> ## Documentation Index
> Fetch the complete documentation index at: https://docs.maia.ai/llms.txt
> Use this file to discover all available pages before exploring further.
# 2026 changelog
export const AIAgents = () => <>the AI Agents icon 
;
export const mission_control = "Mission Control";
export const context_engine = "Context Engine";
export const metl = "Matillion ETL";
export const m_runner = "Maia runner";
export const s_runner = "Streaming runner";
export const designer = "Designer";
export const maia_agents = "Maia AI Agents";
export const maia = "Maia";
Here you'll find the 2026 changelog for {maia}. Just want to read about new features? Read our [New Features](https://roadmap.matillion.com/changelog) blog.
This page covers all releases to Maia during 2026. We have also made available variants of this changelog for features specifically deployed to our [Current runner track](/docs/changelog/current-runner-features) and our [Stable runner track](/docs/changelog/stable-runner-features).
Each item documented on this page includes a minimum {m_runner} version for the Current track, or you'll see "All supported versions" if the feature is available on all supported {m_runner} versions.
For an explanation of runner tracks and minimum {m_runner} versions, read [{m_runner} version tracks](/docs/guides/runner-overview#maia-foundation-runner-version-tracks).
You can filter the {maia} changelog using the topic filters on the right of this page, making it easier to learn about the new features and improvements you're most interested in. Click one or more topics to show the changelog entries for the selected topics. For more information about the topics, read the corresponding [changelog entry](#documentation-new-features).
## Designer - Improvements
* A new property of [iterator components](/docs/guides/iterator-components) allows you to set a maximum number of concurrent iterations on any iterator, giving more flexible control of how pipelines are executed. You can set a concurrency limit that gives optimal performance without overwhelming your downstream systems or data warehouse.
**Minimum {m_runner} version:** All supported versions.
## Connectors - Improvements
A new **Key sanitization strategy** property has been added to some connectors. This property lets you choose how to change the column names in your data so that they conform to your destination warehouse's identifier rules.
The **Key sanitization strategy** property has been added to the following connectors:
* [Amazon Redshift Load](/docs/components/amazon-redshift-load)
* [Databricks](/docs/components/databricks)
* [Gmail](/docs/components/gmail)
* [Google Ads Load](/docs/components/google-ads-load)
* [Google Sheets](/docs/components/google-sheets)
* [IBM Db2 for i Load](/docs/components/ibm-db2-load)
* [JDBC Load](/docs/components/jdbc-load)
* [JDBC](/docs/components/jdbc)
* [Jira Load](/docs/components/jira-load)
* [Mailchimp](/docs/components/mailchimp)
* [MariaDB Load](/docs/components/mariadb-load)
* [Microsoft Dynamics 365 Load](/docs/components/microsoft-dynamics-365-load)
* [Microsoft Exchange](/docs/components/microsoft-exchange)
* [Microsoft SQL Server Load](/docs/components/microsoft-sql-server-load)
* [NetSuite SuiteAnalytics](/docs/components/netsuite-suiteanalytics)
* [NetSuite SuiteAnalytics Load](/docs/components/netsuite-suiteanalytics-load)
* [Oracle Eloqua Load](/docs/components/oracle-eloqua-load)
* [Oracle Fusion Cloud Financials Load](/docs/components/oracle-fcf-load)
* [Oracle Fusion Cloud HCM Load](/docs/components/oracle-hcm-load)
* [Oracle Fusion Cloud Procurement Load](/docs/components/oracle-fcp-load)
* [Oracle Fusion Cloud Project Management Load](/docs/components/oracle-fcpm-load)
* [Oracle Load](/docs/components/oracle-load)
* [PostgreSQL Load](/docs/components/postgresql-load)
* [Sage Intacct Load](/docs/components/sage-intacct-load)
* [Salesforce Load](/docs/components/salesforce-load)
* [SAP NetWeaver Load](/docs/components/sap-netweaver-load)
* [ServiceNow Load](/docs/components/servicenow-load)
* [Shopify Load](/docs/components/shopify-load)
* [Snowflake Load](/docs/components/snowflake-load)
* [SugarCRM Load](/docs/components/sugarcrm-load)
* [SurveyMonkey](/docs/components/surveymonkey)
* [Sybase ASE Load](/docs/components/sybase-ase-load)
* [Workday Load](/docs/components/workday-load)
* [X Ads Load](/docs/components/x-ads-load)
* [X Ads](/docs/components/x-ads)
* [Zuora Bulk](/docs/components/zuora-bulk)
**Minimum {m_runner} version:** All supported versions.
## Pipelines - Improvements
### Observability
Steps within a pipeline run now display a **Pending** status when they have been sent for execution but haven't started yet. Once the runner picks up a step, the status moves to **Running**. This gives you a clearer picture of your pipeline's progress—you can now distinguish between steps that are actively executing and those that are queued. Step statuses are visible from the pipeline run details page and from the **Task history** tab in {designer}. For more information, read [Pipeline run history](/docs/guides/pipeline-run-history).
**Minimum {m_runner} version:** 11.459.0
## Maia AI Agents - New Features
You can now use the **Auto-start** toggle on the [{mission_control}](/docs/guides/mission-control) board to tell {maia_agents} to automatically start tasks that are added to your backlog. For more details, read [Auto-starting tasks](/docs/guides/mission-control#auto-starting-tasks).
**Minimum {m_runner} version:** All supported versions.
## Connectors - New features
* Added a new [Google Ads Load](/docs/components/google-ads-load/) connector for Snowflake and Google BigQuery projects. This connector offers full and incremental load.
**Minimum {m_runner} version:** 11.478.0
* Added a new [Email Load](/docs/components/email-load) connector for Snowflake and Google BigQuery projects. This connector offers full and incremental load.
**Minimum {m_runner} version:** 11.343.0
## API - Updated endpoints
The following endpoints have been updated for the [{maia} REST API](/docs/api-reference/maia-api-overview):
| Method | Endpoint | Description |
| ----------------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------ | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| [GET](/api-reference/pipeline-execution/pipeline-executions) | `/v1/pipeline-executions` | `timeFrame` query parameter description expanded to clarify that it filters to pipelines that started within the time frame relative to now, and accepts an ISO-8601 duration string in the `PnDTnHnMnS` format using days, hours, minutes, and/or seconds (for example, `P1D`, `P7D`, `PT1H`, `PT30M`, `PT90S`). Week, month, and year units (`P1W`, `P1M`, `P1Y`) aren't supported—use the day-based equivalents `P7D`, `P30D`, and `P365D` instead. Examples replaced with seven new examples: past day, past 1 day and 6 hours, past week (7 days), past month (30 days), past hour, past 30 minutes, and past 90 seconds. |
| [GET](/api-reference/connections/retrieve-a-page-of-project-connections-optionally-filtered-by-parameters) | `/v1/projects/{projectId}/connections` | Response schema refined: the `connectionProperties` `oneOf` member schemas are now titled `Cloud Platform Credentials` and `Component Basic`, and the property description now explains that connection properties are determined by the combination of `platformConnectionType`, `authProvider`, and `authType`. `authType` enum gains `COMPOSITE_TOKEN` and `OAUTH_1_0_X_ADS`; the `platformConnectionType` description now documents `COMPONENT_BASIC` and `CLOUD_PLATFORM_CREDENTIALS`, with the deprecated `DWH` value removed; the `secretsLocation` description expanded; and the `authProvider` example changed to `salesforce`. |
| [POST](/api-reference/connections/create-a-new-project-connection) | `/v1/projects/{projectId}/connections` | Request and 201 response schemas updated: the `connectionProperties` `oneOf` member schemas are now titled `Cloud Platform Credentials` and `Component Basic`. The response also adds `COMPOSITE_TOKEN` and `OAUTH_1_0_X_ADS` to the `authType` enum, expands the `platformConnectionType` description (and removes the deprecated `DWH` value), and refines the `connectionProperties`, `secretsLocation`, and `authProvider` descriptions and examples. |
| [PUT](/api-reference/connections/update-an-existing-project-connection-by-name) | `/v1/projects/{projectId}/connections` | Mirrors the POST endpoint: request and 200 response schemas updated with titled `connectionProperties` `oneOf` member schemas (`Cloud Platform Credentials` and `Component Basic`), new `authType` enum values `COMPOSITE_TOKEN` and `OAUTH_1_0_X_ADS`, an expanded `platformConnectionType` description with the deprecated `DWH` value removed, and refined `connectionProperties`, `secretsLocation`, and `authProvider` descriptions and examples. |
| [GET](/api-reference/connections/retrieve-a-page-of-connection-overrides-which-apply-in-a-specific-environment) | `/v1/projects/{projectId}/connections/environment-overrides` | 200 response schema refined: the `connectionOverrides.connectionProperties` `oneOf` member schemas are now titled `Cloud Platform Credentials` and `Component Basic`. |
| [POST](/api-reference/connections/create-an-environment-connection-override) | `/v1/projects/{projectId}/connections/environment-overrides` | Request and 201 response schemas refined: the `connectionOverrides.connectionProperties` `oneOf` member schemas are now titled `Cloud Platform Credentials` and `Component Basic`. |
| [PUT](/api-reference/connections/update-an-existing-environment-connection-override) | `/v1/projects/{projectId}/connections/environment-overrides` | Request and 200 response schemas refined: the `connectionOverrides.connectionProperties` `oneOf` member schemas are now titled `Cloud Platform Credentials` and `Component Basic`. |
| [GET](/api-reference/test-execution/list-test-executions) | `/v1/test-executions` | `timeFrame` query parameter description expanded to clarify that it filters to test executions that started within the time frame relative to now, and accepts an ISO-8601 duration string in the `PnDTnHnMnS` format using days, hours, minutes, and/or seconds (for example, `P1D`, `P7D`, `PT1H`, `PT30M`, `PT90S`). Week, month, and year units (`P1W`, `P1M`, `P1Y`) aren't supported—use the day-based equivalents `P7D`, `P30D`, and `P365D` instead. Examples replaced with seven new examples: past day, past 1 day and 6 hours, past week (7 days), past month (30 days), past hour, past 30 minutes, and past 90 seconds. |
**Minimum {m_runner} version:** All supported versions.
## Designer - Improvements
The **Sample data** tab in {designer} has been updated:
* {maia_agents} can now help you write filter queries when sampling data in {designer}. Click **Ask Maia** in the **Sample data** tab and tell {maia_agents} how you want to filter your data. For more information, read [Filtering sampled data](/docs/guides/sampling-output#filtering-sampled-data).
* You can now sort sampled data by the values in a selected column. Click the header of a column to use it to sort the data in the table. For more information, read [Sort sampled data](/docs/guides/sampling-output#sort-sampled-data).
**Minimum {m_runner} version:** All supported versions.
## Pipelines, Projects - New features
Google BigQuery is now a supported cloud data warehouse in {maia}.
Google BigQuery is available on the Current runner track from today. If you're using the Stable track, expect Google BigQuery to be available around August 1, 2026.
To get started, read:
* [Setup guide—Full SaaS Google BigQuery](/docs/guides/full-saas-bigquery)
* [Setup guide—Hybrid SaaS Google BigQuery](/docs/guides/hybrid-saas-bigquery)
* [Google BigQuery projects](/docs/guides/bigquery-projects)
* [Google BigQuery environments](/docs/guides/bigquery-environments)
The following components are available for Google BigQuery data pipelines.
**Transformation components**
* Aggregate
* Assert View
* Calculator
* Construct Struct
* Convert Type
* Create View
* Detect Changes
* Distinct
* Except
* Extract Structured Data
* Filter
* First/Last
* Fixed Flow
* Generate Sequence
* Intersect
* Join
* Lead Lag
* Rank
* Rename
* Rewrite Table
* Split Field
* SQL
* Table Input
* Table Output
* Transpose Columns
* Transpose Rows
* Unite
* Wildcard Table Input
* Window Calculation
**Orchestration components**
* And
* Append to Grid
* Assert Scalar Variables
* Assert Table
* Assert Table Values
* Azure Queue Storage Message
* Bash Pushdown
* Cloud Pub/Sub
* Compare Table Values
* Create External Table
* Create Table
* Create Table from Values
* dbt Core
* Delete Tables
* End Failure
* End Success
* File Iterator
* Fixed Iterator
* Grid Iterator
* If
* JDBC Table Metadata to Grid
* Or
* Print Variables
* Query Result to Grid
* Query Result to Scalar
* Refresh Materialized View
* Remove from Grid
* Retry
* Run Orchestration
* Run Pipeline to Test
* Run Shared Pipeline
* Run Transformation
* Send Email
* SQL Script
* Start
* Table Iterator
* Table Metadata to Grid
* Table Update
* Update Scalar
* Webhook Post
**Connectors**
* ActiveCampaign
* Amazon Redshift Load
* Amplitude
* Anaplan
* Asana
* Attio
* AVRO Load
* Azure Cosmos DB for NoSQL
* Azure SQL
* Box
* Braze
* Brevo
* Bugcrowd
* Buildkite
* Chargebee
* CircleCI
* ClickUp
* Coalesce
* Concord
* Confluence
* CSV Load
* Data Transfer
* Databricks
* Datadog
* dbt Cloud
* Delighted
* Dixa
* Dropbox
* Dynamics 365 Business Central
* Eventbrite
* Fanatics
* FireHydrant
* Freshdesk
* GitHub
* Gmail
* Gong
* Google Cloud Storage Unload
* Google Drive Table
* Google Sheets
* HubSpot Load
* IBM Db2 for i Load
* Infobip
* Intercom
* JDBC Load
* Jira Load
* Jotform
* JSON Load
* Klaviyo
* LaunchDarkly
* LinkedIn Ads
* Lob
* Maia Flex connector
* Mailchimp
* Mailgun
* Mandrill
* MariaDB Load
* Marketo Load
* Microsoft Dataverse
* Microsoft Dynamics 365 Load
* Microsoft Exchange
* Microsoft SQL Server Load
* Mixpanel
* NetSuite SuiteAnalytics Load
* New Relic
* Notion
* Ongoing WMS
* Open Exchange Rates
* Oracle Eloqua Load
* Oracle Fusion Cloud Financials Load
* Oracle Fusion Cloud HCM Load
* Oracle Fusion Cloud Procurement Load
* Oracle Fusion Cloud Project Management Load
* Oracle Load
* ORC Load
* Orbit
* Ortto
* PagerDuty
* Parquet Load
* PayPal
* Pendo
* Persona
* Pingdom
* Pipedrive
* PostgreSQL Load
* Productboard
* Recurly
* RingCentral
* Rootly
* SAP NetWeaver Load
* SAP ODP
* SAP SuccessFactors
* Sage Intacct Load
* Salesforce Load
* Salesforce Pardot
* Salesloft
* SeatGeek
* SendGrid
* ServiceNow Load
* Shopify Load
* Slack
* Smartsheet
* Snapchat
* Snowflake Load
* Snyk
* Square
* SugarCRM Load
* SurveyMonkey
* Sybase ASE Load
* Ticketmaster
* TikTok
* Toggl
* Twilio
* Unstructured.io
* Workday Custom Reports
* Workday Load
* X Ads Load
* Yelp
* Zendesk Talk
* Zendesk Ticketing
* Zoom
* Zuora
* Zuora Bulk
**Minimum {m_runner} version:** 11.478.0
## Designer - Improvements
You can now bulk export and import both [pipeline and project variables](/docs/guides/variables) via a structured JSON configuration file.
**Minimum {m_runner} version:** All supported versions.
## Designer - Improvements
The {designer}'s [command palette](/docs/guides/designer-ui-basics#command-palette) has been upgraded with the following items:
* You can now execute [Git operations](/docs/guides/git-overview), toggle sidebar panels, move pipelines, and manage schedules directly from the command palette.
* The underlying search engine accounts for typos or partial matches, and automatically surfaces your most frequently used commands over time based on your usage patterns.
**Minimum {m_runner} version:** All supported versions.
## Runners - Improvements
* Updated the [{s_runner} version](/docs/streaming/agent-versions) from `2.119.1` to `2.119.18`.
* Library updates and security fixes.
## Documentation - New features
Added topic filters to the {maia} changelog. Click one or more topics on the right of this page to show the changelog entries for the selected topics. Each topic covers a number of sub-topics, including:
* **Admin:** Roles and permissions, billing, SSO, security, and account-level settings
* **API:** The {maia} API and MCP server
* **Connectors:** Connectors and custom connectors
* **{designer}:** Features available on the {designer} canvas, such as variables, sampling, and Git actions
* **{maia_agents}**: {mission_control}, {context_engine}, {maia_agents} chat features
* **Pipelines:** Orchestration and transformation components, test pipelines, shared pipelines, and pipeline observability
* **Projects:** Project-level features, such as environments, schedules, secret definitions, and OAuths
* **Runners:** {m_runner}s and {s_runner}s
## Pipelines - Improvements
### Observability
You can now subscribe to anomaly alert notifications via Email, Slack, or Webhook to be notified when a pipeline run's duration deviates significantly from historical baselines, so you can investigate performance issues before they escalate. For setup instructions, see [Pipeline notifications](/docs/guides/pipeline-notifications).
**Minimum {m_runner} version:** All supported versions.
## API - New endpoint
The following endpoint has been added to the [{maia} REST API](/docs/api-reference/maia-api-overview):
| Method | Endpoint | Description |
| --------------------------------------------------------------------------------------------------------------------------------- | --------------------------------------------- | ---------------------------------------------------------------------------------------------------------------------- |
| [POST](/api-reference/repositories/rotate-the-access-token-for-a-project-repository) | `/v1/projects/{projectId}/repositories/token` | Rotate the Git provider personal access token for a project repository. The new token is stored in the customer vault. |
**Minimum {m_runner} version:** All supported versions.
## Designer - Improvements
The following improvements have been made in {designer}:
* A new property of [iterator components](/docs/guides/iterator-components) allows the Task History and Observability Dashboard to show the names and values of variables used in each iteration.
* You can now search for variables by name in the **Variables** panel at the bottom of the {designer} canvas.
**Minimum {m_runner} version:** All supported versions.
## API - Updated endpoints
The following endpoint has been updated for the [{maia} REST API](/docs/api-reference/maia-api-overview):
| Method | Endpoint | Description |
| --------------------------------------------------------------------------------------------------------- | --------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------ |
| [POST](/api-reference/environments/create-a-new-environment) | `/v1/projects/{projectId}/environments` | The request body now includes a new `inheritProjectDefaults` property that determines whether the environment inherits the project's default settings. |
## API - Improvements
You can now associate a GitLab repository with a project when using the [Project Provisioning API](/docs/api-reference/provisioning-a-project) to create a project.
**Minimum {m_runner} version:** All supported versions.
## Maia AI Agents - Improvements
You can now filter the tasks shown on the {mission_control} board by their **Source app** and **Source branch** for a clearer overview of where your tasks were created and where {maia_agents} are working on them. For more details, read [Viewing tasks](/docs/guides/mission-control#viewing-tasks).
**Minimum {m_runner} version:** All supported versions.
## Pipelines - Improvements
### Observability
Pipeline notifications now support four completion trigger types: **Success**, **Failure**, **Cancelled**, and **Skipped**. You can subscribe to any combination of trigger types for your scheduled and API-driven pipelines. For more information, see [Pipeline notifications](/docs/guides/pipeline-notifications).
**Minimum {m_runner} version:** All supported versions.
## Projects - Improvements
### Secret definitions
When creating a secret using the {maia} public API, you will now see a `409 Conflict` response if a secret with the same name already exists in the target vault location. Previously, if you created a secret with a name that already existed, the stored value was overwritten without warning.
Secret names must now be unique within a vault location, and existing secrets aren't overwritten. This ensures that the relevant `create` and `edit` permissions are correctly applied through the {maia} API. The change applies to the endpoint for creating a secret value. For more information, read [Managing secret references](/docs/api-reference/secret-references).
**Minimum {m_runner} version:** All supported versions.
## API - New endpoint
The following endpoint has been added to the [{maia} REST API](/docs/api-reference/maia-api-overview):
| Method | Endpoint | Description |
| -------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------ | ---------------------------------------------------------------------------------------------------------------------------------------------------------- |
| [PUT](/api-reference/connections/update-an-existing-environment-connection-override) | `/v1/projects/{projectId}/connections/environment-overrides` | Update an existing environment-specific connection override. Use this endpoint to update an override that was created via the corresponding POST endpoint. |
## API - Updated endpoints
The following endpoints have been updated for the [{maia} REST API](/docs/api-reference/maia-api-overview):
| Method | Endpoint | Description |
| ----------------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------ | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| [GET](/api-reference/ai-agent-tasks/list-ai-agent-tasks) | `/v1/ai/agents/tasks` | Added `createdAfter` and `createdBefore` query parameters for filtering by creation time. Existing `branchId`, `projectId`, and `status` query parameters are now documented. Response now includes a `createdAt` timestamp per task. |
| [GET](/api-reference/consumption/get-a-breakdown-of-your-credit-consumption-for-flat-rated-products) | `/v1/consumption` | Description updated to note that consumption data refreshes every three hours to reflect recent credit usage. |
| [GET](/api-reference/consumption/get-a-breakdown-of-your-credit-consumption-for-etl-users) | `/v1/consumption/etl-users` | Description updated to note that consumption data refreshes every three hours to reflect recent credit usage. |
| [GET](/api-reference/connections/retrieve-a-page-of-project-connections-optionally-filtered-by-parameters) | `/v1/projects/{projectId}/connections` | Vault location schemas in the response now carry titles (`AWS Secrets Manager`, `Azure Key Vault`, `Default Vault`, `Matillion-hosted Vault`) for clearer client-side discrimination. |
| [POST](/api-reference/connections/create-a-new-project-connection) | `/v1/projects/{projectId}/connections` | Major description overhaul, including a new combinations table for valid `platformConnectionType` × `authProvider` × `authType` configurations. Added `COMPOSITE_TOKEN` and `OAUTH_1_0_X_ADS` enum values for `authType`. Removed the deprecated `DWH` enum value from `platformConnectionType`. Added a new `CLOUD_PLATFORM_CREDENTIALS Connection` request example and a corresponding response example. Vault location schemas in the response now carry titles. |
| [PUT](/api-reference/connections/update-an-existing-project-connection-by-name) | `/v1/projects/{projectId}/connections` | Mirrors the POST endpoint updates: added `COMPOSITE_TOKEN` and `OAUTH_1_0_X_ADS` to `authType` enum, removed `DWH` from `platformConnectionType`, descriptions rewritten, and vault location schemas now titled. |
| [DELETE](/api-reference/connections/delete-environment-connection-overrides) | `/v1/projects/{projectId}/connections/environment-overrides` | `connectionNames` query parameter description cleaned up — fixed `Its` to `It's` and removed stray leading whitespace. |
| [GET](/api-reference/connections/retrieve-a-page-of-connection-overrides-which-apply-in-a-specific-environment) | `/v1/projects/{projectId}/connections/environment-overrides` | Vault location schemas in the response now carry titles for clearer client-side discrimination. |
| [POST](/api-reference/connections/create-an-environment-connection-override) | `/v1/projects/{projectId}/connections/environment-overrides` | Endpoint now creates overrides only — it no longer upserts. Returns `409 Conflict` if an override already exists; use the new PUT endpoint to update. Added `COMPOSITE_TOKEN` and `OAUTH_1_0_X_ADS` to the `authType` enum. New `CLOUD_PLATFORM_CREDENTIALS Environment Override` request and response examples. The `403` response description has been corrected from `Unauthorized` to `Forbidden`. |
| [POST](/api-reference/repositories/associate-a-repository-with-a-project) | `/v1/projects/{projectId}/repositories` | Now supports creating GitLab repositories. `provider` enum extended with `gitlab`, a new `providerParameters` property has been added, and a new `GitLab Repository` request example has been included. |
| [POST](/api-reference/secrets/create-a-new-secret-value) | `/v1/secrets/{agentId}/secret-value` | Added a `409 Conflict` response. |
## MCP server - New features
Four new **Test Execution** tools are now available in the [Matillion MCP server](/docs/api-reference/mcp-server), enabling you to validate pipeline changes by running tests from your AI assistant client:
| Tool | Type | Description |
| --------------------------- | ----- | --------------------------------------------------------------- |
| `list-test-executions` | Read | List test executions across all projects, with optional filters |
| `get-test-execution-status` | Read | Get the status and details of a specific test execution |
| `create-test-execution` | Write | Trigger all or specified test executions in a project |
| `cancel-test-execution` | Write | Cancel a running test execution |
**Minimum {m_runner} version:** All supported versions.
## Designer - New features
Three new system variables are available for all iterator components. Use these variables to retrieve counts of attempted iterations, generated iterations, and successful iterations. For details, read [Iterator components](/docs/guides/iterator-components#system-variables).
**Minimum {m_runner} version:** All supported versions.
## Connectors - Improvements
The [Zuora Bulk](/docs/components/zuora-bulk) connector is now available for Databricks and Amazon Redshift projects.
**Minimum {m_runner} version:** 11.334.9-patch.2
## Connectors - New features
The following connectors have been added:
* [Zuora Bulk](/docs/components/zuora-bulk) orchestration connector that allows you to extract large volumes of data from Zuora via the Zuora Bulk API, and load that data into either Snowflake or a cloud storage location (Amazon S3, Azure Blob Storage, or Google Cloud Storage).
**Minimum {m_runner} version:** 11.387.0
* [Oracle Fusion Cloud HCM Load](/docs/components/oracle-hcm-load) orchestration connector that allows you to extract data from Oracle Fusion Cloud HCM and load it directly into your preferred cloud storage location.
**Minimum {m_runner} version:** 11.403.0
## Maia AI Agents - New features
**Navigation:** Added to the left navigation. Use this icon to access {mission_control} and {context_engine}.
[{context_engine}](/docs/guides/context-engine) is now available in **public preview**. Use {context_engine} to add and manage knowledge graphs that capture the structure, relationships, and meaning of your data. {maia_agents} use knowledge graphs to speak the language of your business and build pipelines relevant to your data landscape. Add crawlers to ingest data from your connected data warehouses and pipeline execution history to keep your knowledge graphs up to date.
[{mission_control}](/docs/guides/mission-control) is now available in **public preview**. {mission_control} provides a kanban-style dashboard where you can create, track, and manage tasks being executed by {maia_agents} across all your projects. Tasks move through four columns—**Backlog**, **In progress**, **Needs attention**, and **Completed**—and each task has its own chat panel, so you can switch between tasks without losing context. When creating a task, you can select a knowledge graph to help {maia_agents} work with awareness of your data landscape.
With {mission_control}, we've introduced support for multi-modality. When you create a task in {mission_control}, you can attach images and PDFs to provide additional context. {maia_agents} can see and understand the content of these attachments, allowing them to work directly from diagrams, screenshots, mockups, whiteboard photos, and PDF specifications you share. This is especially useful for tasks like troubleshooting a complex pipeline, where a screenshot of the pipeline canvas or error message can provide critical context that might be missed in a text description. You can reference these attachments in your {mission_control} conversations by typing `@` and then the name of the file.
Text files (`.txt`, `.md`, etc.) aren't supported as attachments. These file types belong in the project itself, where {maia_agents} can already read them.
With {mission_control} tasks, we've also introduced **Bypass permissions** mode. When enabled, {maia_agents} execute tool calls immediately without pausing for approvals, decisions, or check-ins—the opposite of **Ask permission** mode. This is useful for trusted, hands-off runs such as prototyping or clearing a backlog in a sandbox environment. We recommend keeping **Ask permission** as your default and enabling **Bypass permissions** only for specific tasks where you know what {maia_agents} are doing and have scoped your environment accordingly.
You can also watch our new video about [Mission Control](https://www.youtube.com/watch?v=W7HHFcQdKQ4).
**Minimum {m_runner} version:** All supported versions.
## Connectors - Improvements
The [Maia Flex connector](/docs/components/maia-flex-connector/) component has been updated to include three new **Test Execution** endpoints:
* [Create Test Execution](/api-reference/test-execution/create-test-execution)
* [List Test Executions](/api-reference/test-execution/list-test-executions)
* [Test Execution Status](/api-reference/test-execution/test-execution-status)
**Minimum {m_runner} version:** All supported versions.
## Designer - Improvements
The following improvements have been made in {designer}:
* The pipeline file list now updates automatically in real time after importing assets from the Exchange or making changes in another browser tab, removing the need for a manual page refresh.
* When you switch [shared pipelines](/docs/guides/shared-pipelines) within an orchestration job, any pre-configured variable values that no longer exist or match the new pipeline are automatically cleared to prevent runtime errors.
**Minimum {m_runner} version:** All supported versions.
## Designer - New features
The **Stop on condition mode** property for iterator components now includes an **Advanced** option, letting you write stop conditions manually using SQL. For more information, read the documentation for each component:
* [Loop Iterator](/docs/components/loop-iterator)
* [File Iterator](/docs/components/file-iterator)
* [Fixed Iterator](/docs/components/fixed-iterator)
* [Grid Iterator](/docs/components/grid-iterator)
* [Table Iterator](/docs/components/table-iterator)
**Minimum {m_runner} version:** All supported versions.
## Designer - Improvements
The [Variables](/docs/guides/variables) panel has now moved to the bottom of the {designer} UI, sitting between the **Metadata** and **Review results** tabs. This relocation empowers you to manage your pipeline, project, and grid variables, as well as environment overrides—while keeping the {maia_agents} interface open on your canvas.
**Minimum {m_runner} version:** All supported versions.
## Runners - New features
{m_runner}s can now be deployed to Google Cloud on Google Kubernetes Engine (GKE). For details on architecture, prerequisites, and deployment, read the [Google Cloud GKE deployment for {m_runner}s](/docs/guides/gke-deployment-guide).
**Minimum {m_runner} version:** All supported versions.
## Designer - Improvements
The following improvements have been made in {designer}:
* You can now copy and paste [canvas notes](/docs/guides/designer-ui-basics#canvas-notes) individually, or select multiple canvas notes to copy, delete, or change the color of all selected notes.
* When you select multiple components on the canvas, you can now use the [component action bar](/docs/guides/components-overview#component-action-bar) to skip, copy, or delete all selected components.
* You can now deselect a selected component by holding the `Shift` key and clicking the component.
**Minimum {m_runner} version:** All supported versions.
## API - Updated endpoints
The following endpoint has been updated for the [{maia} REST API](/docs/api-reference/maia-api-overview):
| Method | Endpoint | Description |
| --------------------------------------------------------------------------------------------------------- | --------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| [POST](/api-reference/environments/create-a-new-environment) | `/v1/projects/{projectId}/environments` | Added a new request body example, **Snowflake Environment with Cloud Platform Credentials**, demonstrating how to create a Snowflake environment that also defines `CLOUD_PLATFORM` connections (AWS, GCP, and Azure) with credential secret references. |
## API - New endpoints
The following endpoints have been added to the [{maia} REST API](/docs/api-reference/maia-api-overview):
**Test Execution**
| Method | Endpoint | Description |
| ---------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------ | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| [POST](/api-reference/test-execution/create-test-execution) | `/v1/projects/{projectId}/test-executions` | Trigger test executions for a project. Supports `TEST`, `PIPELINE`, and `PROJECT` scopes, with optional `versionName`, `executionTag`, and `agentId` parameters. |
| [GET](/api-reference/test-execution/test-execution-status) | `/v1/projects/{projectId}/test-executions/{testExecutionId}` | Get the status of a specific test execution. |
| [PATCH](/api-reference/test-execution/cancel-test-execution) | `/v1/projects/{projectId}/test-executions/{testExecutionId}` | Cancel a test execution, optionally forcing termination. |
| [GET](/api-reference/test-execution/list-test-executions) | `/v1/test-executions` | List summarized test executions across the account, ordered by start time (most recent first). Supports filtering by project, environment, test name, and time range, with pagination. |
## API - New features
The [Project Provisioning API](/docs/api-reference/provisioning-a-project) now supports Azure DevOps, enabling you to associate an Azure DevOps Git repository with the provisioned project. For more details, read the [Initialize repository](/docs/api-reference/provisioning-a-project#initialize-repository) section of the Project Provisioning API guide.
## API - Improvements
The connections API now supports cloud provider credentials (AWS, Azure, and GCP), enabling you to programmatically create and manage cloud credentials. Combined with the [Project Provisioning API](/docs/api-reference/provisioning-a-project#create-cloud-credential-connection), this completes the zero-touch provisioning flow for full SaaS projects.
## Designer - New features
Post-processing in {designer} now supports updating grid variables. Set static or dynamic values, or map data from a source grid variable column by column. You can also export grid variables from child pipelines. For more information, read [Post-processing](/docs/guides/post-processing).
**Minimum {m_runner} version:** 11.329.1
## Designer - Improvements
You can now search for and add pipeline variables when using dual listbox editors, such as the **Column Names** parameter of the [Table Input](/docs/components/table-input) component. Variables appear in the same selector as static list items and can be added to the selection like any other item.
**Minimum {m_runner} version:** All supported versions.
## Projects - Improvements
### Cloud credentials
Cloud credentials now support environment overrides, enabling you to configure different cloud provider credentials (AWS, Azure, and GCP) per environment. For more information, read [Cloud credentials](/docs/guides/cloud-credentials#cloud-provider-credentials).
The allowed characters in cloud credential names have been updated to align with other connection types such as secrets and OAuths. Any existing names with non-compliant characters have been updated automatically.
**Minimum {m_runner} version:** All supported versions.
## Pipelines - New features
### Test pipelines
You can now create and run tests for your pipelines. Three test types are available:
* **Unit tests:** Test pipeline logic with mock inputs to validate behavior in isolation.
* **Comparison tests:** Compare output between pipelines to verify migration equality.
* **Operational tests:** Validate pipeline behavior for scenarios such as idempotency and temporal patterns.
{maia_agents} can suggest and create tests for you. Enable **Maia test suggestions** in your {maia} chat settings to allow {maia_agents} to offer test creation when you build or modify pipelines. When you update a pipeline, {maia_agents} automatically flag which existing tests are affected.
You can also trigger tests in bulk via API to integrate testing into your CI/CD workflow. For more information, read [Test pipelines](/docs/components/test-pipelines#test-pipelines).
**Minimum {m_runner} version:** 11.348.0
## Connectors - New features
Added a new [Microsoft Dynamics 365 Load connector](/docs/components/microsoft-dynamics-365-load), which offers Full and Incremental load options.
**Minimum {m_runner} version:** 11.305.3
## Designer - Improvements
* **Sample filtering:** The **Filter** field in the **Sample data** panel is now visible as soon as the open pipeline has been validated. If a sample request fails, the **Filter** field and retry controls remain visible, enabling you to change your filter request and sample again without interruption. For more information, read [Filtering sampled data](/docs/guides/sampling-output#filtering-sampled-data).
* **Inspect tables and views:** You can now observe the column metadata in a table or view from the **Schemas** panel. Click **Inspect data** next to a table or view to browse your warehouse data. You can also sample the data from this panel.
* **Select database:** A database selection drop-down has been added to the **Schemas** panel and the **Warehouse data** tab when you click the **Add component** icon on the canvas, allowing you to switch databases without changing your environment configuration.
**Minimum {m_runner} version:** All supported versions.
## MCP server - New features
Ten new AI Agents tools are now available in the [Matillion MCP server](/docs/api-reference/mcp-server), enabling you to create and manage {maia_agents}' tasks from your AI assistant client:
| Tool | Type | Description |
| ----------------------------- | ----- | ------------------------------------------------------------------------------------------------ |
| `create-agent-task` | Write | Create a new {maia_agents} task |
| `list-agent-tasks` | Read | List {maia_agents}' tasks for the account |
| `get-agent-task` | Read | Get a task's status, configuration, and pending decisions |
| `get-agent-task-messages` | Read | Get the message history and events for a task |
| `send-agent-task-message` | Write | Send a follow-up message to a running task to grant permissions or update the task configuration |
| `submit-agent-task-decisions` | Write | Approve or reject tool calls that {maia_agents} are waiting on, and answer pending questions |
| `set-agent-task-permissions` | Write | Replace the full list of granted tool permissions for a task |
| `wait-for-agent-task` | Read | Monitor a task until it reaches a pending decision, terminal state, or timeout |
| `stop-agent-task` | Write | Stop a task that is currently running |
| `delete-agent-task` | Write | Delete a task that has been stopped |
**Minimum {m_runner} version:** All supported versions.
## Admin - Improvements
### Security
The {m_runner} allow list is now enforced on existing schedules. Subsequent pipeline runs will be recorded with a new **Forbidden** status.
## API - Updated endpoints
This release contains a breaking change. Existing API clients that call this endpoint without the `connectionNames` query parameter will now receive a `400 Bad Request` response.
The [Delete environment connection overrides](/api-reference/connections/delete-environment-connection-overrides) endpoint has been updated for the [{maia} REST API](/docs/api-reference/maia-api-overview):
| Method | Endpoint | Description |
| ----------------------------------------------------- | ------------------------------------------------------------ | ------------------------------------------------------------------------------------------------------------------------------------------------------------ |
| DELETE | `/v1/projects/{projectId}/connections/environment-overrides` | The `connectionNames` query parameter is now **required** (previously optional). Requests that omit this parameter will return a `400 Bad Request` response. |
## Pipelines - New features
### Observability
The **Your activity** page now supports saved views. Save your applied filters as a named view, switch between views from the **Saved filters** drop-down, and mark one view as your default to load automatically each time you open the page. For more information, read [Saved views](/docs/guides/pipeline-run-history#saved-views).
**Minimum {m_runner} version:** All supported versions.
## Pipelines - Improvements
### Observability
The root cause analysis that {maia} provides when a scheduled or API-triggered executed pipeline run fails has been improved. The analysis {maia} conducts is more thorough and context-aware, and the analysis report you receive is more detailed and actionable. For more information, read [Root cause analysis](/docs/guides/maia-troubleshooting#root-cause-analysis).
**Minimum {m_runner} version:** All supported versions.
## API - Updated endpoints
The following endpoints have been updated for the [{maia} REST API](/docs/api-reference/maia-api-overview):
| Method | Endpoint | Description |
| ------------------------------------------------- | ------------------------------------------------------------ | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| POST | `/v1/projects/{projectId}/connections` | `authType` now accepts three additional OAuth values: `OAUTH_USER_TOKENS`, `OAUTH_AUTHORIZATION_CODE`, and `OAUTH_CLIENT_CRED`. For `COMPONENT_BASIC` connections, the value determines the type of OAuth secret referenced. |
| PUT | `/v1/projects/{projectId}/connections` | `authType` now accepts three additional OAuth values: `OAUTH_USER_TOKENS`, `OAUTH_AUTHORIZATION_CODE`, and `OAUTH_CLIENT_CRED`. |
| POST | `/v1/projects/{projectId}/connections/environment-overrides` | `authType` is now a **required** field in the request body, accepting `GENERIC_SECRET`, `CLOUD_PROVIDER_SECRET`, `OAUTH_USER_TOKENS`, `OAUTH_AUTHORIZATION_CODE`, or `OAUTH_CLIENT_CRED`. The value must match the parent connection's `authType`. |
| POST | `/v1/projects/{projectId}/environments` | The `connectionOverrides` array now accepts up to 200 items when creating an environment (previously limited to 1). |
| GET | `/v1/projects/{projectId}/published-pipelines` | `page` and `size` query parameters are now explicitly typed as integers with documented defaults (`page`: `0`, `size`: `25`) and a maximum of `1000` for `size`. |
## Runners - New features
* {s_runner}s now support exporting logs to Matillion's systems for support. To enable the feature, set the `EXPORT_LOGS` optional {s_runner} parameter to `true`. For more information, read [Export logs](/docs/streaming/streaming-agent-logs#export-logs).
## Runners - Improvements
* Updated the [{s_runner} version](/docs/streaming/agent-versions) from `2.116.1` to `2.119.1`.
* {s_runner}s log structure updated to a JSON format.
* Library updates and security fixes.
## API - New endpoints
The following endpoints have been added to the [{maia} REST API](/docs/api-reference/maia-api-overview):
**AI Agent Tasks**
| Method | Endpoint | Description |
| ----------------------------------------------------- | ------------------------------------------ | ---------------------------------------------------------------------------------------- |
| GET | `/v1/ai/agents/tasks` | List AI agent tasks for the account, optionally filtered by project, branch, or status. |
| POST | `/v1/ai/agents/tasks` | Create a new AI agent task and start async execution. |
| GET | `/v1/ai/agents/tasks/{taskId}` | Get details of a specific AI agent task, including current status and pending decisions. |
| DELETE | `/v1/ai/agents/tasks/{taskId}` | Permanently delete an AI agent task. The task must not be in `RUNNING` status. |
| GET | `/v1/ai/agents/tasks/{taskId}/messages` | Get paginated events for a task. |
| POST | `/v1/ai/agents/tasks/{taskId}/messages` | Send a follow-up message to a running or stopped task, resuming agent execution. |
| POST | `/v1/ai/agents/tasks/{taskId}/decisions` | Submit approval or question-answer decisions for a task awaiting human input. |
| PUT | `/v1/ai/agents/tasks/{taskId}/permissions` | Replace the full set of granted tool permissions for a task. |
| POST | `/v1/ai/agents/tasks/{taskId}/stop` | Request that a running task be stopped gracefully. |
## API - Updated endpoints
The following endpoints have been updated for the [{maia} REST API](/docs/api-reference/maia-api-overview):
| Method | Endpoint | Description |
| ------------------------------------------------- | ------------------------------------------------------------------ | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| POST | `/v1/projects/{projectId}/repositories` | Request body now accepts `oneOf` `CreateGitHubRepositoryRequest` or `CreateAzureDevOpsRepositoryRequest`, adding support for creating Azure DevOps repositories. The `provider` enum now includes `azure-devops` alongside `github`. A new example has been added for an Azure DevOps repository with `organization`, `repositoryId`, and `tenantId` parameters. |
| GET | `/v1/consumption` | Response schema updated: the `dimensions` object now includes an `environment` field for orchestration and transformation consumption, identifying the environment in which the pipeline ran. |
| POST | `/v1/projects/{projectId}/secret-references` and related endpoints | `agentId` is no longer a required field on `SecretReferenceRequest`, `AwsSecretReferenceRequest`, and `AzureSecretReferenceRequest`. It remains an optional property on `JsonSecretReferenceRequest`. |
## Designer - Improvements
{maia_agents} can now sample up to 200 rows of data when working on pipelines in {designer} to review your data, identify patterns, and troubleshoot transformation steps. For more information about how {maia_agents} sample your data, read [Sampling data](/docs/guides/maia-pipelines#sampling-data). **Minimum {m_runner} version:** All supported versions.
## Designer - New features
Iterator components now support a **Stop on condition** property, allowing you to stop iterations early when a specific condition is met. This brings {maia} iterator behavior in line with {metl}. When enabled, a **Condition** dialog lets you define one or more conditions using a no-code UI, with a **Text mode** toggle available for writing conditions manually as a JavaScript expression. A **Combine conditions** property controls whether multiple conditions are evaluated using **And** or **Or** logic. For more information, read the documentation for each component:
* [Loop Iterator](/docs/components/loop-iterator)
* [File Iterator](/docs/components/file-iterator)
* [Fixed Iterator](/docs/components/fixed-iterator)
* [Grid Iterator](/docs/components/grid-iterator)
* [Table Iterator](/docs/components/table-iterator)
**Minimum {m_runner} version:** All supported versions.
## Designer - New features
Connectors:
The following new connectors are now available for Snowflake projects, offering full and incremental load options to only fetch new and updated rows of data:
* [Oracle Eloqua Load](/docs/components/oracle-eloqua-load)
* [Sage Intacct Load](/docs/components/sage-intacct-load)
**Minimum {m_runner} version for both connectors:** 11.313.7
## Admin - Improvements
### Accounts
{maia} now supports three regions: US, EU, and Australia:
* Customers can select Australia as their region at account creation, and trials will run in AU automatically.
* Hybrid-SaaS Maia runner support is available in the Australia region.
## API - Updated endpoints
The following endpoints have been updated for the [{maia} REST API](/docs/api-reference/maia-api-overview):
| Method | Endpoint | Description |
| ------------------------------------------------- | -------------------------------------------------------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| GET | `/v1/consumption` | Response schema updated: examples for results and consumption properties now include more detailed pipeline/project/user information in the dimensions object. Descriptions for `pipelineName` and `projectId` updated for clarity. |
| GET | `/v1/projects/{projectId}/pipeline-executions/{pipelineExecutionId}` | Summary updated to "Pipeline Execution Details". Description expanded to clarify the endpoint returns full execution details. 200 response description updated for accuracy. |
| POST | `/v1/projects/{projectId}/pipeline-executions` | Description expanded to clarify asynchronous execution, artifact versioning, agent selection, concurrency control, and variable overrides. Request body schema updated: `scalarVariables` and `gridVariables` descriptions now clarify override behavior. |
| POST | `/v1/projects/{projectId}/schedules` | Request body schema updated: `pipeline.scalarVariables` and `pipeline.gridVariables` descriptions now clarify override behavior for scheduled pipeline executions. |
## Admin - Improvements
### Roles and permissions
User management and roles improvements:
* Redesigned user management screen with sorting, filtering, and paging for easier navigation.
* Simplified [account-level roles](/docs/administration/account-roles) to make permissions easier to understand.
* For {maia} users only, [project roles](/docs/administration/project-roles) can now be assigned during user invitation, so access is ready on acceptance.
* For {maia} users only, a new **Roles** section shows clear permissions across account, project, and environment levels.
## API - New endpoint
The following endpoint has been added for the [{maia} REST API](/docs/api-reference/maia-api-overview):
| Method | Endpoint | Description |
| ------- | --------------------------------------------------------------------------- | --------------------------------------------------- |
| **GET** | `/v1/projects/{projectId}/pipeline-executions/{pipelineExecutionId}/status` | Returns the current status of a pipeline execution. |
## API - Updated endpoints
The following endpoints have been updated for the [{maia} REST API](/docs/api-reference/maia-api-overview):
| Method | Endpoint | Description |
| --------- | -------------------------------------------------------------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **GET** | `/v1/consumption` | Response 200: `results.consumption.credits` example changed from '28.4' (string) to 28.4 (number). |
| **GET** | `/v1/pipeline-executions` | Response 200: Added enum value `SCHEDULE_RUN_NOW` to `results.trigger` updated `results.user` description. Response 401/403: Description changes. |
| **POST** | `/v1/projects/{projectId}/pipeline-executions` | Corrected response descriptions (401 Unauthorized, 403 Forbidden). |
| **GET** | `/v1/projects/{projectId}/pipeline-executions/{pipelineExecutionId}` | Description updated to recommend `getExecutionStatusOnly`. Response 200: Added enum value `SCHEDULE_RUN_NOW` to `trigger`; updated `user` description. Response 401/403: Description changes. |
| **PATCH** | `/v1/projects/{projectId}/pipeline-executions/{pipelineExecutionId}` | Corrected response descriptions (401 Unauthorized, 403 Forbidden). |
| **GET** | `/v1/projects/{projectId}/pipeline-executions/{pipelineExecutionId}/steps` | Corrected response descriptions (401 Unauthorized, 403 Forbidden). |
| **POST** | `/v1/projects/{projectId}/reviews` | Request body: `filePaths` property type can now be `null` (previously required an array). |
## Connectors - Improvements
Matillion's own Flex connector has been renamed from **Data Productivity Cloud** to [**{maia}**](/docs/components/maia-flex-connector). The connector's functionality and configuration remain unchanged.
## API - New endpoints
The following endpoints have been added to the [{maia} REST API](/docs/api-reference/maia-api-overview):
**Shared Pipeline Artifacts**
| Method | Endpoint | Description |
| --------- | ---------------------------------------- | --------------------------------------------------------- |
| **GET** | `/v1/shared-pipelines` | List all published shared pipelines across the account. |
| **GET** | `/v1/shared-pipelines/artifacts` | List published artifact versions with optional filters. |
| **GET** | `/v1/shared-pipelines/artifacts/details` | Get artifact metadata and the pipelines inside a version. |
| **POST** | `/v1/shared-pipelines/artifacts` | Publish a new artifact version. |
| **PATCH** | `/v1/shared-pipelines/artifacts` | Enable or disable a specific artifact version. |
## Pipelines - Improvements
### Orchestration
The [Python Script](/docs/components/python-script) component has been updated to use Python 3.12, ahead of the planned Python 3.10 end-of-life in October 2026. For further details, read this [Tech note](/docs/tech-notes/tech-note-matillion-agent-python-update).
**Minimum {m_runner} version:** 11.281.0
## API - New endpoints
The following endpoints have been added to the [{maia} REST API](/docs/api-reference/maia-api-overview):
**Agents**
| Method | Endpoint | Description |
| ------- | --------------------- | ------------------------ |
| **GET** | `/v1/agents/releases` | Get agent track history. |
**Minimum {m_runner} version:** All supported versions.
## API - Updated endpoints
The following endpoints have been updated:
* The `gke` deployment type is now available for both Maia and Streaming agent types for **POST** `/v1/agents` (previously only available for streaming agents).
* A new `FORBIDDEN` status value is now available for the `status` filter parameter of **GET** `/v1/pipeline-executions`.
**Minimum {m_runner} version:** All supported versions.
## Runners - New features
A public Terraform provider is now available for {s_runner}s and Streaming pipelines, enabling you to create and configure {s_runner}s and pipelines programmatically at scale using Terraform scripts. For more information, read [Deploy with Terraform](/docs/streaming/streaming-pipelines/#deploy-with-terraform).
## Admin - New features
### Billing
A **Premium** category has been introduced for selected connectors. These connectors require an additional subscription regardless of your current plan. For more information, contact your Matillion account representative.
**Minimum {m_runner} version:** All supported versions.
## Pipelines - New features
### Transformation
The [Data Cleanse](/docs/components/data-cleanse) component is now available for transformation pipelines in Amazon Redshift projects. This component provides data profiling and cleansing capabilities to help you identify and resolve data quality issues in your datasets.
**Minimum {m_runner} version:** 11.147.0
## Pipelines - Improvements
### Shared pipelines
Updated the [shared pipeline](/docs/guides/shared-pipelines#sharing-a-pipeline) experience. You can now configure scalar and grid variables in the **Share pipeline** dialog, including setting display names, marking variables as required, and reordering them. Variable descriptions are now visible when configuring the pipeline, and the [Run Shared Pipeline](/docs/guides/shared-pipelines#using-a-shared-pipeline) component now provides clearer guidance on required and optional inputs.
**Minimum {m_runner} version:** All supported versions.
## Pipelines - New features
### Observability
Added **Webhook** as a delivery channel for platform notifications, allowing users to route pipeline failure alerts to services such as Microsoft Teams, ServiceNow, or any internal tooling that accepts a webhook. Notifications can be configured at the project or environment level, removing the need to build notification logic into individual pipelines or rely solely on email alerts. For more information, read [Pipeline notifications](/docs/guides/pipeline-notifications).
**Minimum {m_runner} version:** All supported versions.
## Pipelines - New features
### Observability
You can now re-run pipelines directly from the [Pipeline run history](/docs/guides/pipeline-run-history#re-run-pipelines) without opening {designer}. You can re-run an entire pipeline or re-run execution at the step level using dedicated icons to **Run from step** (re-runs that step and all downstream steps) or **Run only step** (re-runs that step in isolation), with each action creating a new pipeline execution.
**Minimum {m_runner} version:** All supported versions.
## API - Updated endpoints
The `Git-Provider-Access-Token` header parameter is now optional for **POST** `/v1/projects/{projectId}/repositories`.
## Admin - New features
### Security
The [IP allow list](/docs/administration/ip-allow-list) enables account access to be restricted to specific IP addresses or IP ranges.
**Minimum {m_runner} version:** All supported versions.
## Connectors - Improvements
Added a new **Column mapping** parameter to Flex connectors and custom connectors. This parameter allows you to select fields from an unstructured dataset to load as individual columns. As a result, you don't need to use an **Extract Nested Data** component to unpack data into columns.
**Minimum {m_runner} version:** 11.234.0
## API - New endpoints
The following endpoints have been added to the [{maia} REST API](/docs/api-reference/maia-api-overview):
**Pipeline Quality Review**
| Method | Endpoint | Description |
| -------- | ---------------------------------- | --------------------------------------- |
| **POST** | `/v1/projects/{projectId}/reviews` | Review pipelines against quality rules. |
**Project Variables**
| Method | Endpoint | Description |
| ---------- | --------------------------------------------------- | --------------------------- |
| **GET** | `/v1/projects/{projectId}/variables` | Retrieve project variables. |
| **POST** | `/v1/projects/{projectId}/variables` | Create a project variable. |
| **DELETE** | `/v1/projects/{projectId}/variables/{variableName}` | Delete a project variable. |
| **PATCH** | `/v1/projects/{projectId}/variables/{variableName}` | Update a project variable. |
**Project Variable Environment Overrides**
| Method | Endpoint | Description |
| ---------- | ------------------------------------------------------------------------------------------- | ------------------------------------------------------------ |
| **GET** | `/v1/projects/{projectId}/variables/{variableName}/environment-overrides` | Retrieve environment-level overrides for a project variable. |
| **POST** | `/v1/projects/{projectId}/variables/{variableName}/environment-overrides` | Create an environment-level override for a project variable. |
| **DELETE** | `/v1/projects/{projectId}/variables/{variableName}/environment-overrides/{environmentName}` | Delete an environment-level override for a project variable. |
**Minimum {m_runner} version:** All supported versions.
## API - Updated endpoints
A new **Full SaaS Agent on Snowflake** example has been added to the request body for **POST** `/v1/projects`.
**Minimum {m_runner} version:** All supported versions.
## Projects - New features
### Environments
Added a new **Allow inherit project defaults** toggle to control whether an environment inherits project-level variable values by default or requires manual overrides. For more information on managing these configurations, read [Environments](/docs/guides/environments).
**Minimum {m_runner} version:** All supported versions.
## Connectors - New features
Added a new [Oracle Load](/docs/components/oracle-load) connector, which offers Full and Incremental load options.
**Minimum {m_runner} version:** 11.134.0
## Admin - Improvements
### Roles and permissions
Users with the **Contributor** role on a project can now create and modify project variables, but can't delete them. For more information, read [Project roles](/docs/administration/project-roles#role-permissions).
**Minimum {m_runner} version:** All supported versions.
## Pipelines - Improvements
### Observability
Added a new table to the **Audit event types** section of the **Audit Service** page to list the [{maia_agents}](/docs/guides/audit-service#maia-runner) event types supported by the Audit API. Also updated the [Account](/docs/guides/audit-service#account) table to include additional event types.
**Minimum {m_runner} version:** All supported versions.
## Designer - New features
New system variables for environment defaults have been added, allowing you to reference default connection values in your pipelines. These variables provide access to the default database, schema, warehouse, role, catalog, compute resource, S3 bucket, and application configured for the environment. For more information, read [System variables](/docs/guides/system-variables#list-of-system-variables).
**Minimum {m_runner} version:** 11.217.1
## Pipelines - New features
### Orchestration
The following Load components are now available for Databricks:
* [AVRO Load](/docs/components/avro-load)
* [CSV Load](/docs/components/csv-load)
* [JSON Load](/docs/components/json-load)
* [ORC Load](/docs/components/orc-load)
* [Parquet Load](/docs/components/parquet-load)
**Minimum {m_runner} version:** 11.154.0
## Pipelines - Improvements
### Observability
Added a new table to the **Audit event types** section of the **Audit Service** page that lists the [Project](/docs/guides/audit-service#project) event type supported by the Audit API.
**Minimum {m_runner} version:** All supported versions.
## Projects - Improvements
### Artifacts
Updated the information shown when [viewing artifacts](/docs/guides/artifacts#view-artifacts). You can now see the Git hash of the commit that created the artifact, and the name of the user who created the artifact.
**Minimum {m_runner} version:** All supported versions.
## Runners - Improvements
* Updated the [{s_runner} version](/docs/streaming/agent-versions) from `2.115.1` to `2.116.1`.
* Library updates and security fixes.
* Added a new advanced property called `matillion.snowflake.maintain-primary-keys` (default: `false`). When enabled, the {s_runner} will mirror primary key constraints from your source database onto the destination Snowflake tables. This includes table creation and schema changes. If changes to the primary key of a table are required, writes to the source table must be stopped and outstanding changes allowed to flow through to Snowflake before stopping the pipeline. Once the primary key update is complete, the pipeline can be restarted. To enable this feature, add the following to your Streaming pipeline's advanced properties:
* `matillion.snowflake.maintain-primary-keys` = `true`
This is a Snowflake-specific property and has no effect on other target types. Snowflake primary keys are unenforced (informational only), but they can be useful for query optimization hints and downstream tooling that reads table metadata.
## Pipelines - Improvements
### Observability
The **Your activity** page that displays the pipeline run history has been updated to improve visibility and make troubleshooting faster. The update simplifies the run history view and adds new information to help you quickly understand pipeline runs and identify issues during investigations. For more information, read [Pipeline run history](/docs/guides/pipeline-run-history#pipeline-run-history).
The [Audit service](/docs/guides/audit-service) now records audit events for [project connections](/docs/guides/audit-service#project) and [environment connections](/docs/guides/audit-service#environments). Project connection events are produced for SSH tunnels, OAuths, and secret definitions. Environment connection events are produced for OAuths and secret definitions.
**Minimum {m_runner} version:** All supported versions.
## Projects - New features
### Secret definitions, OAuths
Added support for environment-level secrets and OAuth credentials, allowing you to manage secrets and OAuth credentials at the environment level for more granular control and simplified cross-environment credential management. For more information, read [Configure values and overrides](/docs/guides/secrets-and-secret-definitions#configure-values-and-overrides).
**Minimum {m_runner} version:** All supported versions.
## Projects - Improvements
### Secret definitions, OAuths
* You can now edit existing secrets and OAuths, making it easier to rotate credentials.
**Minimum {m_runner} version:** All supported versions.
## API - New endpoints
Credential management has been simplified into a single API endpoint for connections.
The following endpoints have been added to the [{maia} REST API](/docs/api-reference/maia-api-overview):
**Connections**
| Method | Endpoint | Description |
| ---------- | ------------------------------------------------------------ | ------------------------------------------------------------------------------ |
| **DELETE** | `/v1/projects/{projectId}/connections` | Delete a project connection by name. |
| **GET** | `/v1/projects/{projectId}/connections` | Retrieve a page of project connections, optionally filtered by parameters. |
| **POST** | `/v1/projects/{projectId}/connections` | Create a new project connection. |
| **PUT** | `/v1/projects/{projectId}/connections` | Update an existing project connection by name. |
| **DELETE** | `/v1/projects/{projectId}/connections/environment-overrides` | Delete environment connection overrides. |
| **GET** | `/v1/projects/{projectId}/connections/environment-overrides` | Retrieve a page of connection overrides which apply in a specific environment. |
| **POST** | `/v1/projects/{projectId}/connections/environment-overrides` | Create or update an environment connection override. |
**Minimum {m_runner} version:** All supported versions.
## API - Updated endpoints
A new enum value of `SKIPPED` has been added to the `status` query parameter for **GET** `/v1/pipeline-executions`.
**Minimum {m_runner} version:** All supported versions.
## Connectors - New features
[Microsoft Exchange](/docs/components/data-transfer#__tabbed_1_10) has been added as a new source to the Data Transfer component. This enables users to transfer attachments from Exchange email messages and calendar events using OAuth-based authentication.
**Minimum {m_runner} version:** 11.183.0
## API - New endpoints
The following endpoint has been added to the [{maia} REST API](/docs/api-reference/maia-api-overview):
**Secrets**
| Method | Endpoint | Description |
| -------- | ------------------------------------ | ------------------------------------ |
| **POST** | `/v1/secrets/{agentId}/secret-value` | Set a secret value for a {m_runner}. |
## Connectors - New features
Added a new [Sybase ASE Load](/docs/components/sybase-ase-load) connector, which offers Full and Incremental Load options. We recommend using this connector instead of using the Database Query component to connect to Sybase ASE.
**Minimum {m_runner} version:** 11.191.2
## Runners - New features
{m_runner}s can now be deployed using Kubernetes on Amazon EKS or Azure AKS. This provides horizontal pod autoscaling, standardized Helm chart deployment, and native cloud authentication. A public [deployment library](https://github.com/matillion-public/deployment-library) is available with Terraform modules, Helm charts, and pre-deployment validation scripts. For an overview, read the [Kubernetes deployment guide](/docs/guides/kubernetes-deployment-guide).
**Minimum {m_runner} version:** All supported versions.
## Maia AI Agents - New features
{maia_agents} can now notify you when they finish processing or require your input. This means you can prompt {maia_agents}, then switch to another task, and return as soon as {maia_agents} have finished working on your prompt. For more information, read [Background notifications](/docs/guides/maia-sessions-and-tools#background-notifications).
**Minimum {m_runner} version:** All supported versions.
## Maia AI Agents - Improvements
In the **Fix with Maia** dialog, {maia_agents} use contextual awareness of the executed pipeline's codebase and history to suggest the most relevant branch based on commit and publish history, and automatically select the environment where the failure occurred. For more information, read [{maia_agents} assisted - intelligent pipeline recovery](/docs/guides/maia-troubleshooting#pipeline-recovery-with-maia).
**Minimum {m_runner} version:** All supported versions.
## Projects - Improvements
### Git providers
The [Matillion App for GitHub](/docs/guides/installing-matillion-app-github-marketplace#required-github-app-permissions) now requests **read** and **write** access to GitHub Actions workflow files. This enables {maia} to update and manage workflow files (`.github/workflows/`) in connected repositories as part of GitHub pipeline management.
**Minimum {m_runner} version:** All supported versions.
## Pipelines - New features
### Transformation
A new **Warehouse data** tab is available when you click the **Add component** icon on the canvas, allowing you to directly add data from your warehouse as a [Table Input](/docs/components/table-input) component in transformation pipelines.
**Minimum {m_runner} version:** All supported versions.
## Pipelines - New features
### Observability
* The [Pipeline run history](/docs/guides/pipeline-run-history#pipeline-run-history) page now displays **Skipped** pipeline runs when a schedule has **Allow concurrent schedule runs** disabled and a new execution is triggered while a previous run for the same schedule is still in progress. Skipped runs are recorded with an amber status indicator.
* You can now search pipeline runs by **Execution ID**, in addition to pipeline name.
**Minimum {m_runner} version:** All supported versions.
## Maia AI Agents - Improvements
{maia_agents} now support [workload conversion](/docs/guides/convert-workloads-from-other-platforms) from Apache NiFi, IBM DataStage, WhereScape, and Oracle ODI, enabling you to convert existing ETL/ELT workloads into production-ready {maia} pipelines.
**Minimum {m_runner} version:** All supported versions.
## Maia AI Agents - New features
* {maia_agents} can now ask questions in your conversations to better understand your prompts.
* When approving [tool permissions](/docs/guides/maia-sessions-and-tools#tool-permissions) in your chats with {maia_agents}, you can now provide a reason when declining permissions. This can help {maia_agents} to better understand your prompt or to adjust their approach.
* {maia_agents} now dynamically update your conversation titles to accurately reflect the topic of discussion, making it easier to understand your [conversation history](/docs/guides/maia-sessions-and-tools#conversation-history) and return to previous conversations.
**Minimum {m_runner} version:** All supported versions.
## Pipelines - New features
### Transformation
A [Data Cleanse](/docs/components/data-cleanse) component is now available for transformation pipelines in Snowflake projects. This component provides data profiling and cleansing capabilities to help you identify and resolve data quality issues in your datasets.
**Minimum {m_runner} version:** 11.147.0
## Runners - Improvements
* Updated the [{s_runner} version](/docs/streaming/agent-versions) from `2.113.75` to `2.115.1`.
* Db2 for IBM i: Added Heartbeat functionality.
* Library updates and security fixes.
## Maia AI Agents - New features
You can now create [skills](/docs/guides/maia-skills) to help {maia_agents} respond to your prompts more effectively. Skills are sets of reusable instructions that are particularly useful if you often give {maia_agents} repetitive prompts. They provide guidelines and standards for {maia_agents} to follow, which means you don't need to include this information in every prompt.
**Minimum {m_runner} version:** All supported versions.
## Pipelines - New features
### Observability
* You can now automatically fix selected pipeline quality rule violations using [Fix with Maia](/docs/guides/pipeline-quality#fixing-rule-violations-with-maia) from the **Review results** panel, helping resolve common issues such as component naming, missing start or end components, and unused components.
* Added a new [Review all](/docs/guides/pipeline-quality#reviewing-all-pipelines-in-a-branch) pipelines option to the {designer} canvas. This lets you review all orchestration and transformation pipelines in a branch in a single action, making it easier to identify pipeline quality issues across your branch.
* You can now choose to run pipeline quality checks from the **Commit changes** dialog by enabling the [review option before committing](/docs/guides/pipeline-quality#reviewing-pipelines-before-committing). When enabled, commits are blocked if rule violations with an **error** enforcement level are detected, while commits with warnings or no issues proceed successfully.
**Minimum {m_runner} version:** All supported versions.
## API - Updated endpoints
The following endpoints have been updated:
* Added a `[FULL_SAAS]` enum value to property `agentDeploymentType` for **POST** `/v1/projects`.
* Added a 403 response to **DELETE** `/v1/projects/{projectId}/secret-references/{secretReferenceName}`.
## Pipelines - New features
### Observability
A new [Additional Users](/docs/administration/usage-dashboard#additional-users) tab is now available in the **Consumption Dashboard**, providing visibility into individual user credit charges incurred during the current billing period.
**Minimum {m_runner} version:** All supported versions.
## Connectors - Improvements
A new **OAuth 2.0 Client Credentials** option is now available in the Authentication type drop-down menu for the [Salesforce Load](/docs/components/salesforce-load) connector.
**Minimum {m_runner} version:** 11.67.0
## Runners - New features
Hybrid SaaS {m_runner}s can now export {m_runner} logs to Matillion's systems for support purposes. This facilitates faster response if support is needed. This feature can be enabled or disabled in the {m_runner} settings. For more information, read [Export logs](/docs/guides/export-logs-hybrid-saas).
**Minimum {m_runner} version:** 11.46.1
## Connectors - Improvements
Added a new property to the [Workday Load](/docs/components/workday-load) component. The new **Advanced Mode Type** property lets you choose whether to use an SQL query or SOAP operation to fetch your data.
**Minimum {m_runner} version:** 11.120.0
## Maia AI Agents - New features
[Convert Workloads](/docs/guides/convert-workloads-from-other-platforms) has been added to {maia_agents}. This new feature allows users to leverage {maia_agents}' intelligence to seamlessly convert existing ETL/ELT workloads from other platforms into production-ready {maia} pipelines.
**Minimum {m_runner} version:** All supported versions.
## API - Updated endpoints
The `pipelineName` field has been added to the `dimensions` object of the response for `GET /v1/consumption`. Applies to orchestration and transformation pipeline credit usage types.
## Pipelines - New features
### Observability
Added **Slack** as a delivery channel for platform notifications, allowing users to receive pipeline failure alerts directly in Slack. Notifications can be configured at the project or environment level, removing the need to build notification logic into individual pipelines or rely solely on email alerts. For more information, read [Pipeline notifications](/docs/guides/pipeline-notifications).
**Minimum {m_runner} version:** All supported versions.
## Connectors - New features
Added an [Amazon Redshift Load](/docs/components/amazon-redshift-load) connector for Snowflake projects. This connector offers Incremental Load and we recommend using it instead of the Database Query component to connect to Amazon Redshift.
**Minimum {m_runner} version:** 11.94.0
## Admin - Improvements
### Roles and permissions
The **Runner** role is now renamed to **Operator**. This role can now modify environment-level overrides for project-level variables, secret definitions, and OAuths. For more information, read [Environment roles](/docs/administration/environment-roles).
**Minimum {m_runner} version:** All supported versions.
## Maia AI Agents - New features
[Fix with Maia](/docs/guides/maia-troubleshooting#pipeline-recovery-with-maia) has been added to Root Cause Analysis. This new feature allows users to leverage Maia's capabilities to automatically identify and fix issues in their scheduled or API-executed pipeline, reducing downtime and improving efficiency.
**Minimum {m_runner} version:** All supported versions.
## Connectors - New features
Added a new [HubSpot Load](/docs/components/hubspot-load) connector, which uses the HubSpot API to retrieve your data. This connector offers Full and Incremental Load options.
**Minimum {m_runner} version:** 11.72.0