Integration · Automate and collaborate

Prisma Postgres and Meteor AI agents

Connect Prisma Postgres to an internal Met: check databases, backups and schemas, plus your Prisma Compute deployments and logs.

Book a demo See every integration

How does Prisma Postgres connect to Meteor?

With the Prisma Postgres ready connector, your Met uses Prisma's official MCP server on the workspace you choose when connecting, and it gets 31 tools. It lists your Prisma Postgres databases, reads their schema, checks the automated backups and your Prisma Compute apps, with their builds, deployments and logs. Running SQL, changing the schema, creating databases or buckets, restoring a backup, starting, stopping, promoting or rolling back deployments, setting environment variables and everything that deletes start turned off until the admin turns them on. Creating connection strings and bucket keys comes turned on. Prisma warns that changing or deleting resources can cause data loss or interrupt live traffic.

What it automates

What a Meteor Met does with Prisma Postgres

Find out why the last deployment failed

When someone on the team reports on WhatsApp that the app is not responding, the Met lists the app's builds and deployments in Prisma Compute, reads the latest one's logs and tells you which error shows up.

Confirm there is a backup before a change

Before your team touches the database, the Met lists its automated backups and tells you which is the latest. Restoring it creates a new database; it does not overwrite the current one.

Check a database's structure

The Met reads the database schema and explains its tables and columns, without running SQL.

Go back to the previous version

If the last deployment broke something and the admin turned on the rollback action, which starts turned off, the Met makes an earlier deployment of the app live and confirms which one is running.

31 tools

Everything your Met can ask Prisma Postgres for

Each one is a real action in the connector. You can switch off the ones you would rather it never used.

Tool	What it does
`fetch_workspace_details`	Returns the details of the connected Prisma Postgres workspace.
`list_prisma_postgres_databases`	Lists the workspace's Prisma Postgres databases.
`create_prisma_postgres_database`	Creates a managed Prisma Postgres database and a new project. Starts turned off.
`delete_prisma_postgres_database`	Permanently deletes a database by ID; it cannot delete a project's default database. Starts turned off.
`list_prisma_postgres_connection_strings`	Lists a database's available connection strings.
`create_prisma_postgres_connection_string`	Creates a new connection string for a database.
`delete_prisma_postgres_connection_string`	Deletes a connection string by ID. Starts turned off.
`list_prisma_postgres_backups`	Lists a database's available automated backups.
`create_prisma_postgres_recovery`	Restores a database into a new database from a backup. Starts turned off.
`introspect_database_schema`	Reads a database's schema to show its structure.
`execute_sql_query`	Runs SQL to read, modify or delete data in a database, without changing its schema. Starts turned off.
`execute_prisma_postgres_schema_update`	Runs a schema update on a database. Starts turned off.
`list_object_store_buckets`	Lists the workspace's Object Storage buckets, optionally filtered by project.
`create_object_store_bucket`	Creates an Object Storage bucket in a project. Starts turned off.
`delete_object_store_bucket`	Permanently deletes a bucket with all its objects and access keys. Starts turned off.
`create_object_store_bucket_key`	Creates an S3-compatible access key for a bucket.
`delete_object_store_bucket_key`	Deletes a bucket access key, which stops working immediately. Starts turned off.
`search_prisma_documentation`	Answers a question about Prisma with the official documentation, with links to the sources.
`list_prisma_compute_apps`	Lists the workspace's Prisma Compute apps, optionally filtered by project.
`list_prisma_compute_builds`	Lists builds, newest first, with filters by project, branch and state.
`list_prisma_compute_deployments`	Lists a Prisma Compute app's deployments, newest first.
`get_prisma_compute_deployment_logs`	Reads a deployment's runtime logs, one page at a time.
`promote_prisma_compute_deployment`	Makes a running deployment the live version of an app. Starts turned off.
`rollback_prisma_compute_app`	Makes an existing deployment live, starting it first if it is stopped. Starts turned off.
`start_prisma_compute_deployment`	Starts an existing deployment from an already uploaded build; it does not build or upload the app. Starts turned off.
`stop_prisma_compute_deployment`	Stops a deployment and disconnects it from the public URL if it is the live version. Starts turned off.
`delete_prisma_compute_deployment`	Permanently deletes a deployment, stopping it first if needed. Starts turned off.
`delete_prisma_compute_app`	Permanently deletes an app; active deployments block it unless you ask to delete them too. Starts turned off.
`set_prisma_compute_env_var`	Sets a production or preview environment variable, or overrides it for one branch. Values are encrypted and never returned. Starts turned off.
`delete_prisma_compute_env_var`	Permanently deletes an environment variable by ID. Starts turned off.
`get_prisma_composer_topology`	Shows the saved connections between a project's Composer services and resources, with console links; it may be out of date.

The limits

What the Prisma Postgres integration does NOT do

We would rather say it here than have you find out during rollout.

It does not ask you for confirmation in the chat before writing. Deleting, running SQL, changing the schema, creating databases or buckets, restoring backups, moving deployments and setting variables start turned off; creating connection strings and bucket keys stays on unless the admin turns it off in Connections.
It does not upload or build your code. The Compute tools handle apps and deployments that already exist; to deploy a new app you follow the Prisma Compute guide.
It does not return environment variable values. Prisma stores them encrypted and never returns them.
It does not see other workspaces. It works on the workspace you choose when connecting.
It does not pay your Prisma plan. The databases and buckets it creates, if the admin turns those actions on, stay in your account; Meteor does not resell the service.

Before you start

What you need to connect Prisma Postgres

1

A Prisma account with access to a workspace

If you do not have one, sign up in Prisma Console. The workspace groups your projects and is the one the Met will see.
2

Connect it in Met

In Met, open Connections → Prisma Postgres → Connect. A Prisma window opens so you sign in, choose the workspace and approve access. Then you choose which actions stay on.

Frequently asked questions

Questions about Prisma Postgres and Meteor

No. According to Prisma, the restore creates a new database from the backup. The current database stays as it is.

It does not upload or build code. It can start, promote, stop or roll back deployments that already exist if the admin turns those actions on in Connections, since they start turned off.

Environment variables are never returned. Connection strings and bucket keys can be listed or created, so it is best to assign the connector to a Met that only serves the team.

You do, with your own Prisma plan. Meteor does not resell Prisma: the connector uses your company's account and what it creates stays there.

Step by step