Hunting a slow page
Maya Okafor, backend lead at Acme Ltd, asked which queries took the most time on production yesterday. Memtro read Insights for the branch and listed the top patterns with their tables, count and p99 latency.
PlanetScale + Memtro
Which schema changes are waiting to deploy, and which queries are slowest?
PlanetScale hosts your databases, and Memtro reads them so you can ask about schema changes and performance. It reads databases with kind, region and plan, branches, and deploy requests with their state, approval and reviews. It also reads backups, the heaviest or slowest query patterns on a branch, and invoices. When asked and granted, it can create or delete a branch, and open, review, deploy, apply or revert a deploy request. It can also take or delete a backup. PlanetScale can tell Memtro when a deploy request errors or a backup fails, and a job can start on it.
In Memtro
What you can do with PlanetScale in Memtro
Ask in plain English from chat, your editor, a scheduled job or your own agent. The model picks the right PlanetScale tool, reads what it needs, and answers with the source cited.
What it does
- 01Reads databases with kind, state, plan, region and whether deploys need approval, and branches with their state and size.
- 02Reads deploy requests with state, approval, deployment progress and reviews.
- 03Reads backups with size and cost, the query patterns taking most time on a branch with p50 and p99 latency, and invoices.
- 04When asked and granted, creates or deletes a branch, or opens, comments on or approves a deploy request. A new request is visible to everyone on the database.
- 05When asked and granted, deploys, cancels, closes, applies or reverts a deploy request, or skips its revert period. These change the target branch, often production.
- 06With events on, a deploy request waiting to cut over, an errored deploy request or a failed backup can start a job.
Acts when asked
When you ask it to
These tools change something in PlanetScale. The model is told to use them only when you ask for exactly that, and an administrator grants each one by name to the teams that may use it; a personal connection acts only after its owner allows it.
- 01Create a branch
- 02Delete a branch
- 03Open a deploy request from one branch into another
- 04Comment on or approve a deploy request
- 05Deploy a deploy request (queue its schema change)
- 06Cancel a queued deploy request
- 07Close a deploy request without deploying it
- 08Apply a gated deploy request (cut over)
- 09Revert a deployed schema change
- 10Skip the revert period of a deployed schema change
- 11Take a backup of a branch
- 12Delete a backup
Events
When it happens
Turn events on in the connection's Events tab on the Connections page, then add the address it shows as a webhook in PlanetScale, choosing the same kinds; the tab shows PlanetScale's own steps. Each event can start a job, with the event as its input. PlanetScale signs every delivery, and one whose signature does not check out is refused before it is read.
Events and webhooks- 01A deploy request was opened
- 02A deploy request was queued
- 03A deploy request started running
- 04A deploy request is waiting to cut over
- 05A deploy request finished applying its schema
- 06A deploy request stopped with an error
- 07A deploy request was reverted
- 08A deploy request was closed
- 09A branch is ready to connect
- 10A branch has a new insights anomaly
- 11A schema recommendation was created
- 12A branch went to sleep
- 13A production branch is about to start maintenance
- 14A Postgres branch ran out of memory
- 15A new primary was promoted
- 16A backup completed
- 17A backup failed
- 18A keyspace reached a storage threshold
- 19A Postgres database reached a storage threshold
- 20PlanetScale staff asked for access
- 21Maintenance was scheduled
- 22Maintenance started
- 23Maintenance finished
- 24Maintenance failed and will be retried
- 25A test event
In practice
Ask it like you would a colleague
Once PlanetScale is connected, these are the kinds of things people ask. Plain questions, no syntax. If the answer lives partly in another system you have connected, Memtro looks there too.
PlanetScale + Memtro
Which deploy requests are open on the main database?
What are the slowest queries on production over the last day?
When did the last backup of production run?
Is our latest invoice paid?
Create a development branch called add-orders-index.
Cutover needs a person
When PlanetScale told Memtro a deploy request at Northwind was waiting to cut over, a job started. It read the request's notes and reviews and posted them to Tom Reilly, who asked Memtro to apply it.
Teams
Where it fits
The teams that lean on PlanetScale most, a question each of them asks, and what they connect beside it.
- Data and analyticsQuestions answered from the warehouse, with the SQL shownAnalyst, Data engineer, Anybody with a question, Head of data“Revenue by segment for Q3 from the warehouse, with the SQL you used”Used alongside Postgres, Stripe, Slack, Google Drive, Snowflake and 3 more.
- EngineeringKeep decisions from evaporatingEngineer, Tech lead, On call, New joiner“Who owns the billing migration, and what did we decide about retries?”Used alongside GitHub, Linear, Jira, Confluence, Slack and 5 more.
- FinanceAsk the ledger questions in plain EnglishFinance lead, Credit control, Management accountant, Founder or CFO“Which invoices over £5k are more than 30 days overdue?”Used alongside Xero, Stripe, Postgres, HubSpot, QuickBooks Online and 10 more.
Better together
Works with
A record in PlanetScale is rarely the whole story. Connect the systems that hold the rest and Memtro checks them together, then remembers what it learned so the next question starts further along.
- AnalyticsPlanetScale with Google Analytics 4Ask your traffic questions in plain English and get the numbers back.
- AnalyticsPlanetScale with Plausible AnalyticsAsk about your website traffic in plain English.
- AnalyticsPlanetScale with UmamiAsk about your website traffic in plain English, without sending anybody to a dashboard.
- Finance and commercePlanetScale with XeroInvoices, bills, contacts and reports, answered in plain English with figures cited to Xero.
- Finance and commercePlanetScale with StripeCustomers, charges, invoices and subscriptions through a restricted key.
- Finance and commercePlanetScale with FellohBookings, payments, refunds, chargebacks, supplier payments and payouts from the travel payments platform.
- Work managementPlanetScale with JiraJQL search, issue details and projects from Jira Cloud.
- Work managementPlanetScale with Clio ManageAsk the matter: its client, its fee earner, the time on it and what is still unbilled.
- Work managementPlanetScale with LinearIssues, projects and teams, synced hourly into memory.
Also in Databases and warehouses
Using something else for the same job, or both? Each connector has its own tools; connect as many as you like.
- Postgres
- MySQL and MariaDB
- SQL Server
- BigQuery
- Snowflake
- AWS Athena
- MongoDB
- ClickHouse
- Amazon Redshift
- Databricks
- Elasticsearch
- dbt Cloud
- Fivetran
- Neon
- Appwrite
Grist- Hasura
- Meilisearch
NocoDB- Qdrant
Typesense
Weaviate
Tinybird
Hightouch
Census- Airbyte
- Budibase
- CockroachDB Cloud
- Cube Cloud
- DataHub
kintone
Monte Carlo
OpenMetadata- Pinecone
- Quickbase
Secoda
Softr
Teable- Turso
- Upstash
Vectara- Xata
- Zilliz Cloud
Setup
How to connect
Connect withService token, plus token ID and organization where needed
- 1
In PlanetScale, open the organization Settings, then Service tokens, and create a token. Add read permissions, plus branch, deploy request and backup permissions only for the actions you allow.
- 2
In Memtro, open Connections → PlanetScale and paste the Service token. Add the Service token ID for an older token, and the Organization if the token sees more than one.
- 3
An administrator grants the branch, deploy and backup actions by name. Events are turned on from the connection's Events tab, then added in each database's Webhooks settings.
- 4
Ask which deploy requests are open.
Also works from
Chat in the dashboard, the OpenAI-compatible API, scheduled jobs, and any MCP client: Claude Code, OpenCode, Cursor, Claude Desktop. The PlanetScale tools are the same everywhere.
FAQ
Questions
Can Memtro change anything in PlanetScale?+
Only when you ask, and only what it has been granted: create a branch; delete a branch; open a deploy request from one branch into another; comment on or approve a deploy request; deploy a deploy request (queue its schema change); cancel a queued deploy request; close a deploy request without deploying it; apply a gated deploy request (cut over); revert a deployed schema change; skip the revert period of a deployed schema change; take a backup of a branch; delete a backup. Every other tool reads, searches or lists. The Connections page labels the tools that act, an administrator grants each one by name to the teams that may use it, and a personal connection acts only after its owner allows it.
Who can use a PlanetScale connection?+
A personal connection serves only your own requests. A shared connection serves the organisation: everyone, or only the teams an admin chooses. Admins can always add shared connections; other members need the permission.
Does Memtro store my PlanetScale data?+
Credentials are encrypted at rest. Tool results are cached briefly so agent loops do not repeat identical calls, and the facts the model extracts are saved to your organisation's memory with provenance (system, record and link) so they can be checked and superseded.
How do I connect PlanetScale?+
Service token, plus token ID and organization where needed. In PlanetScale, open the organization Settings, then Service tokens, and create a token. Add read permissions, plus branch, deploy request and backup permissions only for the actions you allow.
Can PlanetScale start a job?+
Yes. Turn events on in the connection's Events tab on the Connections page, then add the address it shows as a webhook in PlanetScale, choosing the same kinds; the tab shows PlanetScale's own steps. A job set to start "When an event arrives" then runs once for each matching event, with the event as its input. The run treats the event as outside data, never as instructions, and anything the run remembers is kept as unverified.
Connect PlanetScale in a few minutes
Seven-day free trial, then one monthly price per organisation. Connectors, jobs and memory are never metered.