)

Example input: `{"warehouseId":"<warehouseId>"}`

### `holdedmcp_get_warehouse_stock`

Get stock levels for a warehouse · Read-only

**Required permission:** `inventory:warehouses.read`

Returns the stock levels for all products stored in a specific warehouse.

The response includes each product's ID, variant ID (if applicable), and current stock quantity. Use this endpoint to audit inventory or synchronize stock with external systems.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `warehouseId` (`string`, required): Warehouse ID (24-char hex ObjectId)

Example input: `{"warehouseId":"<warehouseId>"}`

### `holdedmcp_get_waybill`

Get a waybill · Read-only

**Required permission:** `inventory:shipments.read`

Retrieves the full details of a single waybill by its unique identifier.

The response includes line items, shipping tracking information, custom fields, and pipeline stage.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `waybillId` (`string`, required): Waybill ID (24-char hex ObjectId)

Example input: `{"waybillId":"<waybillId>"}`

### `holdedmcp_get_waybill_attachment`

Get waybill attachment · Read-only

**Required permission:** `inventory:shipments.read`

Downloads a specific attachment from a waybill. The response contains the raw binary content of the file.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `attachmentId` (`string`, required): Attachment identifier (filename)
- `waybillId` (`string`, required): Waybill ID (24-char hex ObjectId)

Example input: `{"attachmentId":"<attachmentId>","waybillId":"<waybillId>"}`

### `holdedmcp_get_waybill_pdf`

Get waybill PDF · Read-only

**Required permission:** `inventory:shipments.read`

Downloads the waybill as a PDF file. The response Content-Type is `application/pdf`.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `waybillId` (`string`, required): Waybill ID (24-char hex ObjectId)

Example input: `{"waybillId":"<waybillId>"}`

### `holdedmcp_get_webhook`

Get a webhook · Read-only

**Required permission:** `developers:webhooks.read`

Retrieves a webhook by its unique identifier.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `webhookId` (`string`, required): Webhook ID

Example input: `{"webhookId":"<webhookId>"}`

### `holdedmcp_get_webhook_events`

List supported webhook event types · Read-only

**Required permission:** `developers:webhooks.read`

Returns all supported webhook event types grouped by domain.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `holdedmcp_list_accounting_accounts`

List chart of accounts · Read-only

**Required permission:** `accounting:chart-of-accounts.read`

Returns the list of all accounting accounts in your chart of accounts.

The response includes each account's number, name, and group. Use this endpoint to discover available accounts before creating ledger entries.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `archived` (`boolean`, optional): Include archived accounts when true
- `end_date` (`string`, optional): Filter accounts with activity up to this date (YYYY-MM-DD)
- `include_empty` (`boolean`, optional): Include accounts with no transactions when true
- `start_date` (`string`, optional): Filter accounts with activity from this date (YYYY-MM-DD)
- `text` (`string`, optional): Case-insensitive partial search on account name or number

Example input: `{}`

### `holdedmcp_list_all_employee_times`

List all employee time entries · Read-only

**Required permission:** `team:time-tracking.read`

Returns a paginated list of all employee time tracking entries for the account.

Results are cursor-paginated: the response includes a `cursor` value and a `has_more` boolean. Pass the `cursor` to the next request to fetch the following page. Default page size is 50, configurable up to 200.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `cursor` (`string`, optional): Opaque cursor for pagination. Use the value returned in the previous response to fetch the next page.
- `limit` (`integer`, optional): Maximum number of items to return. Default: `50`.

Example input: `{}`

### `holdedmcp_list_all_project_times`

List all project time entries · Read-only

**Required permission:** `projects:projects.read`

Returns a paginated list of the project time tracking entries of the account, across every project the caller can see. Without `startDate`, only entries from the 18 months before `endDate` (or before today) are returned; without `endDate`, entries dated after now are left out.

Results are paginated with `limit` and `cursor` parameters. Default page size is 50, configurable up to 200.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `cursor` (`string`, optional): Pagination cursor returned by the previous response. Omit to start from the first page.
- `employee_id` (`string`, optional): Filter by employee: only the entries logged by the employee's linked user are returned, and an employee with no linked user has none. Takes an employee id, not a user id (neither an entry's `user_id` nor an employee's `holded_user_id`). An id that is not an employee of the account returns 404.
- `endDate` (`string`, optional): Filter entries on or before this date (ISO 8601, YYYY-MM-DD)
- `limit` (`integer`, optional): Maximum number of items to return. Default: `50`.
- `startDate` (`string`, optional): Filter entries on or after this date (ISO 8601, YYYY-MM-DD)

Example input: `{}`

### `holdedmcp_list_banking_account_bank_movements`

List banking account movements · Read-only

**Required permission:** `accounting:banks.read`

Returns a cursor-paginated list of raw bank movements for a banking account, with the bank-reported amount, balance after the movement, value date and reconciliation status. Use this endpoint to build reconciliation flows, cash-flow analysis, balance alerts or banking dashboards. Unlike Payment records (available at `/api/v2/payments`), which link bank movements to invoices and documents, this endpoint exposes the raw entries that come from the bank statement. Results are ordered by value date descending (then by id descending to break ties). Returns an empty list for cash-type accounts (their movements live in the ledger, not the bank statement) — use `GET /api/v2/treasury/accounts/{id}/cash-movements` instead.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `id` (`string`, required): Banking account identifier
- `cursor` (`string`, optional): Opaque cursor returned by the previous page
- `end_date` (`string`, optional): Inclusive upper bound on the movement value date (ISO 8601 date or date-time)
- `limit` (`integer`, optional): Maximum number of movements per page. Defaults to 50, maximum 200. Default: `50`.
- `start_date` (`string`, optional): Inclusive lower bound on the movement value date (ISO 8601 date or date-time)
- `status` (`array`, optional): Reconciliation statuses to filter by: 'pending' (not reconciled), 'reconciled' (fully matched), 'partial' (partially matched), 'forced_reconciled' (manually marked).

Example input: `{"id":"<id>"}`

### `holdedmcp_list_banking_account_cash_movements`

List banking account cash movements · Read-only

**Required permission:** `accounting:banks.read`

Returns a cursor-paginated list of ledger entries for a cash-type banking account. Use this endpoint to list cash register movements — entries created manually or automatically from payments linked to a cash account. This endpoint only works for accounts of type cash; for bank, card, or gateway accounts use `/api/v2/treasury/accounts/{id}/bank-movements` instead (which returns raw bank-statement entries from providers).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `id` (`string`, required): Banking account identifier
- `cursor` (`string`, optional): Opaque cursor returned by the previous page
- `end_date` (`string`, optional): Inclusive upper bound on the movement accounting date (ISO 8601 date or date-time)
- `limit` (`integer`, optional): Maximum number of movements per page. Defaults to 50, maximum 200. Default: `50`.
- `start_date` (`string`, optional): Inclusive lower bound on the movement accounting date (ISO 8601 date or date-time)

Example input: `{"id":"<id>"}`

### `holdedmcp_list_banking_accounts`

List banking accounts · Read-only

**Required permission:** `accounting:banks.read`

Returns a cursor-paginated list of banking accounts in your account.

Use `archived=true` to retrieve only archived accounts, `archived=false` for only active accounts, or omit the parameter to return both. Use the `cursor` returned by `ApiListResponse` to fetch the next page; `limit` defaults to the standard list page size and is capped at the API-wide maximum.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `archived` (`boolean`, optional): Filter by archived state. `true` returns only archived accounts, `false` returns only active accounts. Omit to return both (default).
- `cursor` (`string`, optional): Opaque cursor for pagination
- `limit` (`integer`, optional): Maximum number of banking accounts to return per page. Default: `50`.
- `type` (`string`, optional): Filter banking accounts by account type. Allowed values: `bank` (bank account), `card` (credit or debit card), `gateway` (payment gateway such as Stripe or PayPal), `cash` (cash register / petty cash). One of: `bank`, `card`, `gateway`, `cash`.

Example input: `{}`

### `holdedmcp_list_booking_locations`

List all booking locations · Read-only

**Required permission:** `crm:booking-services.read`

Returns all booking locations for your account. Locations are not paginated — all are returned in a single response.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `holdedmcp_list_bookings`

List all bookings · Read-only

**Required permission:** `crm:booking-services.read`

Returns a paginated list of bookings for your account.

Results are cursor-paginated: the response includes a `cursor` value and a `has_more` boolean. Default page size is 50, configurable up to 200 with `limit`.

You can filter by date range and service.

To query bookings for a single day pass the same date for both `start_date` and `end_date` (e.g. `start_date=2024-01-15&end_date=2024-01-15`). A date-only value for `end_date` automatically includes the full day up to 23:59:59.

When `items` is an empty array and `has_more` is false, there are no bookings matching the given criteria — this is a definitive result, do not retry with different parameters.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `cursor` (`string`, optional): Opaque cursor for pagination
- `end_date` (`string`, optional): Filter bookings up to this date (inclusive). Accepts ISO 8601 datetime (e.g. 2024-12-31T23:59:59+00:00) or date only (e.g. 2024-12-31). A date-only value includes the full day up to 23:59:59.
- `limit` (`integer`, optional): Number of items per page. Default: `50`.
- `service_id` (`string`, optional): Filter by service identifier
- `sort_field` (`string`, optional): Field to sort by. One of: `start_time`, `created_at`.
- `sort_order` (`string`, optional): Sort direction. One of: `asc`, `desc`.
- `start_date` (`string`, optional): Filter bookings starting from this date. Accepts ISO 8601 datetime (e.g. 2024-01-01T00:00:00+00:00) or date only (e.g. 2024-01-01)

Example input: `{}`

### `holdedmcp_list_business_plan_goals`

List business plan goals vs actuals · Read-only

**Required permission:** `accounting:cashflow.read`

Returns Business Plan goals for the authenticated account, one row per accounting account, broken down by month. Each month carries both the `goal` (the target the account set for itself) and the `actual` (what really happened, from the ledger) for the same period, so goal and reality can be compared without a second call.

Use when the user wants to compare a client's Business Plan targets against what actually happened, e.g. to prepare a monthly review.

Only `tree=profit_and_loss` is supported today; Cashflow goals are not available through this endpoint yet.

`start_date` must not be after `end_date`, and the range between them must not exceed 36 months.

Returns an empty list, not an error, when the account has no Business Plan configured or no goals exist in the requested range.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `end_date` (`string`, required): End of the reporting range (inclusive — any day within the last desired month)
- `start_date` (`string`, required): Start of the reporting range (inclusive, snapped to the first day of its month)
- `tree` (`string`, required): Goal tree to query. Only "profit_and_loss" is supported today. One of: `profit_and_loss`.
- `accounting_account_id` (`string`, optional): Restrict the response to a single accounting account. Omit to return every account in the tree.
- `cursor` (`string`, optional): Opaque cursor for pagination
- `limit` (`integer`, optional): Maximum number of accounting accounts to return per page. Defaults to 50, maximum 200. Default: `50`.
- `only_with_goals` (`boolean`, optional): When true, excludes accounting accounts with no goal set anywhere in the requested range. Without it, a chart of accounts with few goals mostly returns rows with goal: null — this filters those out before pagination, so pages are not spent on empty accounts. Default: `false`.
- `scenario_id` (`string`, optional): Business Plan scenario to query — see GET /api/v2/business-plan/scenarios. Defaults to the account's main scenario.

Example input: `{"end_date":"<end_date>","start_date":"<start_date>","tree":"profit_and_loss"}`

### `holdedmcp_list_business_plan_scenarios`

List business plan scenarios · Read-only

**Required permission:** `accounting:cashflow.read`

Returns the Business Plan scenarios configured for the authenticated account. A Business Plan can hold several scenarios, each with its own goals; use the returned `id` as `scenario_id` when calling the goals endpoint.

Use when the user asks which business plan scenarios exist, or before requesting goals for a scenario other than the main one.

Returns an empty list, not an error, when the account has no Business Plan configured yet.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `holdedmcp_list_contact_attachments`

List contact attachments · Read-only

**Required permission:** `contacts:contacts.read`

Returns the list of file names attached to a contact.

Each item is a plain file name string (e.g. `invoice.pdf`). Use the file name to download individual files via the Get attachment endpoint.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `contactId` (`string`, required): Contact ID (24-char hex ObjectId)

Example input: `{"contactId":"<contactId>"}`

### `holdedmcp_list_contact_groups`

List contact groups · Read-only

**Required permission:** `contacts:contacts.read`

Returns a paginated list of contact groups.

Results are cursor-paginated: the response includes a `cursor` value and a `has_more` boolean. Pass the `cursor` to the next request to fetch the following page.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `cursor` (`string`, optional): Pagination cursor returned by the previous response. Omit to start from the first page.
- `limit` (`integer`, optional): Maximum number of items to return. Default: `50`.

Example input: `{}`

### `holdedmcp_list_contact_shipping_addresses`

List the shipping addresses of a contact · Read-only

**Required permission:** `contacts:contacts.read`

Returns the shipping addresses of a contact. Each one carries the `shipping_id` that identifies it.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `contactId` (`string`, required): Contact ID (24-char hex ObjectId)

Example input: `{"contactId":"<contactId>"}`

### `holdedmcp_list_contacts`

List contacts · Read-only

**Required permission:** `contacts:contacts.read`

Returns a paginated list of contacts in your account.

You can filter by `phone`, `mobile`, `email`, `code`, or `custom_id` to find specific contacts. All filters use exact matching. Results are cursor-paginated: the response includes a `cursor` value and a `has_more` boolean. Pass the `cursor` to the next request to fetch the following page.

The default page size is 25 contacts. You can increase it up to 100 using the `limit` parameter.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `code` (`string`, optional): Filter contacts by exact tax identification number (NIF/CIF)
- `cursor` (`string`, optional): Opaque cursor for pagination. Use the value returned in the previous response to fetch the next page.
- `custom_id` (`string`, optional): Filter contacts by your external reference identifier
- `email` (`string`, optional): Filter contacts by exact email address match
- `limit` (`integer`, optional): Maximum number of contacts to return per page. Defaults to 25, maximum 100. Default: `50`.
- `mobile` (`string`, optional): Filter contacts by exact mobile number match
- `phone` (`string`, optional): Filter contacts by exact phone number match

Example input: `{}`

### `holdedmcp_list_credit_note_attachments`

List credit note attachments · Read-only

**Required permission:** `sales:invoices.read`

Returns the list of files attached to a credit note.

Each attachment includes its identifier, file name, and size.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `creditNoteId` (`string`, required): Credit Note ID (24-char hex ObjectId)

Example input: `{"creditNoteId":"<creditNoteId>"}`

### `holdedmcp_list_credit_notes`

List credit notes · Read-only

**Required permission:** `sales:invoices.read`

Returns a paginated list of credit notes in your account.

Results are cursor-paginated: the response includes a `cursor` value and a `has_more` boolean. Pass the `cursor` to the next request to fetch the following page.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `approval_status` (`string`, optional): Filter by approval status. Use "draft" for unapproved credit notes, "approved" for approved ones. One of: `draft`, `approved`.
- `contact_id` (`string`, optional): Filter credit notes by contact identifier
- `cursor` (`string`, optional): Pagination cursor returned by the previous response. Omit to start from the first page.
- `end_date` (`string`, optional): Return only credit notes issued on or before this date. Accepts ISO 8601 datetime (e.g. 2026-12-31T23:59:59+00:00) or date only (e.g. 2026-12-31). A date without a time (YYYY-MM-DD) includes that whole day; a value with a time is applied exactly as sent.
- `limit` (`integer`, optional): Maximum number of credit notes to return per page. Defaults to 50, maximum 200. Default: `50`.
- `sort` (`string`, optional): Sort order for results. Prefix with - for descending. One of: `date`, `-date`, `dueDate`, `-dueDate`, `total`, `-total`.
- `start_date` (`string`, optional): Return only credit notes issued on or after this date. Accepts ISO 8601 datetime (e.g. 2026-01-01T00:00:00+00:00) or date only (e.g. 2026-01-01)
- `status` (`string`, optional): Filter credit notes by payment status. "pending" matches credit notes with no payment recorded and "partial" those paid in part; both still owe a balance. "outstanding" returns the two together, and "overdue" narrows that set to a due date already past. One of: `pending`, `completed`, `partial`, `cancelled`, `failed`, `overdue`, `outstanding`.

Example input: `{}`

### `holdedmcp_list_employee_times`

List time entries for an employee · Read-only

**Required permission:** `team:time-tracking.read`

Returns a paginated list of time tracking entries for a specific employee.

Results are cursor-paginated: the response includes a `cursor` value and a `has_more` boolean. Pass the `cursor` to the next request to fetch the following page.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `employeeId` (`string`, required): Employee ID
- `cursor` (`string`, optional): Opaque cursor for pagination. Use the value returned in the previous response to fetch the next page.
- `endDate` (`string`, optional): Filter entries starting on or before this date (ISO 8601, YYYY-MM-DD)
- `limit` (`integer`, optional): Maximum number of items to return. Default: `50`.
- `startDate` (`string`, optional): Filter entries starting on or after this date (ISO 8601, YYYY-MM-DD)

Example input: `{"employeeId":"<employeeId>"}`

### `holdedmcp_list_employees`

List all employees · Read-only

**Required permission:** `team:employees.read`

Returns a paginated list of employees.

Results are paginated with `limit` and `cursor` parameters. Default page size is 50, configurable up to 200.

When resolving an employee the user named, match against BOTH first name and surname, accent- and case-insensitively, and gather ALL matches — never stop at the first result. If more than one employee matches, present the complete list of matches (name plus email or employee code) and let the user choose; never pick one on your own.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `cursor` (`string`, optional): Pagination cursor returned by the previous response. Omit to start from the first page.
- `limit` (`integer`, optional): Maximum number of items to return. Default: `50`.
- `search` (`string`, optional): Server-side free-text filter over name, surname, email and code, ignoring case and accents (e.g. "pablo diaz" matches "Pablo Díaz"). ALWAYS use this parameter to resolve an employee the user named — pass the user's exact text. Do NOT filter by name yourself client-side (jmespath/contains): that only sees the current page and cannot fold accents, so it misses matches and picks the wrong person. This search scans ALL employees and returns every match; if more than one comes back, present them and let the user choose.

Example input: `{}`

### `holdedmcp_list_estimate_attachments`

List estimate attachments · Read-only

**Required permission:** `sales:estimates.read`

Returns the list of files attached to an estimate.

Each attachment includes its identifier, file name, and size.

A signed-in user with limited access to estimates only reaches the estimates assigned to them or created by them; any other is reported as not found. API keys still need the scope of this endpoint.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `estimateId` (`string`, required): Estimate ID (24-char hex ObjectId)

Example input: `{"estimateId":"<estimateId>"}`

### `holdedmcp_list_estimates`

List estimates · Read-only

**Required permission:** `sales:estimates.read`

Returns a paginated list of estimates in your account.

Results are cursor-paginated: the response includes a `cursor` value and a `has_more` boolean. Pass the `cursor` to the next request to fetch the following page.

A signed-in user with limited access to estimates gets only the estimates assigned to them. API keys still need the scope of this endpoint.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `approval_status` (`string`, optional): Filter by approval status. Use "draft" for unapproved estimates, "approved" for approved ones. One of: `draft`, `approved`.
- `billing_status` (`string`, optional): Filter estimates by their invoicing status (the **Billed** column in Sales > Quotes). Independent of `status`. One of: `pending`, `paid`, `partial`.
- `contact_id` (`string`, optional): Filter estimates by contact identifier
- `cursor` (`string`, optional): Pagination cursor returned by the previous response. Omit to start from the first page.
- `end_date` (`string`, optional): Filter up to this date. Accepts ISO 8601 datetime or date only (YYYY-MM-DD). A date without a time (YYYY-MM-DD) includes that whole day; a value with a time is applied exactly as sent.
- `limit` (`integer`, optional): Maximum number of estimates to return per page. Defaults to 50, maximum 200. Default: `50`.
- `sort` (`string`, optional): Sort order for results. Prefix with - for descending. One of: `date`, `-date`, `dueDate`, `-dueDate`, `total`, `-total`.
- `start_date` (`string`, optional): Filter from this date. Accepts ISO 8601 datetime or date only (YYYY-MM-DD)
- `status` (`string`, optional): Filter estimates by the customer's decision on the quote: "completed" is shown as **Accepted** in the app, "failed" as **Rejected**, "pending" as awaiting an answer. Independent of `billing_status`. One of: `pending`, `completed`, `partial`, `cancelled`, `failed`, `overdue`.

Example input: `{}`

### `holdedmcp_list_events`

List all events · Read-only

**Required permission:** `crm:events.read`

Returns a paginated list of CRM events.

Results are cursor-paginated: the response includes a `cursor` value and a `has_more` boolean. Default page size is 50, configurable up to 100 with `limit`.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `cursor` (`string`, optional): Pagination cursor from a previous response
- `limit` (`integer`, optional): Maximum number of items to return. Default: `50`.

Example input: `{}`

### `holdedmcp_list_expenses_accounts`

List all expenses accounts · Read-only

**Required permission:** `accounting:expenses-accounts.read`

Returns the list of all expenses accounts in your account.

The response includes both active and archived expenses accounts. Use expenses account IDs when setting defaults on contacts or assigning costs to purchase lines.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `holdedmcp_list_funnels`

List all funnels · Read-only

**Required permission:** `crm:funnels.read`

Returns a paginated list of funnels.

Results are paginated with `limit` and `cursor` parameters. Default page size is 50, configurable up to 200.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `cursor` (`string`, optional): Pagination cursor returned by the previous response. Omit to start from the first page.
- `limit` (`integer`, optional): Maximum number of items to return. Default: `50`.

Example input: `{}`

### `holdedmcp_list_inbox`

List inbox incoming documents · Read-only

**Required permission:** `inbox:inbox.read`

Returns a cursor-paginated list of incoming documents in your inbox.

Optional filters: `status` (comma-separated list of statuses: `new`, `processing`, `error`, `pending_to_verify`, `done`, `rejected`), `start_date`/`end_date` (ISO 8601 date range on creation date), `user_id` (comma-separated user IDs).

Results are ordered by creation date descending (newest first). Pass `cursor` from the previous response to fetch the next page.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `cursor` (`string`, optional): Opaque pagination cursor from the previous response
- `end_date` (`string`, optional): Filter documents created on or before this ISO 8601 date
- `limit` (`integer`, optional): Number of items per page. Defaults to 50, maximum 200. Default: `50`.
- `start_date` (`string`, optional): Filter documents created on or after this ISO 8601 date
- `status` (`string`, optional): Comma-separated statuses to filter by (e.g. pending_to_verify,done)
- `user_id` (`string`, optional): Comma-separated user IDs to filter by uploader

Example input: `{}`

### `holdedmcp_list_invoice_attachments`

List invoice attachments · Read-only

**Required permission:** `sales:invoices.read`

Returns the list of files attached to an invoice.

Each attachment includes its identifier, file name, and size. Use the attachment ID to download or delete individual files.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `invoiceId` (`string`, required): Invoice ID (24-char hex ObjectId)

Example input: `{"invoiceId":"<invoiceId>"}`

### `holdedmcp_list_invoice_recurrings`

List recurring invoices · Read-only

**Required permission:** `sales:invoices.read`

Returns a paginated list of recurring invoices in your account.

Results are cursor-paginated: the response includes a `cursor` value and a `has_more` boolean. Pass the `cursor` to the next request to fetch the following page.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `contact_id` (`string`, optional): No description.
- `cursor` (`string`, optional): Pagination cursor returned by the previous response. Omit to start from the first page.
- `limit` (`integer`, optional): Maximum number of recurring invoices to return per page. Defaults to 50, maximum 200. Default: `50`.

Example input: `{}`

### `holdedmcp_list_invoices`

List invoices · Read-only

**Required permission:** `sales:invoices.read`

Returns a paginated list of invoices in your account.

Results are cursor-paginated: the response includes a `cursor` value and a `has_more` boolean. Pass the `cursor` to the next request to fetch the following page.

The list includes payment summary fields (`payments_total`, `payments_pending`) so you can display outstanding balances without fetching each invoice individually.

Payment status: an invoice with no payment recorded is `pending`, one collected in part is `partial`, and both still owe a balance. `status=outstanding` returns the two together; `status=overdue` narrows that set to invoices whose due date has passed.

Date parameters (`start_date`, `end_date`) filter by ISSUE date; `due_date_start`/`due_date_end` filter by DUE date; all are optional and unfiltered means the whole account. Use `status=overdue` to list all unpaid invoices (pending or partial) whose due date has passed. All date parameters accept RFC 3339 (e.g. `2026-01-01T00:00:00+00:00`) or date-only (`2026-01-01`).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `accounting_account_num` (`integer`, optional): Filter invoices that have at least one line booked to this accounting account, identified by its numeric PGC code (e.g. 70000000 for "Ventas de mercaderias"). Only the numeric code is accepted; a non-numeric value returns 400. Returns 404 if no accounting account with that number exists in the tenant. When called through the AI assistant, an account name is resolved to its code automatically before the request reaches this endpoint.
- `approval_status` (`string`, optional): Filter by approval status. Use "draft" for unapproved invoices, "approved" for approved ones. One of: `draft`, `approved`.
- `contact_id` (`string`, optional): Filter invoices by contact identifier
- `cursor` (`string`, optional): Pagination cursor returned by the previous response. Omit to start from the first page.
- `due_date_end` (`string`, optional): Filter invoices whose DUE date is on or before this date. Accepts ISO 8601 datetime or date only (e.g. 2026-03-12). A date without a time (YYYY-MM-DD) includes that whole day; a value with a time is applied exactly as sent.
- `due_date_start` (`string`, optional): Filter invoices whose DUE date is on or after this date. Accepts ISO 8601 datetime or date only (e.g. 2026-01-01).
- `end_date` (`string`, optional): Filter invoices up to this date. Accepts ISO 8601 datetime (e.g. 2026-12-31T23:59:59+00:00) or date only (e.g. 2026-12-31). A date without a time (YYYY-MM-DD) includes that whole day; a value with a time is applied exactly as sent.
- `limit` (`integer`, optional): Maximum number of invoices to return per page. Defaults to 50, maximum 200. Default: `50`.
- `sort` (`string`, optional): Sort order for results. Prefix with - for descending. One of: `date`, `-date`, `dueDate`, `-dueDate`, `total`, `-total`.
- `start_date` (`string`, optional): Filter invoices from this date. Accepts ISO 8601 datetime (e.g. 2026-01-01T00:00:00+00:00) or date only (e.g. 2026-01-01)
- `status` (`string`, optional): Filter invoices by payment status. "pending" matches invoices with no payment recorded and "partial" those collected in part; both still owe a balance. "outstanding" returns the two together, "overdue" narrows that set to a due date already past, and "completed" returns the settled ones. One of: `pending`, `completed`, `partial`, `cancelled`, `failed`, `overdue`, `outstanding`.

Example input: `{}`

### `holdedmcp_list_invoicing_forecasts`

List invoicing forecasts · Read-only

**Required permission:** `accounting:cashflow.read`

Returns a cursor-paginated list of invoicing forecasts for the authenticated account. Each forecast represents a scheduled payment milestone linked to an estimate or purchase order.

Use when the user asks to list, search or browse cashflow invoicing forecasts, optionally filtered by the source document.

Distinct from the Remittances endpoint: invoicing forecasts model future payment milestones on open documents, not batch SEPA transfers.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `cursor` (`string`, optional): Opaque cursor for pagination
- `document_id` (`string`, optional): Filter by the related estimate or purchase order identifier
- `document_type` (`string`, optional): Filter by related document type. One of: `estimate`, `order`.
- `limit` (`integer`, optional): Maximum number of forecasts to return per page. Defaults to 50, maximum 200. Default: `50`.

Example input: `{}`

### `holdedmcp_list_leads`

List all leads · Read-only

**Required permission:** `crm:deals.read`

Returns a paginated list of leads.

Results are paginated with `limit` and `cursor` parameters. Default page size is 50, configurable up to 200.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `cursor` (`string`, optional): Pagination cursor returned by the previous response. Omit to start from the first page.
- `limit` (`integer`, optional): Maximum number of items to return. Default: `50`.

Example input: `{}`

### `holdedmcp_list_ledger_entries`

List ledger entries · Read-only

**Required permission:** `accounting:daily-ledger.read`

Returns a paginated list of ledger entries in your accounting books.

Results are cursor-paginated: the response includes a `cursor` value and a `has_more` boolean. Default page size is 50, configurable up to 200 with `limit`.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `end_date` (`string`, required): Filter entries up to and including this date
- `start_date` (`string`, required): Filter entries from this date onward
- `account` (`integer`, optional): Filter entries by accounting account number (e.g. 70500001). Returns only entries posted to this exact account.
- `cursor` (`string`, optional): Opaque cursor for pagination
- `limit` (`integer`, optional): Maximum number of ledger entries to return per page. Defaults to 50, maximum 200. Default: `50`.

Example input: `{"end_date":"<end_date>","start_date":"<start_date>"}`

### `holdedmcp_list_numbering_series`

List numbering series by document type · Read-only

**Required permission:** `sales:invoicing-settings.read`

Returns the list of all numbering series configured in your account, grouped by document type.

Use this endpoint to discover available series before creating documents with a specific `number_line_id`.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `type` (`string`, required): Document type for the numbering series. One of: `invoice`, `salesreceipt`, `creditnote`, `estimate`, `proform`, `waybill`, `salesorder`, `purchase`, `purchaserefund`, `order`, `purchaseshipment`, `receiptnote`, `productionorder`.

Example input: `{"type":"invoice"}`

### `holdedmcp_list_payment_methods`

List payment methods · Read-only

**Required permission:** `sales:invoicing-settings.read`

Returns all payment methods configured for the account, including both manual methods (bank transfer, cash on delivery, cheque) and online connected providers (Stripe, PayPal, GoCardless, Square, Monei, Wallet). Use when the user asks which payment methods they have set up or how they collect payments from customers. Different from the payments endpoint which returns individual transactions — this returns the configured methods themselves. A signed-in user who can write estimates reads them too, as the estimate editor shows them every payment method. API keys still need the scope of this endpoint.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `cursor` (`string`, optional): Opaque cursor for pagination. Pass the value returned in the previous response to fetch the next page.
- `limit` (`integer`, optional): Maximum number of payment methods to return per page. Defaults to 50, maximum 200. Default: `50`.

Example input: `{}`

### `holdedmcp_list_payments`

List payments and collections · Read-only

**Required permission:** `accounting:payments.read`

Lists the money received (cobros) and paid out (pagos) recorded in the account, newest first, with its date, amount, banking account and the document it settles. Use it when the user asks about a collection or a payment, e.g. "show me the last collection" (último cobro) or "payments made this month" (pagos de este mes): pass `type` collection or payment, and narrow by dates, banking account or document. Not the payment methods configured in the account (métodos de pago: transfer, card, cash), which have their own list. It lists recorded movements, not what is still owed: for supplier bills or invoices pending payment, list purchases or invoices with `status=outstanding` and their due dates.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `banking_account_id` (`string`, optional): Filter payments linked to this banking account
- `cursor` (`string`, optional): Opaque cursor for pagination
- `document_id` (`string`, optional): Filter payments linked to this document. Combine with document_type to match a specific document
- `document_type` (`string`, optional): Filter payments by document type. Can be used alone, or combined with document_id to match a specific document
- `end_date` (`string`, optional): Filter payments up to this date. Accepts ISO 8601 datetime (e.g. 2024-12-31T23:59:59+00:00) or date only (e.g. 2024-12-31)
- `limit` (`integer`, optional): Maximum number of payments to return per page. Default: `50`.
- `start_date` (`string`, optional): Filter payments from this date onward. Accepts ISO 8601 datetime (e.g. 2024-01-01T00:00:00+00:00) or date only (e.g. 2024-01-01)
- `type` (`string`, optional): Filter by direction: "collection" for money received (cobro), "payment" for money paid out (pago). One of: `collection`, `payment`.

Example input: `{}`

### `holdedmcp_list_price_lists`

List price lists · Read-only

**Required permission:** `inventory:rates.read`

Lists all price lists (rates) of the authenticated account.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `holdedmcp_list_product_images`

List product images · Read-only

**Required permission:** `inventory:products.read`

Retrieves the list of all images associated with a product.

Each image includes its URL, available sizes (original, large, medium, small, thumbnail), position in the gallery, and description.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `productId` (`string`, required): Product ID (24-char hex ObjectId)

Example input: `{"productId":"<productId>"}`

### `holdedmcp_list_production_orders`

List production orders · Read-only

**Required permission:** `inventory:production-orders.read`

Returns a paginated list of production orders in your account.

Results are cursor-paginated: the response includes a `cursor` value and a `has_more` boolean. Pass the `cursor` to the next request to fetch the following page.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `cursor` (`string`, optional): Pagination cursor returned by the previous response. Omit to start from the first page.
- `limit` (`integer`, optional): Maximum number of production orders to return per page. Defaults to 50, maximum 200. Default: `50`.

Example input: `{}`

### `holdedmcp_list_products`

List all products · Read-only

**Required permission:** `inventory:products.read`

Returns a cursor-paginated list of products in your catalog.

Default page size is 50, configurable up to 200 with `limit`. Pass the returned `cursor` value to retrieve the next page.

Provide `name` to return only products whose name matches, searched across your whole catalog. Pagination works the same way (`cursor` + `limit`).

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `cursor` (`string`, optional): Opaque cursor for pagination
- `limit` (`integer`, optional): Number of products per page. Defaults to 50, maximum 200. Default: `50`.
- `name` (`string`, optional): Search products by name across your whole catalog. Optional.

Example input: `{}`

### `holdedmcp_list_proforma_attachments`

List proforma attachments · Read-only

**Required permission:** `sales:proforms.read`

Returns the list of files attached to a proforma invoice.

Each attachment includes its identifier, file name, and size.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `proformaId` (`string`, required): Proforma ID (24-char hex ObjectId)

Example input: `{"proformaId":"<proformaId>"}`

### `holdedmcp_list_proformas`

List proformas · Read-only

**Required permission:** `sales:proforms.read`

Returns a paginated list of proforma invoices in your account.

Results are cursor-paginated: the response includes a `cursor` value and a `has_more` boolean. Pass the `cursor` to the next request to fetch the following page.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `contact_id` (`string`, optional): Filter proformas by contact identifier
- `cursor` (`string`, optional): Pagination cursor returned by the previous response. Omit to start from the first page.
- `end_date` (`string`, optional): Filter up to this date. Accepts ISO 8601 datetime or date only (YYYY-MM-DD). A date without a time (YYYY-MM-DD) includes that whole day; a value with a time is applied exactly as sent.
- `limit` (`integer`, optional): Maximum number of proformas to return per page. Defaults to 50, maximum 200. Default: `50`.
- `sort` (`string`, optional): Sort order for results. Prefix with - for descending. One of: `date`, `-date`, `dueDate`, `-dueDate`, `total`, `-total`.
- `start_date` (`string`, optional): Filter from this date. Accepts ISO 8601 datetime or date only (YYYY-MM-DD)
- `status` (`string`, optional): Filter proformas by status. One of: `pending`, `completed`, `partial`, `cancelled`, `failed`, `overdue`.

Example input: `{}`

### `holdedmcp_list_project_times`

List time entries for a project · Read-only

**Required permission:** `projects:projects.read`

Returns a paginated list of time tracking entries for a specific project.

Results are paginated with `limit` and `cursor` parameters.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `projectId` (`string`, required): Project ID
- `cursor` (`string`, optional): Pagination cursor returned by the previous response. Omit to start from the first page.
- `employee_id` (`string`, optional): Filter by employee: only the entries logged by the employee's linked user are returned, and an employee with no linked user has none. Takes an employee id, not a user id (neither an entry's `user_id` nor an employee's `holded_user_id`). An id that is not an employee of the account returns 404.
- `endDate` (`string`, optional): Filter entries on or before this date (ISO 8601, YYYY-MM-DD)
- `limit` (`integer`, optional): Maximum number of items to return. Default: `50`.
- `startDate` (`string`, optional): Filter entries on or after this date (ISO 8601, YYYY-MM-DD)

Example input: `{"projectId":"<projectId>"}`

### `holdedmcp_list_projects`

List projects · Read-only

**Required permission:** `projects:projects.read`

Returns a paginated list of projects. Supports filtering by status to retrieve active (in-progress), completed, cancelled, waiting, or budgeted projects.

Use `status=active` or `status=in_progress` to list only active projects. Results are paginated with `limit` and `cursor` parameters. Default page size is 50, configurable up to 200.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `cursor` (`string`, optional): Pagination cursor returned by the previous response. Omit to start from the first page.
- `limit` (`integer`, optional): Maximum number of items to return. Default: `50`.
- `status` (`string`, optional): Filter projects by status. Use "active" or "in_progress" to retrieve only active/in-progress projects. One of: `active`, `in_progress`, `completed`, `cancelled`, `waiting`, `budgeted`.
- `text` (`string`, optional): Search on the project name, case and accents aside: every word (up to eight) must appear.

Example input: `{}`

### `holdedmcp_list_purchase_attachments`

List purchase attachments · Read-only

**Required permission:** `accounting:purchases.read`

Returns the list of files attached to a purchase invoice.

Each attachment includes its identifier, file name, and size.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `purchaseId` (`string`, required): Purchase ID (24-char hex ObjectId)

Example input: `{"purchaseId":"<purchaseId>"}`

### `holdedmcp_list_purchase_order_attachments`

List purchase order attachments · Read-only

**Required permission:** `inventory:purchase-orders.read`

Returns the list of files attached to a purchase order.

Each attachment includes its identifier, file name, and size.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `purchaseOrderId` (`string`, required): Purchase Order ID (24-char hex ObjectId)

Example input: `{"purchaseOrderId":"<purchaseOrderId>"}`

### `holdedmcp_list_purchase_orders`

List purchase orders · Read-only

**Required permission:** `inventory:purchase-orders.read`

Returns a paginated list of purchase orders in your account.

Results are cursor-paginated: the response includes a `cursor` value and a `has_more` boolean. Pass the `cursor` to the next request to fetch the following page.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `contact_id` (`string`, optional): Filter by contact identifier
- `cursor` (`string`, optional): Pagination cursor returned by the previous response. Omit to start from the first page.
- `end_date` (`string`, optional): Return only purchase orders issued on or before this date. Accepts ISO 8601 datetime or date only (YYYY-MM-DD). A date without a time (YYYY-MM-DD) includes that whole day; a value with a time is applied exactly as sent.
- `limit` (`integer`, optional): Maximum number of purchase orders to return per page. Defaults to 50, maximum 200. Default: `50`.
- `sort` (`string`, optional): Sort field and direction. Prefix with - for descending order. One of: `date`, `-date`, `dueDate`, `-dueDate`, `total`, `-total`.
- `start_date` (`string`, optional): Return only purchase orders issued on or after this date. Accepts ISO 8601 datetime or date only (YYYY-MM-DD)
- `status` (`string`, optional): Filter by payment status. One of: `pending`, `completed`, `partial`, `cancelled`, `failed`, `overdue`.

Example input: `{}`

### `holdedmcp_list_purchase_refunds`

List purchase refunds · Read-only

**Required permission:** `accounting:purchases.read`

Returns a paginated list of purchase refunds in your account.

Results are cursor-paginated: the response includes a `cursor` value and a `has_more` boolean. Pass the `cursor` to the next request to fetch the following page. Results are ordered by creation time, newest first.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `contact_id` (`string`, optional): Filter by contact identifier
- `cursor` (`string`, optional): Pagination cursor returned by the previous response. Omit to start from the first page.
- `end_date` (`string`, optional): Return only purchase refunds issued on or before this date. Accepts ISO 8601 datetime or date only (YYYY-MM-DD). A date without a time (YYYY-MM-DD) includes that whole day; a value with a time is applied exactly as sent.
- `limit` (`integer`, optional): Maximum number of purchase refunds to return per page. Defaults to 50, maximum 200. Default: `50`.
- `start_date` (`string`, optional): Return only purchase refunds issued on or after this date. Accepts ISO 8601 datetime or date only (YYYY-MM-DD)
- `status` (`string`, optional): Filter by payment status. "pending" matches purchase refunds with no payment recorded and "partial" those paid in part; both still owe a balance. "outstanding" returns the two together, "overdue" narrows that set to a due date already past, and "completed" returns the settled ones. One of: `pending`, `completed`, `partial`, `cancelled`, `failed`, `overdue`, `outstanding`.

Example input: `{}`

### `holdedmcp_list_purchase_shipments`

List purchase shipments · Read-only

**Required permission:** `inventory:purchase-shipments.read`

Returns a paginated list of purchase shipments in your account.

Results are cursor-paginated: the response includes a `cursor` value and a `has_more` boolean. Pass the `cursor` to the next request to fetch the following page.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `contact_id` (`string`, optional): Filter by contact (supplier) identifier
- `cursor` (`string`, optional): Pagination cursor returned by the previous response. Omit to start from the first page.
- `end_date` (`string`, optional): Filter purchase shipments issued on or before this date. Accepts ISO 8601 datetime or date only (YYYY-MM-DD). A date without a time (YYYY-MM-DD) includes that whole day; a value with a time is applied exactly as sent.
- `limit` (`integer`, optional): Maximum number of purchase shipments to return per page. Defaults to 50, maximum 200. Default: `50`.
- `sort` (`string`, optional): Sort order for results. Prefix with - for descending. One of: `date`, `-date`, `dueDate`, `-dueDate`, `total`, `-total`.
- `start_date` (`string`, optional): Filter purchase shipments issued on or after this date. Accepts ISO 8601 datetime or date only (YYYY-MM-DD)
- `status` (`string`, optional): Filter by status. "pending" matches purchase shipments with no payment recorded and "partial" those paid in part; both still owe a balance. "outstanding" returns the two together, and "overdue" narrows that set to a due date already past. One of: `pending`, `completed`, `partial`, `cancelled`, `failed`, `overdue`, `outstanding`.

Example input: `{}`

### `holdedmcp_list_purchases`

List purchases · Read-only

**Required permission:** `accounting:purchases.read`

Returns a paginated list of purchase invoices in your account.

Results are cursor-paginated: the response includes a `cursor` value and a `has_more` boolean. Pass the `cursor` to the next request to fetch the following page.

Payment status: a purchase with no payment recorded is `pending`, one paid in part is `partial`, and both still owe a balance. `status=outstanding` returns the two together; `status=overdue` narrows that set to purchases whose due date has passed.

Date parameters (`start_date`, `end_date`) filter by ISSUE date; `due_date_start`/`due_date_end` filter by DUE date; all are optional and unfiltered means the whole account. `status=overdue` is applied server-side across the whole account and matches pending and partially paid purchases whose due date has passed.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `approval_status` (`string`, optional): Filter by approval status. Use "draft" for unapproved purchases, "approved" for approved ones. One of: `draft`, `approved`.
- `contact_id` (`string`, optional): Filter by contact identifier
- `cursor` (`string`, optional): Pagination cursor returned by the previous response. Omit to start from the first page.
- `due_date_end` (`string`, optional): Return only purchases whose DUE date is on or before this date. Combine with status=pending to list overdue purchases (set this to the cut-off date). Accepts ISO 8601 datetime or date only (YYYY-MM-DD). A date without a time (YYYY-MM-DD) includes that whole day; a value with a time is applied exactly as sent.
- `due_date_start` (`string`, optional): Return only purchases whose DUE date is on or after this date. Accepts ISO 8601 datetime or date only (YYYY-MM-DD).
- `end_date` (`string`, optional): Return only purchases issued on or before this date. Accepts ISO 8601 datetime or date only (YYYY-MM-DD). A date without a time (YYYY-MM-DD) includes that whole day; a value with a time is applied exactly as sent.
- `limit` (`integer`, optional): Maximum number of purchases to return per page. Defaults to 50, maximum 200. Default: `50`.
- `sort` (`string`, optional): Sort field and direction. Prefix with - for descending order. One of: `date`, `-date`, `dueDate`, `-dueDate`, `total`, `-total`.
- `start_date` (`string`, optional): Return only purchases issued on or after this date. Accepts ISO 8601 datetime or date only (YYYY-MM-DD)
- `status` (`string`, optional): Filter by payment status. "pending" matches purchases with no payment recorded and "partial" those paid in part; both still owe a balance. "outstanding" returns the two together, "overdue" narrows that set to a due date already past, and "completed" returns the settled ones. One of: `pending`, `completed`, `partial`, `cancelled`, `failed`, `overdue`, `outstanding`.

Example input: `{}`

### `holdedmcp_list_receipt_note_attachments`

List receipt note attachments · Read-only

**Required permission:** `sales:invoices.read`

Returns the list of files attached to a receipt note.

Each attachment includes its identifier, file name, and size.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `receiptNoteId` (`string`, required): Receipt Note ID (24-char hex ObjectId)

Example input: `{"receiptNoteId":"<receiptNoteId>"}`

### `holdedmcp_list_receipt_notes`

List receipt notes · Read-only

**Required permission:** `sales:invoices.read`

Returns a paginated list of receipt notes for the account.

Use the `cursor` from the response to fetch the next page. Filter by contact, status, or date range.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `approval_status` (`string`, optional): Filter by approval status. Use "draft" for unapproved receipt notes, "approved" for approved ones. One of: `draft`, `approved`.
- `contact_id` (`string`, optional): Filter receipt notes by contact identifier
- `cursor` (`string`, optional): Pagination cursor returned by the previous response. Omit to start from the first page.
- `end_date` (`string`, optional): Filter up to this date. Accepts ISO 8601 datetime (e.g. 2026-12-31T23:59:59+00:00) or date only (e.g. 2026-12-31). A date without a time (YYYY-MM-DD) includes that whole day; a value with a time is applied exactly as sent.
- `limit` (`integer`, optional): Maximum number of receipt notes to return per page. Defaults to 50, maximum 200. Default: `50`.
- `sort` (`string`, optional): Sort order for results. Prefix with - for descending. One of: `date`, `-date`, `dueDate`, `-dueDate`, `total`, `-total`.
- `start_date` (`string`, optional): Filter from this date. Accepts ISO 8601 datetime (e.g. 2026-01-01T00:00:00+00:00) or date only (e.g. 2026-01-01)
- `status` (`string`, optional): Filter receipt notes by status. "pending" matches receipt notes with no payment recorded and "partial" those paid in part; both still owe a balance. "outstanding" returns the two together, and "overdue" narrows that set to a due date already past. One of: `pending`, `completed`, `partial`, `cancelled`, `failed`, `overdue`, `outstanding`.

Example input: `{}`

### `holdedmcp_list_remittances`

List remittances · Read-only

**Required permission:** `accounting:remittances.read`

Returns a paginated list of bank remittances for the account, filterable by direction (`type`), processing `status`, date range, treasury bank account, creditor (SWIFT/IBAN/SEPA id) or issuer. Use when the user asks to list, view, or filter their remittances. This endpoint returns remittance batches only; for individual invoices use the invoices list, and for individual payments not grouped into a remittance use the payments list.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `bank_id` (`string`, optional): Filter by treasury bank account identifier
- `creditor_iban` (`string`, optional): Filter by creditor IBAN
- `creditor_id` (`string`, optional): Filter by SEPA creditor identifier
- `creditor_swift` (`string`, optional): Filter by creditor SWIFT/BIC code (8 or 11 alphanumeric characters)
- `cursor` (`string`, optional): Opaque cursor for pagination
- `end_date` (`string`, optional): Inclusive upper bound on the remittance date (ISO 8601 date, e.g. 2026-12-31)
- `issuer_id` (`string`, optional): Filter by issuer identifier
- `limit` (`integer`, optional): Maximum number of remittances to return per page. Defaults to 25, maximum 100. Default: `50`.
- `start_date` (`string`, optional): Inclusive lower bound on the remittance date (ISO 8601 date, e.g. 2026-01-01)
- `status` (`string`, optional): Filter by processing status. One of: `pending`, `completed`, `processing`.
- `type` (`string`, optional): Filter by remittance direction. `inbound` returns collection remittances received from customers. `outbound` returns payment remittances issued to suppliers. One of: `inbound`, `outbound`.

Example input: `{}`

### `holdedmcp_list_sales_channels`

List all sales channels · Read-only

**Required permission:** `sales:sales-channels.read`

Returns the list of all sales channels in your account.

The response includes both active and archived sales channels. Use sales channel IDs when setting defaults on contacts or assigning revenue to document lines.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `holdedmcp_list_sales_order_attachments`

List sales order attachments · Read-only

**Required permission:** `inventory:sales-orders.read`

Returns the list of files attached to a sales order.

Each attachment includes its identifier, file name, and size.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `salesOrderId` (`string`, required): Sales Order ID (24-char hex ObjectId)

Example input: `{"salesOrderId":"<salesOrderId>"}`

### `holdedmcp_list_sales_orders`

List sales orders · Read-only

**Required permission:** `inventory:sales-orders.read`

Returns a paginated list of sales orders in your account.

Results are cursor-paginated: the response includes a `cursor` value and a `has_more` boolean. Pass the `cursor` to the next request to fetch the following page.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `contact_id` (`string`, optional): Filter by contact identifier
- `cursor` (`string`, optional): Pagination cursor returned by the previous response. Omit to start from the first page.
- `end_date` (`string`, optional): Filter up to this date. Accepts ISO 8601 datetime (e.g. 2026-12-31T23:59:59+00:00) or date only (e.g. 2026-12-31). A date without a time (YYYY-MM-DD) includes that whole day; a value with a time is applied exactly as sent.
- `limit` (`integer`, optional): Maximum number of sales orders to return per page. Defaults to 50, maximum 200. Default: `50`.
- `sort` (`string`, optional): Sort field and direction. One of: `date`, `-date`, `dueDate`, `-dueDate`, `total`, `-total`.
- `start_date` (`string`, optional): Filter from this date. Accepts ISO 8601 datetime (e.g. 2026-01-01T00:00:00+00:00) or date only (e.g. 2026-01-01)
- `status` (`string`, optional): Filter by sales order status. One of: `pending`, `completed`, `partial`, `cancelled`, `failed`, `overdue`.

Example input: `{}`

### `holdedmcp_list_sales_receipt_attachments`

List sales receipt attachments · Read-only

**Required permission:** `sales:receipts.read`

Returns the list of files attached to a sales receipt.

Each attachment includes its identifier, file name, and size.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `salesReceiptId` (`string`, required): Sales Receipt ID (24-char hex ObjectId)

Example input: `{"salesReceiptId":"<salesReceiptId>"}`

### `holdedmcp_list_sales_receipts`

List sales receipts · Read-only

**Required permission:** `sales:receipts.read`

Returns a paginated list of sales receipts in your account.

Results are cursor-paginated: the response includes a `cursor` value and a `has_more` boolean. Pass the `cursor` to the next request to fetch the following page.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `approval_status` (`string`, optional): Filter by approval status. Use "draft" for unapproved sales receipts, "approved" for approved ones. One of: `draft`, `approved`.
- `contact_id` (`string`, optional): Filter by contact identifier
- `cursor` (`string`, optional): Pagination cursor returned by the previous response. Omit to start from the first page.
- `end_date` (`string`, optional): Filter up to this date. Accepts ISO 8601 datetime (e.g. 2026-12-31T23:59:59+00:00) or date only (e.g. 2026-12-31). A date without a time (YYYY-MM-DD) includes that whole day; a value with a time is applied exactly as sent.
- `limit` (`integer`, optional): Maximum number of sales receipts to return per page. Defaults to 50, maximum 200. Default: `50`.
- `sort` (`string`, optional): Sort field and direction. One of: `date`, `-date`, `dueDate`, `-dueDate`, `total`, `-total`.
- `start_date` (`string`, optional): Filter from this date. Accepts ISO 8601 datetime (e.g. 2026-01-01T00:00:00+00:00) or date only (e.g. 2026-01-01)
- `status` (`string`, optional): Filter by sales receipt status. "pending" matches sales receipts with no payment recorded and "partial" those paid in part; both still owe a balance. "outstanding" returns the two together, and "overdue" narrows that set to a due date already past. One of: `pending`, `completed`, `partial`, `cancelled`, `failed`, `overdue`, `outstanding`.

Example input: `{}`

### `holdedmcp_list_services`

List services · Read-only

**Required permission:** `sales:services.read`

Returns a paginated list of services in your catalog.

Results are cursor-paginated. Default page size is 25, configurable up to 100 with `limit`.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `cursor` (`string`, optional): Pagination cursor returned by the previous response. Omit to start from the first page.
- `include_archived` (`boolean`, optional): Whether to include archived services in the results. Default: `false`.
- `limit` (`integer`, optional): Maximum number of services to return per page. Defaults to 25, maximum 100. Default: `50`.

Example input: `{}`

### `holdedmcp_list_tags`

List all tags · Read-only

**Required permission:** `account:tags.read`

Returns the complete list of tags configured in your account.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `holdedmcp_list_tasks`

List all tasks · Read-only

**Required permission:** `projects:projects.read`

Returns a paginated list of tasks.

Use `project_id` for the tasks of one project and `text` to search the task name; both filter server-side across every page. Use `status=open` for the tasks still to do: not in a done status and not archived. Requests authenticated as a user also include the private projects that user is a member of; requests made with an API key list public projects only. Results are paginated with `limit` and `cursor` parameters. Default page size is 50, configurable up to 200.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `cursor` (`string`, optional): Pagination cursor returned by the previous response. Omit to start from the first page.
- `limit` (`integer`, optional): Maximum number of items to return. Default: `50`.
- `project_id` (`string`, optional): Only the tasks of this project. A project the caller cannot see lists nothing.
- `status` (`string`, optional): `open`: tasks not in a done status and not archived. `done`: tasks in a done status, archived ones included. Omit to list every task, archived ones included. One of: `open`, `done`.
- `text` (`string`, optional): Search on the task name: every word (up to eight) must appear, case and accents aside. Filters server-side across every page.

Example input: `{}`

### `holdedmcp_list_taxes`

List all taxes · Read-only

**Required permission:** `accounting:taxes.read`

Returns the complete list of taxes configured in your account. Taxes define the VAT, GST, or other rates applied to document line items.

Use this endpoint to discover available tax IDs before creating documents. Each tax includes its percentage rate, type, and applicable region.

A signed-in user who can write estimates reads them too, as the estimate editor shows them every tax. API keys still need the scope of this endpoint.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `holdedmcp_list_warehouses`

List all warehouses · Read-only

**Required permission:** `inventory:warehouses.read`

Returns the list of all warehouses in your account.

The response includes each warehouse's name, address, and default status. Unlike document endpoints, warehouses are not paginated — all warehouses are returned in a single response.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs: none.

Example input: `{}`

### `holdedmcp_list_waybill_attachments`

List waybill attachments · Read-only

**Required permission:** `inventory:shipments.read`

Returns the list of files attached to a waybill.

Each attachment includes its identifier, file name, and size.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `waybillId` (`string`, required): Waybill ID (24-char hex ObjectId)

Example input: `{"waybillId":"<waybillId>"}`

### `holdedmcp_list_waybills`

List waybills · Read-only

**Required permission:** `inventory:shipments.read`

Returns a paginated list of waybills in your account.

Results are cursor-paginated: the response includes a `cursor` value and a `has_more` boolean. Pass the `cursor` to the next request to fetch the following page.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `contact_id` (`string`, optional): Filter by contact identifier
- `cursor` (`string`, optional): Pagination cursor returned by the previous response. Omit to start from the first page.
- `end_date` (`string`, optional): Filter waybills issued on or before this date. Accepts ISO 8601 datetime or date only (YYYY-MM-DD). A date without a time (YYYY-MM-DD) includes that whole day; a value with a time is applied exactly as sent.
- `limit` (`integer`, optional): Maximum number of waybills to return per page. Defaults to 50, maximum 200. Default: `50`.
- `sort` (`string`, optional): Sort order for results. Prefix with - for descending. One of: `date`, `-date`, `dueDate`, `-dueDate`, `total`, `-total`.
- `start_date` (`string`, optional): Filter waybills issued on or after this date. Accepts ISO 8601 datetime or date only (YYYY-MM-DD)
- `status` (`string`, optional): Filter by waybill status. One of: `pending`, `completed`, `partial`, `cancelled`, `failed`, `overdue`.

Example input: `{}`

### `holdedmcp_list_webhooks`

List webhooks · Read-only

**Required permission:** `developers:webhooks.read`

Returns a paginated list of webhooks for the account.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `cursor` (`string`, optional): Opaque cursor for pagination
- `limit` (`integer`, optional): Maximum number of webhooks to return per page. Defaults to 50, maximum 200. Default: `50`.

Example input: `{}`

### `holdedmcp_preview_inbox_file`

Get a processed thumbnail of a file from an inbox document · Read-only

**Required permission:** `inbox:inbox.read`

Returns a PNG thumbnail of the file, resized to one of a closed set of platform widths. Omitting both width and height returns the original image unresized.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `filename` (`string`, required): Name of the file to retrieve
- `incomingDocumentId` (`string`, required): Incoming document identifier
- `height` (`integer`, optional): Target height in pixels. Optional: only allowed when equal to width and width is one of the platform square presets, to request an exact square crop instead of an auto height. One of: `32`, `40`.
- `width` (`integer`, optional): Target width in pixels. Must be one of the allowed platform presets; any other value returns 404. Omit together with height to get the original image unresized. One of: `32`, `40`, `86`, `248`.

Example input: `{"filename":"<filename>","incomingDocumentId":"<incomingDocumentId>"}`

### `holdedmcp_search_contacts`

Search contacts by name · Read-only

**Required permission:** `contacts:contacts.read`

Search for contacts by name. Matches against the `name` field using prefix matching (autocomplete-style): `?name=Joh` matches `John Due`, `John Smith`, etc. Token-aware, case- and diacritic-insensitive.

The response includes a `cursor` value and a `has_more` boolean. Pass the `cursor` to the next request to fetch the following page.

The default page size is 25 contacts. You can increase it up to 100 using the `limit` parameter.

Read only: yes. Destructive: no. Idempotent (safe to retry): yes.

Inputs:

- `name` (`string`, required): Search term matched against the contact `name` field (prefix matching, case- and diacritic-insensitive)
- `cursor` (`string`, optional): Opaque cursor for pagination. Use the value returned in the previous response to fetch the next page.
- `limit` (`integer`, optional): Maximum number of contacts to return per page. Defaults to 25, maximum 100. Default: `50`.

Example input: `{"name":"<name>"}`

## Related

Other Accounting & Finance connectors ([all 82](/agentkit/connectors/?category=accounting-finance)):

- [QuickBooks](/agentkit/connectors/quickbooks/): Scalekit connector, OAuth, 117 tools
- [Xero](/agentkit/connectors/xero/): Scalekit connector, OAuth, 94 tools
- [Stripe](/agentkit/connectors/stripe/): Scalekit connector, Access token, 178 tools


---

## More Scalekit documentation

| Resource | What it contains | When to use it |
|----------|-----------------|----------------|
| [/llms.txt](/llms.txt) | Structured index with routing hints per product area | Start here — find which documentation set covers your topic before loading full content |
| [/llms-full.txt](/llms-full.txt) | Complete documentation for all Scalekit products in one file | Use when you need exhaustive context across multiple products or when the topic spans several areas |
| [sitemap-0.xml](https://docs.scalekit.com/sitemap-0.xml) | Full URL list of every documentation page | Use to discover specific page URLs you can fetch for targeted, page-level answers |
