---
title: "Microsoft Dataverse integration for AI agents | Langdock"
description: "Connect Microsoft Dataverse to Langdock. Agents and workflows can use 17 actions, such as “Add note” and “Create multiple records”."
url: "https://langdock.com/integrations/microsoft-dataverse"
language: "en"
---

[Integrations](https://langdock.com/products/integrations)

# Microsoft Dataverse integration for AI agents

Query, analyze, and update the Dataverse tables behind your Power Apps

Connect Microsoft Dataverse to Langdock and use it in chat, in agents and in workflows.

17 actions · 2 triggers

[Start free trial](https://app.langdock.com/signup) [Talk to sales](https://langdock.com/talk-to-sales)

## Microsoft Dataverse actions

Each action is one step an agent or workflow can run in Microsoft Dataverse.

- Add note

  Adds a note to a Dataverse record's timeline, optionally with one attached file, e.g. a call summary on a ticket or a signed offer on an account.
- Create multiple records

  Adds many rows to one Dataverse table in one step, e.g. importing 40 contacts from a spreadsheet or logging a batch of tickets, so the import is confirmed once instead of per row. Up to 1000 rows, each with its own column values. Returns the new IDs by row number (row 1 is the first object in the list) and the rows that could not be created. For a single row, use 'Create record'.
- Create record

  Creates a row in a Dataverse table and returns its ID and a link to open it. Lookups accept the linked record's GUID and choice columns accept their label, so no Web API binding syntax is needed. To add many rows at once, use 'Create multiple records'.
- Delete record

  Permanently deletes a Dataverse record. Related records may be deleted or unlinked according to the table's relationship settings.
- Describe table

  Describes a Dataverse table's columns: logical name, display name, type, whether it is required when creating a record, the allowed values of choice columns, and which tables a lookup column points to. For tables with many columns, pass search to get only the matching ones. Call this before writing SQL or creating and updating records to get exact column names and choice values.
- Download file

  Downloads a file stored in Dataverse so it can be read or analyzed: the file or image column of a record (e.g. a signed service report or a contact photo), or the attachment of a note (table 'annotation' with the note ID from 'List notes').
- Get record

  Gets one Dataverse record by its ID or record URL, with linked record names and choice labels resolved. Returns all non-empty columns unless specific columns are requested.
- List notes

  Lists the notes (comments) on a Dataverse record (the Notes section of its timeline), newest first, with their text and attached file details. Download an attachment with 'Download file', using table 'annotation' and the note ID.
- List records

  Lists rows from one Dataverse table, optionally filtered and sorted, e.g. the 20 most recently modified accounts or all open service tickets for one customer. Lookup columns include the linked record's name and choice columns their label. For joins, counts, or totals across tables, use 'Query with SQL' instead.
- List tables

  Lists the tables in the Dataverse environment, with each table's logical name (used in SQL and other actions), display name, and description. Your organization's own tables are listed first. Use this to find the right table before describing or querying it.
- List views

  Lists the saved views of a Dataverse table, such as 'My Active Accounts' or 'Active Service tickets': the public views shared in the app and the user's personal views. Run one with 'Run view'.
- Query with SQL

  Runs a read-only SQL SELECT against Dataverse tables and returns the rows. Use it for joins across tables, counts, sums, averages, and grouped reports, e.g. open tickets per customer. Respects the user's Dataverse security roles. Aggregate queries (COUNT, SUM, AVG, MIN, MAX, GROUP BY) can cover at most 50,000 rows per query; narrow larger tables with WHERE. Call 'Describe table' first to get exact column names and choice values.
- Run view

  Returns the rows of a saved Dataverse view, such as 'My Active Accounts' or a personal view, with the view's own filter, columns and sort order, exactly as users see it in the app. Use 'List views' to see the available views.
- Search records

  Full-text search across Dataverse tables, like the search box in Power Apps, e.g. everything mentioning 'Müller' or 'pump error'. Finds records by name and other indexed text, tolerates partial words, and returns the best matches from all searchable tables with their table and ID. Use it when the table or exact value is unknown; to filter one known table precisely, use 'List records' instead. It does not match record IDs (GUIDs): for a known record use 'Get record', and for its notes use 'List notes'.
- Update multiple records

  Applies the same column values to many records of one Dataverse table in one step, e.g. closing 40 resolved tickets or reassigning all of a customer's open cases, so the change is confirmed once instead of per record. Up to 1000 records. Only the given columns change; records are never created. Get the record IDs first with 'List records', 'Run view', or 'Query with SQL'. For a single record, use 'Update record'.
- Update record

  Updates columns of an existing Dataverse record. Only the given columns change; the record is never created if it does not exist. Lookups accept the linked record's GUID and choice columns accept their label.
- Upload file

  Stores a file in the file or image column of a Dataverse record, e.g. a signed service report on a ticket or a photo on a contact. Replaces the file currently in that column. To attach a file to the record's timeline instead, use 'Add note'.

## Microsoft Dataverse triggers

A trigger starts a workflow when something happens in Microsoft Dataverse.

- New record

  Triggers when a row is added to a Dataverse table, e.g. a new service ticket, case or contact. Each event carries the full row with choice labels and the names of linked records. Optionally only rows that match a filter trigger.
- Updated record

  Triggers when an existing row in a Dataverse table is changed, e.g. a ticket's status or owner. Does not fire for newly created rows; use New record for that. Each event carries the full row as it is after the change, not which columns changed; several saves within one polling interval (about 2 minutes) arrive as one event with the latest values.

## Setup and permissions

An action can ask for your confirmation before it runs. In an agent, each action is set to “Always allow” or “Needs approval”.

### Documentation

- [Using integrations](https://docs.langdock.com/en/using-langdock/guides/integrations/using-integrations)
- [How connections work](https://docs.langdock.com/en/using-langdock/integrations/connections)
- [Manage access to actions](https://docs.langdock.com/en/admin/manage-integrations/manage-action-access)

[All integrations](https://langdock.com/products/integrations)
